Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A study of the developmental activities that describe highly elite men and women soccer players and the… Hendry, David T. 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2018_may_hendry_david_thomas.pdf [ 1.84MB ]
Metadata
JSON: 24-1.0365946.json
JSON-LD: 24-1.0365946-ld.json
RDF/XML (Pretty): 24-1.0365946-rdf.xml
RDF/JSON: 24-1.0365946-rdf.json
Turtle: 24-1.0365946-turtle.txt
N-Triples: 24-1.0365946-rdf-ntriples.txt
Original Record: 24-1.0365946-source.json
Full Text
24-1.0365946-fulltext.txt
Citation
24-1.0365946.ris

Full Text

 A study of the developmental soccer activities that describe highly elite men and women soccer players and the relation of these activities to indices of motivation, soccer-related skills and progressions  by David T Hendry BSc (Hons) Edinburgh Napier University, 2003 MSc University of British Columbia, 2012 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Kinesiology)  THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)   © David T. Hendry, 2018ii   Abstract A multivariable measurement approach was used to determine relationships between talent development pathways in soccer and outcomes related to future elite success, soccer skill ratings and motivation. In Study 1, elite “Academy” youth soccer players in the UK were followed over 5-years to ascertain the developmental activities that distinguished players which progressed to youth and adult professional levels. Professional players followed an early majority engagement pathway characterized by predominant involvement in high volumes of soccer practice and play (i.e., self-led activities) from an early age. They participated in other sports but the majority time was in soccer. Adult professional players accumulated more hours in soccer play compared to youth professionals, but not practice. In Study 2, coach ratings of soccer skills were related to attainment of youth and adult professional status. Coach ratings of technical, tactical and creative skill were higher in players that achieved professional youth status, versus de-selected academy players. Only tactical skill, and somewhat, technical skill, differentiated adult from youth professionals. Technical and tactical skill ratings were primarily related to hours in soccer practice, but there were no relations to hours in play. In Study 3, practice amounts were related to markers of self-determined motivation (SDM) but soccer play hours were not related to SDM. Through comparisons with recreational, age-matched soccer players, SDM was shown to become less self-determined from 15 -17 yr, but only among elite players. In Study 4, the developmental pathways engaged in by elite (National-team) and sub-elite (Varsity) adult women soccer players were assessed. National players followed an early majority engagement pathway, engaging in more soccer play than sub-elites, which was also rated as more challenging. In summary, success in elite soccer is characterized by an early, iii  majority engagement pathway consisting of early childhood involvement in soccer, relatively high volumes of practice and play, and majority time in soccer, in comparison to other sports.   iv  Lay summary I investigated outcomes associated with different developmental pathways associated with success in soccer. These pathways differed with respect to the importance placed upon soccer play (e.g., street soccer), soccer practice and engagement in multisport activities during childhood. Adult professional male players participated in more hours in soccer play those players that failed to make the transition from professional youth status. Soccer practice was associated with coach ratings of technical, tactical and creative skill. Contrary to ideas forwarded by some researchers, soccer play hours were unrelated to measures of motivation. Elite (world-class) female players followed a similar path to males but tended to engage in less soccer play. Overall, elite male and female soccer players followed an early engagement pathway, in which they engaged primarily in soccer related play and practice in comparison to other sports from an early age.   v  Preface  A version of Study 1 (chapter 2) has been accepted for publication. The citation for this publication is: Hendry, D.T., & Hodges, N.J., (2018). Early majority engagement pathway best defines transitions from youth to adult elite men’s soccer in the UK: A three time point prospective and retrospective study. Psychology of Sport and Exercise. (36).DOI10.1016/j.psychsport.2018.01.009. As first author I was responsible for the overall research design, data analysis and interpretation, as well the subsequent write-up. Dr. Nicola Hodges helped significantly at each stage of the process. Before data collection, this research project (H11-01946) was approved by the UBC Behavioral Research Ethics Board on 8th September 2011. Study 2 (chapter 3) has been accepted for publication. The citation for this publication is: Hendry, D.T., Williams, A.M., & Hodges, N.J. (2018). Coach ratings of skills and their relations to practice, play and success: Transitions from youth-elite to adult-Professional status in soccer.  Journal of Sport Sciences. DOI10.1080/02640414.2018.1432236. I was involved in all aspects of the study including; design, ethical application, participant recruitment, data collection, analysis and interpretation of the data, and manuscript preparation. Dr. Nicola Hodges was involved at each stage from study design through to the reviewer responses. Dr. A. Mark Williams provided conceptual advice and contributed to manuscript preparation. Player practice history data from Study 1 was used in the present study. New data pertaining to skill ratings, collected as part of Study 1, but not included in the manuscript for Study 1 formed the basis of Study 2. As per study 1, ethical approval was granted by the UBC Behavioral Research Ethics Board (H10-01946) on 8th September, 2011.  vi    A version of Study 3 (chapter) is in preparation to be submitted. Dr. Peter Crocker, Dr. A. Mark Williams and Dr. Nicola Hodges are co-authors on this paper. As first author I was involved in the design, ethical application, participant recruitment, data collection, analysis and interpretation of data, and manuscript preparation. Dr. Nicola Hodges was involved at each stage the process. Dr. Peter Crocker and Dr. A. Mark Williams provided conceptual input and aided the preparation of the manuscript. Some data from Study 1, was included in Study 3, but this study also included comparisons with two new groups of competitive, yet recreational youth athletes in Canada. Ethical approval was granted by the University of British Columbia Behavioral Research Ethics Board (H11-01946).    Study 4 (chapter 5) is in preparation to be submitted for publication. Dr. Nicola Hodges, Dr. Paul Ford, and Dr. A. Mark Williams are co-authors on the manuscript. Some of the data from this study will also be included in a cross-country analysis of development activities engaged in by world-class female soccer players (led by Dr. Ford). Dr. Nicola Hodges was involved at each stage the process. I was involved in all aspects of this investigation including study design, ethical application, participant recruitment, data collection, analysis and interpretation of the data, and manuscript preparation. Ethical approval was granted by the University of British Columbia Behavioral Research Ethics Board (H14-02779) on 31st, October 2014.     vii  Acknowledgements  First and foremost, I would like to thank my supervisor and friend Dr. Nicola Hodges. I owe you so much, from taking a chance on me in the first place to supporting and guiding me through the struggles that I have faced over the duration of graduate school. I am eternally grateful and thank you for the lasting contribution that you have made to me personally and to my family. I simply would not have got to this point without you. Thank you.  I am also very grateful to my committee members Dr. Mark Williams and Dr. Peter Crocker who have provided insightful feedback at every stage of this PhD process. Dr. Williams, watching your interview on Sky Sports created the initial spark of intellectual curiosity that prompted me towards pursuing a career in academia. Dr. Crocker, seeing your mind work has been one of the greatest privileges of studying at UBC. Your graduate seminar was a huge inspiration, I could have listened to you talk all day.   To my fellow graduate students; Des Mulligan, Nicole Ong, Beverly Larsson, April Karlinsky, Paul Campagnaro, Tom Coppolla and Keith Lohse, thank you all for your friendship and support over the last several years. We have had some great laughs and fun adventures! I am sure there will be many more to come.  Thank you to my wonderful family, Mum, Dad, Craig, Gran, Grampa and Granny Grace. Thank you for your love and support. I hope this makes you proud.  I would also like to thank those people that have helped give me the strength of character and belief to pursue my goals; Lawrence Haggart, Mike Mosher; Glyn Roberts, Gerard McCafferty, Billy Irvine, Matthew Stewart, John McBeth, Scott Dingwall, Richard Fox, Barry Main, Dave Partridge, Scott Lescak, viii  Andy Gould, Doreen O’Carrol, Catriona Simpson, Peter Laird, Dr. Tony Westbury, Alan Boyd, Jimmy McNee, Andy Kennedy, Robin Gibson, Craig Mulholland, and Stewart Taylor.  Finally, thank you to my beautiful wife, Kimberly and son Louis. You both make me happier than I will ever be able to describe. Kim, without you, none of this would have been possible. I would never have even dreamed about this. Thank you for your unwavering love and support throughout this whole PhD process and for looking after Louis with such love when I have been busy writing.      ix    To Kim, Louis and Baby This is all for you.     x  Table of contents  Abstract ........................................................................................................................................................ ii Lay summary .............................................................................................................................................. iv Preface .......................................................................................................................................................... v Acknowledgements ................................................................................................................................... vii Table of contents ......................................................................................................................................... x List of tables............................................................................................................................................... xv List of figures ............................................................................................................................................. xvi Chapter 1: Introduction ............................................................................................................................. 1 1.1 Deliberate practice ..................................................................................................................... 2 1.2 The DMSP .................................................................................................................................. 3 1.3 Early engagement hypothesis ................................................................................................... 5 1.4 Measuring success ..................................................................................................................... 6 1.5 Discriminability of skill ............................................................................................................. 6 1.6 Developing skill .......................................................................................................................... 9 1.7 Activity relationships with motivation ................................................................................... 11 1.8 Self-Determined Motivation ................................................................................................... 12 1.9 Developing female soccer experts ........................................................................................... 14 1.10 Challenge point framework .................................................................................................... 16 1.11 Unique contributions to the sport expertise literature ......................................................... 19 1.12 Overall research objectives ..................................................................................................... 22 Chapter 2: Early majority engagement pathway best defines transitions from youth to adult elite men’s soccer in the UK: A three time point prospective and retrospective study .............................. 24 2.1 Introduction ............................................................................................................................. 24 2.2 Methods .................................................................................................................................... 30 2.2.1 Participants .......................................................................................................................... 30 2.2.2 Procedure ............................................................................................................................. 31 2.2.3 Measures .............................................................................................................................. 32 2.2.3.1 Practice Questionnaires ............................................................................................. 32 2.2.4 Statistical analyses ............................................................................................................... 34 2.2.4.1 Reliability and validity ............................................................................................... 34 2.2.4.2 Group differences ....................................................................................................... 35 xi  2.2.4.3 Stage 1 .......................................................................................................................... 35 2.2.4.4 Stage 2 .......................................................................................................................... 36 2.3 Results ....................................................................................................................................... 37 2.3.1 Reliability and validity ........................................................................................................ 37 2.3.2 Between group comparisons ............................................................................................... 37 2.3.2.1 Stage 1 .......................................................................................................................... 37 2.3.2.2 Stage 2 .......................................................................................................................... 39 2.4 Discussion ................................................................................................................................. 39 Chapter 3: Coach ratings of skills and their relations to practice, play and successful transitions from youth-elite to adult-professional status in soccer .......................................................................... 52 3.1 Introduction ............................................................................................................................. 52 3.1.1 Conceptual models of high level sports skill development .............................................. 52 3.1.2 Developmental sport activities and their relations to perceptual-cognitive skills ......... 53 3.1.3 Relations between various skills and adult-success (professional status) in soccer ....... 56 3.2 Methods .................................................................................................................................... 58 3.2.1 Participants .......................................................................................................................... 58 3.2.2 Procedure ............................................................................................................................. 59 3.2.3 Measures .............................................................................................................................. 59 3.2.3.1 Practice Questionnaires .............................................................................................. 59 3.2.3.2 Skill ratings ................................................................................................................. 61 3.2.3 Statistical analyses ............................................................................................................... 61 3.2.3.1 Player and coach skill ratings ..................................................................................... 62 3.2.3.2 Group differences in skill ratings................................................................................ 62 3.2.3.3 Relations between skill ratings and soccer activities .................................................. 63 3.3 Results ....................................................................................................................................... 63 3.3.1 Player and coach skill ratings ............................................................................................ 63 3.3.2 Group differences in skill ratings ...................................................................................... 64 3.3.3 Relations between skill ratings and soccer activities ........................................................ 65 3.4 Discussion ................................................................................................................................. 66 Chapter 4: Factors associated with self-determined motivation in youth soccer players: Comparisons across age and skill in a combined prospective and cross-sectional study ................... 77 4.1 Introduction ............................................................................................................................. 77 4.2 Methods .................................................................................................................................... 83 4.2.1 Participants .......................................................................................................................... 83 xii  4.2.2 Procedures ........................................................................................................................... 84 4.2.3 Measures .............................................................................................................................. 85 4.2.3.1 Retrospective questionnaires ..................................................................................... 85 4.2.3.2 Motivation .................................................................................................................... 87 4.2.4 Statistical analyses ............................................................................................................... 88 4.2.4.1 Soccer development and demographics ...................................................................... 89 4.2.4.2 Motivation comparison across age and skill ............................................................... 89 4.2.4.3 Soccer activity relationships with motivation ............................................................. 89 4.3 Results ....................................................................................................................................... 89 4.3.1 Soccer development and demographics ............................................................................ 89 4.3.2 Motivation comparisons across age, time and skill .......................................................... 90 4.3.3 Soccer activity relationships with motivation ................................................................... 91 4.4 Discussion ................................................................................................................................. 92 Chapter 5: Influence of developmental activities and perceptions of challenge in the development of expertise in women’s soccer. .................................................................................................................. 101 5.1 Introduction ........................................................................................................................... 101 5.1.1      Developmental practice pathways for elite sport ........................................................... 102 5.1.2 Challenge perceptions ....................................................................................................... 105 5.2 Methods .................................................................................................................................. 108 5.2.1 Participants ........................................................................................................................ 108 5.2.2 Procedures ......................................................................................................................... 109 5.2.3 Questionnaires ................................................................................................................... 109 5.2.4 Statistical analyses ............................................................................................................. 111 5.2.4.1 Developmental milestones and activities ................................................................ 111 5.2.4.1 Reliability .................................................................................................................. 112 5.2.4.3 Challenge perceptions .............................................................................................. 112 5.3 Results ..................................................................................................................................... 113 5.3.1 Developmental activity milestones ................................................................................... 113 5.3.2 Developmental activity hours ........................................................................................... 113 5.3.3 Within questionnaire player-player reliability ............................................................... 115 5.3.4 Childhood and adolescent challenge ratings ................................................................... 115 5.4 Discussion ............................................................................................................................... 116 Chapter 6: General discussion ............................................................................................................... 128 xiii  6.1 Synthesis of study findings .................................................................................................... 129 6.1.1 Men’s soccer ........................................................................................................................... 129 6.1.1.1 Early majority engagement ..................................................................................... 129 6.1.1.2 Development of soccer-specific skills ...................................................................... 131 6.1.1.3 Soccer activities and Self-Determined Motivation (SDM) .................................... 132 6.1.2 Women’s soccer ................................................................................................................. 134 6.1.2.1 Early engagement pathway ..................................................................................... 134 6.2 Conceptual contributions ...................................................................................................... 137 6.2.1 Early majority engagement in soccer .............................................................................. 137 6.2.2 Discriminability of skill ratings ........................................................................................ 141 6.2.3 Development of skill .......................................................................................................... 145 6.2.4 Motivation .......................................................................................................................... 147 6.2.5 Developmental activity challenge ..................................................................................... 150 6.3 Practical implications ............................................................................................................ 153 6.3.1 Benefits of early majority engagement ............................................................................ 153 6.3.2 Evaluation of the DMSP ................................................................................................... 156 6.3.3 Talent identification .......................................................................................................... 157 6.4 Limitations and future recommendations ........................................................................... 158 6.4.1 Retrospective recall techniques ........................................................................................ 158 6.4.2 Subjective testing ............................................................................................................... 160 6.4.3 Sample size ......................................................................................................................... 161 6.4.4 Cause or consequence of expertise ................................................................................... 162 6.5 Conclusion .............................................................................................................................. 163 References ................................................................................................................................................ 165 Appendices ............................................................................................................................................... 191 Appendix A: T1 questionnaire package ............................................................................................ 191 A.1: Club contact letter .................................................................................................................. 191 A.2:  Parent/Guardian Information Letter ................................................................................... 193 A.3: Coach Information Letter ...................................................................................................... 195 A.4: Practice History Questionnaire (Time 1) .............................................................................. 196 A.5: Football Enjoyment Questionnaire ....................................................................................... 203 A.6: Player skill ratings .................................................................................................................. 205 A.7: Parent questionnaire............................................................................................................... 206 xiv  A.8: Coach Questionnaire and Skill Ratings ................................................................................ 211 Appendix B: T2 questionnaire package ............................................................................................ 219 B.1: Professional club email. .......................................................................................................... 219 B.2: Parent/Guardian Information Letter .................................................................................... 220 B.3: Player Consent Form .............................................................................................................. 222 B.4: Practice history questionnaire (new players) ....................................................................... 224 B.5: Player skill ratings .................................................................................................................. 231 B.6: Follow-up player questionnaire ............................................................................................. 232 B.7: Coach questionnaire and skill ratings ................................................................................... 239 Appendix C: Professional status contact letters .......................................................................... 246 C.1: Adult professional status email .......................................................................................... 246 Appendix D: Study 4 question package ......................................................................................... 247 D.1: Recruitment letter ................................................................................................................ 247 D.2: Women’s soccer questionnaire .......................................................................................... 249   xv  List of tables  Table 1. Soccer milestones and soccer activity average estimates (and SDs) of accumulated hours during childhood (i.e., until T1) for the Academy-only and the professional (pro) groups (groups determined at T2) and for accumulated hours up until T2 for the Professional-youth group, latterly subdivided into Youth-professional only and Adult-professional (at T3). ... 48 Table 2. Average coach and player ratings and within group SDs at T1 for technical, tactical, physical, and creative skills. Comparisons are made across the Academy-only (Acad) with the future Professional groups (Pro) as well as within this latter group dependent on whether they were youth-Professional only (YPro) or adult-Professional (APro). ............................ 73 Table 3. Pearson r correlations between T1 skill ratings for all players assessed at T1 as well as separately for the Academy-only (Acad) group and players selected to play professional-youth at T2 (Pro) and Accumulated (Accum) practice /play during childhood as well as % of overall time in play relative to practice (Play %). ............................................................ 74 Table 4. Pearson r correlations between T2 skill ratings for all professional players (Pro) and for the subdivided, professional-youth only (YPro) and adult-professional (APro) groups and accumulated (Accum) practice /play during childhood (and Play % as a function of play+practice) and across the player’s career. ...................................................................... 75 Table 5. Means, SDs and 95% confidence intervals corresponding to accumulated and weekly hours in practice and play (during childhood and across the player’s careers) for the elite and non-elite groups, as well as start age in soccer activities and number of sports participated in childhood. Statistical analyses are also presented based on independent t-tests (df = 61). Cohen’s d is given as a measure of effect size. ............................................ 98 Table 6. Mean (and SD) self-determined motivation scores of the current U15 & U17 elite and non-elite soccer players at time 1 (T1) and time 2 (T2). ...................................................... 99 Table 7. Mean ages (SD) for soccer milestones for National and Varsity women soccer players and number of other sports participated in childhood (5-12 yr) and adolescence (13-19 yr)............................................................................................................................................. 123 Table 8. Mean (SD) challenge rating and accumulated hours in “high challenge” (ratings of 3 and 4) soccer activities (competition, practice, play), during childhood and adolescence for the National and Varsity women athletes. ................................................................................ 124   xvi  List of figures Figure 1. Developmental Model of Sport Participation. Adapted from Côté et al. (2007). ........... 4 Figure 2. Schematic to show the two stages of analysis designed to discern group based differences based on variables related to soccer milestones, soccer activity amounts and sport diversity. At stage 1, two orthogonal preplanned contrasts were conducted to compare 1) Academy-only to the two professional groups and then 2) the two professional groups to each other. ............................................................................................................................. 50 Figure 3a & b. Average numbers of hours accumulated (and between-subjects SDs) in a) soccer practice and b) play as a function of skill and age. ............................................................... 51 Figure 4. Mean skill ratings (tactical, technical, physical and creative) for the academy-only group (T1 only) and all professional (pro.) youth players (T1 & T2, left-side). The professional youth groups are subdivided on the right side into the youth-professional only, and adult-professional groups (T1 & T2). ............................................................................ 76 Figure 5. Group means (and SD bars) for global self-determined motivation (SDI) and controlled extrinsic motivation (EM) as a function of time (time 1, T1 or time 2, T2) and current (T2) age group (U15 & U17 yr) for the Elite players. ................................................................ 100 Figure 6a & b. Mean (and SD bars) for accumulated hours in soccer activities (practice, play and competition) by National (a) and Varsity (b) soccer players from the under 6 yr age-group (U6) to under 19 yr (U19). .................................................................................................. 125 Figure 7. Mean accumulated hours (and SD bars) in soccer practice and practice in other sports as a function of age period (childhood or adolescence) and skill (National, Varsity). ....... 126 Figure 8 a & b: Mean (and SD bars) for ratings of challenge across age-group and soccer activity (practice, play, competition) for a) National and b) Varsity players. ................................. 127   1  Chapter 1: Introduction  Over the last twenty years, research endeavors have been directed towards determining the type and quantity of practice that leads to expert performance across a variety of domains including music (e.g., Ericsson, Krampe, & Tesch-Römer, 1993), chess (Gobet & Campitelli, 2007), business (Unger, Keith, Hilling, Gielnik, & Frese, 2009) medicine (McGaghie, Issenberg, Cohen, Barsuk, & Wayne, 2011) and sport (e.g., Hodges, Kerr, Starkes, Weir, & Nananidou, 2004; Coughlan, Williams, McRobert, & Ford, 2014). The world of sport, and in particular soccer, offers researchers a valuable environment to facilitate understanding of expert performance, primarily due to the vast worldwide participation base and the relatively small proportion of players that reach the professional (expert) level. According to FIFA, (2007) estimates, only 0.04 % of registered male players are registered as professionals. Thus, there appears to be a necessity to engage in large volumes of soccer specific activity during development (e.g., Ford et al., 2012; Ford, Ward, Hodges, & Williams, 2009; Ford & Williams, 2012; Haugaasen & Jordet, 2012; Zibung & Conzelmann, 2013).  The necessity for research into the talent development process has been exacerbated through a trend toward talent development models/systems recruiting players at increasingly younger ages, with a view to optimizing the volume and quality of practice accumulated during development (Côté, Coakley, & Bruner, 2011). However, the overall efficacy of this early selection approach and its overall benefit to developing expertise, and its psycho-social impact on players is been questioned (e.g., Côté & Erickson, 2015). A number of theoretical models have been forwarded to explain the multidimensional nature of developing expertise. In this thesis, I focus on the three most prevalent models used to describe the pathway towards expertise in soccer; Deliberate Practice Framework (DPF; Ericsson, Krampe, & Tesch-Römer, 1993); 2  Developmental Model of Sports Participation (DMSP; e.g., Côté, Lidor, & Hackfort, 2009; Côté, 1999) and the Early Engagement Hypothesis (Ford et al., 2009).  1.1 Deliberate practice Much of the scientific study of expertise development in sport over the past quarter century stems from Ericsson, Krampe and Tesch-Römer's (1993) classic study of expert musicians, in which they reported a strong (monotonic) relationship between cumulative hours in “deliberate practice” with musical attainment. Deliberate practice was considered hours spent in solitary practice, because this was practice that was considered as structured and engaged in for the primary purpose of improving performance (Ericsson & Pool, 2016). Deliberate practice was further classified as being and activity that is not always inherently enjoyable, requiring full attention and high levels of effort, typically structured by a coach or teacher (Ericsson & Pool, 2016; Ericsson et al., 1993). Following this initial research, a number of researchers showed similar positive relationships between accumulated hours in practice activities (both team and individual) and performance attainment (e.g., Ford & Williams, 2012; Hodges et al., 2004; Ward, Hodges, Starkes, & Williams, 2007; Zibung & Conzelmann, 2013). Recently, the deliberate practice framework has been scrutinized by researchers, particularly in sports (e.g., Côté, 1999; Côté, Baker, & Abernethy, 2007; Côté, Murphy-Mills, & Abernethy, 2012; Hambrick et al., 2014; Macnamara, Hambrick, & Oswald, 2014). Recent meta-analyses have shown that in sports, hours spent in deliberate practice activities have only accounted for 18% of the variance in sport expertise when comparing across a skill class (Macnamara et al., 2014). Importantly, within a skill class of elite sports performers, accumulated deliberate practice accounted for as little as ~1% of the variance in expert status attainment (Macnamara et al., 2016). One explanation for this finding relates to the idea that the 3  necessity for deliberate practice may be circumvented by participating in other, potentially related sports, during childhood (5-12yr) or by participation in play-type activities rather than formal, structured practice (e.g., street soccer/hockey, Côté et al., 2007, 2012). A strong view is that play-type activities during childhood circumvent the need for deliberate practice and that they are necessary for the development of sport expertise (e.g., Côté et al., 2012; Côté & Hancock, 2014). 1.2 The DMSP The Developmental Model of Sport Participation (DMSP; Côté, 1999; Côté et al., 2007, 2012) is one of the most prominent models explaining the pathway towards sports expertise (Bruner, Erickson, Wilson, & Côté, 2010). The DMSP materialized as a counterpoint to deliberate practice theory (Bonneville-Roussy & Bouffard, 2015) and ideas that success and domain-specific expertise were (only) a result of significant investments in sport-specific deliberate practice (i.e., effortful activities designed to improve performance). As a conceptual framework, the DMSP highlights the childhood years (5-12 yr) as a particularly important time period for long-term athlete development and continued engagement in sport (see Côté, et al., 2012). Based largely on the type, intensity and variety of sport experiences during childhood the DMSP offers two primary pathways towards sports expertise: (i) early specialization, involving high volumes of domain specific deliberate practice in one sport from an early age; and (ii) early diversification/sampling, involving participation in a variety of different sports and play activities during childhood and later specialization.  There has been some debate as to which pathway is most beneficial for the attainment of expertise and related, positive youth sport development (e.g., Côté et al., 2012; Ford, Coughlan, Hodges, & Williams, 2015). Some researchers claim distinct benefits from sampling a variety of 4  different sports and play experiences during childhood that are associated with success as well as decreased risk of drop-out or injury (Baker, Cote, & Abernethy, 2003; Bompa, 2000; Côté et al., 2007, Côté, Lidor, & Hackfort, 2009) and potentially increased intrinsic motivation (Côté et al., 2009, 2012). Others present evidence to show that early specialization in the primary domain frequently defines individuals who attain success, based on retrospectively collected developmental histories (e.g., Ford et al., 2009; Helsen, Hodges, Kel, & Starkes, 2000; Law, Côté, & Ericsson, 2008).  Figure 1. Developmental Model of Sport Participation. Adapted from Côté et al. (2007).    5  1.3 Early engagement hypothesis A third pathway exists which offers a less extreme version of the “specialization” pathway. This third pathway is in line with what has been termed “the early engagement hypothesis” (Ford et al., 2009). Two of the primary elements of both the early specialization pathway and early diversification/sampling pathway; practice and play, are important components of the early engagement pathway (Ford et al., 2009). Emphasis is placed not only on early engagement in the primary sport but also engagement in play-type activities (i.e., informal, sport-specific activities that are primarily engaged in for fun, rather than improvement). Evidence in support of this pathway has been primarily from studies of elite soccer players. Elite youth soccer players engaged in more soccer specific practice and more informal “play” activities during childhood than non-elite youth players (Ford et al., 2009). Follow-up of the elite players showed that only time in soccer-specific play during childhood distinguished between players that later achieved a professional contract at age 16 yr from those that did not (Ford et al., 2009). Considering the implications of these ideas for formal versus informal practice time during childhood, additional verification of these findings is important. The importance of one pathway over another is likely predicated by sport, culture and context. For example, sports such as gymnastics and ice skating, which have an early age for peak sport attainment, demand engagement in an early specialization pathway (Côté, Ericsson, & Law, 2005). Equally, on the basis of specific anthropometric or physiological capacities, adult athletes from one sport (e.g., 100m sprinters) can be turned into Olympians (e.g., skeleton) via systematic talent transfer programs in a 14 month period (see Bullock et al., 2009). In sports such as soccer, where the participation base is large and the opportunities to achieve expertise are low, there appears to be a greater necessity for early and large volumes of soccer specific activity 6  during development (e.g., Ford & Williams, 2012; Haugaasen & Jordet, 2012; Zibung & Conzelmann, 2013).  1.4 Measuring success  In previous research, the attainment of a professional contract at age 16 years has been used as a benchmark of expertise “attainment” in adulthood (e.g., Ford, Ward, Hodges, & Williams, 2009). However, this method has been criticized due to its relatively early age of assessment and that many of the players that achieve this milestone, do not go on to play first team, professional soccer (e.g., Swann, Moran, & Piggott, 2015). The transition from receiving a professional youth team player contract to establishing playing time in the club’s first team(s) as an adult can be extremely difficult (e.g., Cook, Crust, Littlewood, Nesti, & Allen-Collinson, 2014). Therefore, there is reason to further define success in professional youth team players beyond the original awarding of a professional contract at age 16 yrs (i.e., adult professional status). A prospective research design would afford opportunity to fully assess the dynamic nature of youth development, allowing comparison between those factors contributing to attainment at the youth and professional level. This type of research design would allow us to see if play differentiates across the successful athletes at this more elite (i.e., adult) level as detailed by previous authors (Ford & Williams, 2012; Hornig, Aust, & Güllich, 2016). 1.5 Discriminability of skill Although skilled performance is multifaceted, the essential components required for soccer success primarily relate to technical, tactical and physical skill (e.g., Meylan, Cronin, Oliver, & Hughes, 2010; Williams & Reilly, 2000). The demands of professional soccer match-play dictate that players possess the ability to execute technical actions consistently (technical expertise) and be able to perform the right action at the right time, adapting quickly to new 7  configurations of play and ball circulation (tactical expertise; Elferink-Gemser, Visscher, Richart, & Lemmink, 2004). Further players must have the physiological capacities to perform high intensity actions with and without the ball for 90 minutes (e.g., Meylan et al., 2010). Due to a general improvement in the organization and fitness levels in professional soccer teams over the last 2 decades, creative skill has become an increasingly desirable facet of sport performance (Memmert, Baker, & Bertsch, 2010). Creativity within team sports is characterized by the generation of several solutions to a problem in specific individual groups or in team tactical games situations which can be denoted as surprising, rare and ore original (Memmert, 2015). Several authors have investigated the predictive and discriminatory capabilities of these fundamental soccer skills at the youth level using isolated tests of technical (e.g., Höner, Votteler, Schmid, Schultz, & Roth, 2014; Huijgen, Elferink-Gemser, Post, & Visscher, 2009;), tactical (e.g., Kannekens, Elferink-Gemser, Post, & Visscher, 2009) and psychological (e.g., Zuber, Zibung, & Conzelmann, 2014) skills to determine future professional status. These studies have generally shown that skilled and future professional players achieve higher scores on individual measures of skill than less skilled and future amateurs. However, differences across skill levels are often dependent upon the sensitivity of the test protocol and the stage of athlete development (e.g., Höner et al., 2014; Kannekens et al., 2009; Vaeyens et al., 2006). For instance, attempts to  track tactical skill development over time (Kannekens et al., 2009) have been criticized due to the reliance on verbal report measures (see Araújo, Travassos, & Vilar, 2010) ). Despite decisions regarding selection to professional status being primarily based upon the subjective opinions of the clubs coaching staff, skills are typically assessed in research through specific, isolated tests, such as tests of sprint speed as a proxy for fitness or number of keep-ups as a measure of technical skill (e.g., (Huijgen, Elferink-Gemser, Post, & Visscher, 8  2010). Only a few researchers have made comparisons between multiple skills to assess which skill(s), at the youth level, best predict later selection to the adult, professional level (for exceptions see; Huijgen et al., 2009; Vaeyens et al., 2006). This type of comparison allows for conclusions about skills which are important (or at least valued) within an advanced group of athletes for later success at adult levels (rather than across different skill levels) and how these earlier identified skills are related to past practice histories. At the youth level, the tendency for talent development programs to focus on more physically mature players is well documented, which is in part explained by the relative age effect (e.g., Helsen et al., 2012; Meylan, Cronin, Oliver, & Hughes, 2010; Mujika et al., 2009). The relative age effect (RAE) relates to a potential 12-month age difference between youth players within a single age category (e.g., U13) based upon his/her proximity to the selection date. Thus, players born towards the beginning of the selection date (e.g., January 1st) may have distinct physiological and cognitive advantages over players born towards the end of the selection period (e.g., December 31st). Such advantages tend to result in a predomination of players born towards the beginning of the selection period being over represented in youth sports (see Wattie, MacDonald, & Cobley, 2015). However, there is a growing body of literature showing that RAE, and concomitant enhanced physical capabilities during youth careers dissipate with age, such that at elite levels of sport the RAE is less pronounced or in some cases reversed as players are compared at the adult level (Cobley, Baker, Wattie, & McKenna, 2009; McCarthy, Collins, & Court, 2015). The rationale behind any reversal or diminished RAE effect is that relatively less mature players develop their technical and tactical skills, as a means of compensation for any lack of physicality, to a greater extent than relatively older players. Therefore, once the less physically mature players “catch up” physically, they are better 9  equipped to cope with the enhanced temporal and cognitive demands associated with higher levels of play (i.e., adult professional level). With this, it may be that any positive relationship between physical skills with attainment of a professional youth contract is no longer present at the adult professional level. It may also be the case that technical, tactical and even creative skills may better differentiate between groups at the professional adult level.  1.6 Developing skill Based upon the DMSP, Côté and colleagues made a series of posulates relating to potential outcomes associated from engagement in an early sampling pathway. These authors proposed that high amounts of deliberate play during early childhood provides a broad base of motor and cognitive experiences that can be later applied to facilitate development of the principal sport of interest (Côté et al., 2009, 2012). Moreover, play can develop these essential skills without the potential drawbacks associated with sport-specific practice (e.g., increased injury, Post et al., 2017), psychological burnout, (see Côté & Erickson, 2015; Côté et al., 2012). Because of the emphasis which has been placed on the importance of soccer play in the development of sport experts (e.g., Ford et al., 2012, 2009; Haugaasen & Jordet, 2012) we were intersted in testing the postulate of Cote et al., by gathering evaluations of cognitive and motor skills considered fundamental for success in soccer (i.e., technical, tactical, physical, and creative skills; Höner, Votteler, Schmid, Schultz, & Roth, 2014; Huijgen, Elferink-Gemser, Post, & Visscher, 2009;  Kannekens, Elferink-Gemser, & Visscher, 2009; le Gall, Carling, Williams, & Reilly, 2010; Memmert, 2015; Williams & Reilly, 2000) and relating these to childhood deliberate play and practice. Relationships between developmental activities and specific skill development has been studied, but rarely have multiple skills been considered across different time periods. In line with 10  the deliberate practice framework (Ericsson et al., 1993), there is evidence that tactical and technical skills are related to accumulated hours in soccer-specific practice (Huijgen et al., 2009; Kannekens et al., 2009). In a cross-sectional comparison of Academy vs. non-Academy players, perceptual-cognitive skills in soccer (i.e., memory and decision making), in children as young as 8 years old, were attributed to hours in high quality sport-specific practice, although in this study play was not measured (Ward & Williams, 2003). As a counterpoint to the studies above, several researchers have shown positive relationships between high volumes of soccer-specific play with performance on laboratory-based assessments of perceptual-cognitive skills, including decision making and anticipation (e.g., Roca et al., 2012, Williams, Ward, Bell-Walker, & Ford, 2012). In both studies, skilled performers also accumulated more hours in soccer practice compared to less skilled counterparts. The authors suggest that the natural affordances of soccer-specific play offer opportunities to innovate, improvise and respond strategically in game-related contexts which mirror the same underlying structures involved in match play (e.g., Williams et al., 2012).  The institutionalization of sport (through formalized coaching academies) has been suggested by some to have limited the capacity for creative decisions (i.e., tactical creativity, Memmert & Roth, 2007). Self-led play experiences are thought to afford opportunities to experiment with new ideas and techniques, more than formal practice, thus potentially enhancing the development of creativity (Côté et al., 2012; Memmert et al., 2010). Conversely, Ericsson (2006) views creativity as the ultimate goal of deliberate practice, in that it is only once an individual has fully mastered their domain, can they ascend to producing performances that are both unique and effective. The empirical literature on the topic of creativity is also mixed, with retrospective accounts of individuals showing creative outcomes associated with engagement in 11  both domain specific practice and play developmental activities (Memmert, 2015; Memmert et al., 2010). The idea that practice can constrain players’ creative impulses is largely based on the assumption that the coaching environment is overly prescriptive and controlling. However, employment of practice principles that provide prescriptive instruction and feedback in a controlled (limited) manner and where the athletes experience some autonomy in their practice decisions (Hendry & Hodges, 2013; Williams & Hodges, 2005), can also promote freedom of expression which may enhance creativity in ball sports (Memmert, 2015; Memmert & Roth, 2007).  1.7 Activity relationships with motivation  Similar to the proposed positive associations between involvement in childhood play with skill development, Côté and colleagues (2012) postulated that “high amounts of deliberate play during the sampling years (~5-12 years) builds a solid foundation of intrinsic motivation…and promotes intrinsic regulation…” (Côté et al., 2012, pp. 278–279). In previous research, I have reported (Hendry, Crocker, & Hodges, 2014) that accumulated hours in childhood play in elite-youth soccer (Under 13/U13– U17 yr) showed no relation to self-determined motivation (SDM; Deci & Ryan, 2017). However, the oldest players (U17 group) were less autonomously motivated than younger age-group players and for these players only there was a negative relationship between accumulated hours in Academy practice and global SDM. One might expect that prolonged exposure would enhance the degree to which the players valued soccer. These findings may be indicative of a general, developmental trend towards interests in other activities and change in motivations during adolescence. Alternatively, motivational-related changes may be specific to elite, youth soccer, reflecting negative effects associated with 12  prolonged, highly competitive, sport involvement and the proximity to rewards associated with adult, professional status.  Motivation is considered a primary influence on the emergence of expertise (Baker & Horton, 2004; Baker & Young, 2014; Ericsson et al., 1993), in that learners must be willing to invest maximal effort over an extended period of time in deliberate practice activities (Ericsson & Towne, 2010). Different types of motivation are required to engage in deliberate practice activities since they are often described as not always being inherently enjoyable (e.g., (Coughlan et al., 2014; Ericsson et al., 1993; yet see Hodges et al., 2004; Ward et al., 2007). Furthermore, the reasons for engaging in deliberate practice may change across age. Early engagement in practice may primarily be for enjoyment in practice itself and interactions with friends (e.g., intrinsic motivation), to enjoyment from the rewards of practice (e.g. improved performance, Ward & Williams, 2003). The complex nature of motivation involved in  practice engagement is encompassed with Self-Determination Theory (SDT; Deci & Ryan, 1985; Ryan & Deci, 2017). 1.8 Self-Determined Motivation Self-determination theory (SDT) is a meta-theoretical framework which assumes that motivation lies along a continuum of self-determination. At the forefront of SDT is the idea that humans have an innate tendency to seek growth and embrace challenges. In this regard, humans are inherently intrinsically motivated (Deci & Ryan, 2011). However, SDT recognizes that the social environment in which people operate can either positively or negatively influence a person’s predilection towards growth and intrinsic motivation (Deci & Ryan, 2002). The interaction of personal and societal factors can influence the degree to which one’s basic psychological needs (competence, autonomy and relatedness) are met. According to Deci and 13  Ryan (2002), when these basic needs have been met, an individual becomes autonomously motivated (i.e., engaging in an activity through a sense of volition), whereas when these needs are thwarted, motivation is controlled (i.e., engaging in an activity through a sense of obligation or pressure) or the individual becomes amotivated (i.e., lacks motivation).  SDT offers a nuanced, multidimensional account of motivation, placing motivation along a continuum of self-determination consisting of three broad types of motivation; intrinsic and extrinsic motivation and amotivation. Underpinning these motivations are six behavioral regulations, namely, intrinsic, integrated, identified, introjected, external and amotivation. Intrinsic regulation occurs when an individual performs for enjoyment or interest. Autonomous, extrinsic motivation, consisting of integrated regulation (i.e., behavior is congruent with individual’s values/beliefs) and identified regulation (i.e., benefits of behavior are important to the individual) are considered to be more self-determined forms of extrinsic motivation, while controlled, extrinsic motivation, consisting of introjected regulation (i.e., individual participates to avoid shame/guilt) and external regulation (i.e., individual participates to achieve reward or avoid punishment) are considered non-self-determined forms of extrinsic motivation (Deci & Ryan, 1985). Amotivation denotes a lack of motivation. Behavioral regulations can be encompassed within two higher order themes; autonomous (including intrinsic, integrated and identified regulations) and controlled motivation (including introjected and external regulations). Generally, autonomous forms of motivation are associated with positive outcomes, whereas controlled motivations are related to largely negative outcomes (Ryan & Deci, 2017).  Although one might expect experts to be more self-determined in their motivational regulations, Gillet and colleagues (Gillet, Berjot, & Gobancé, 2009; Gillet, Berjot, Vallerand, Amoura, & Rosnet, 2012) have shown evidence that athletes may be best characterized by 14  several co-existing types of motivation. These authors showed that high level athletes involved in competitive settings scored highly in both autonomous and controlled motivation and low in amotivation. Competitive settings, such as those inherent in sport, do not typically foster predominantly autonomous (i.e., more self-determined) forms of motivation (Ratelle, Guay, Vallerand, Larose, & Senécal, 2007), which to some extent may explain reports of physical and emotional exhaustion (Gillet et al., 2012) and conflicts that can exist in athletes as they try to balance their sporting life with their relationships (e.g., Vallerand, 2010). Given the findings from our previous work, showing differences across age groups in autonomous motivation, and positive associations between academy practice with controlled motivation, we were keen to conduct further research. This would allow us to assess the dynamic nature of motivations in elite youth athletes as well as potential relationships between developmental soccer activities and SDM.  1.9 Developing female soccer experts Women’s soccer is a rapidly growing sport with approximately 30 million females participating worldwide, making it one of the most popular female sports in the world (FIFA, 2015). Yet, there remains a lack of research into the developmental experiences of elite female soccer players which could be used as a benchmark and guide to future generations of elite players (Gledhill & Harwood, 2014). Much of what is known about the development of elite soccer players is based upon studies of male soccer players (Ford et al., 2009; Ford & Williams, 2012), yet it is unclear whether these finding are generalizable to female players (Gill, 2001).  The financial disparity between the men’s and women’s game allied to a lack of professional opportunities have resulted in fewer structured developmental opportunities for aspiring female players at developmental levels. With respect to play it is unclear to what extent 15  negative socio-cultural expectations may impact the amount of soccer play engaged in by female soccer players (Williams, 2003). With this being the case, it is unclear if and what existing talent development models (e.g., deliberate practice framework; Developmental Model of Sports Participation, Early Engagement Hypothesis) apply to female soccer.  While studies of female soccer experts are scarce, there has been qualitative research directed to uncovering key developmental markers of elite youth female soccer players in England (Gledhill & Harwood, 2014). One interesting developmental trend, viewed as beneficial to development, was early competition experience with boys at an early age (i.e., 10-12 yr). Competitive, co-recreation opportunities tended to cease from 12-14 years, at which point elite players participated in organized female-only teams. Although this allowed them to display competency amongst other female players this period was viewed as being the least beneficial in their overall soccer development. However, as a means of compensation, during this stage, female players engaged in informal soccer activities (“play”) against boys. From age 14-17 yr, participation in female only competition gave players the opportunity to display competency and become selected for national junior squads. Despite the insights derived from the work of Gledhill and Harwood, (2014), as with qualitative research, the sample size was small and the cohort was not likely representative of the North American system where female soccer participation has been a mainstay sport for several decades. The findings were not interpreted within extant models of talent development which would allow testing of ideas related to early specialization, practice diversity and the relative roles of informal and formal practice experiences.  16  1.10 Challenge point framework Because of the increasing institutionalization of many youth, talent development programs, it is becoming increasingly difficult to use differences in the amount of practice hours engaged in by prospective athletes as a predictive measure of success. As such, calls have been made to find new and innovative means of measuring practice quality since practice quantity may not be sensitive to differences across a skill class of high performing athletes (e.g., Ford et al., 2015).  One of the major definitional differences between play and practice is based upon whether or not an individual is primarily engaged in an activity to improve performance (Côté, 1999; Côté & Erickson, 2015; Ericsson et al., 1993). As such, intention to improve performance should more likely be the primary goal of practice (although not necessarily) and would be more prevalent within practice activities (both individual and team) than in play. However, the extant evidence showing positive relationships between soccer specific play and soccer expertise (Ford et al., 2009; Ford & Williams, 2012; Hornig et al., 2016) suggests that skill acquisition can emerge as a by-product from engagement in play, irrespective of any specific intention. Therefore, it may be that the degree of challenge experienced, whether in play, practice or competition, may be as important, if not more, than the specific intention behind engagement in the behavior. Therefore assessing the degree of challenge associated with each developmental activity at different milestones may provide an indication of the quality of the activity and (in retrospect), it’s importance to the attainment of expertise.  One possible method of establishing activity quality is based upon the challenge point framework (Guadagnoli & Lee, 2004). According to the authors, there is a theoretical optimal challenge point that emerges when the constant degree of difficulty inherent in the task (termed 17  nominal task difficulty) is equal to or slightly higher than the skill of the learner relative to the task (termed functional task difficulty). When an individual is at this point, the learner is able to process an optimal amount of information which maximizes the potential for learning. If the level of challenge is either too high or too low learning will not be facilitated to the same extent. The optimal challenge point is considered dynamic so that as the learner becomes more skilled so to can the degree of challenge presented within the task. In this sense there are clear parallels between the challenge point framework with several aspect of Ericsson et al.’s (1993) deliberate practice theory in which the limits of the learners’ current capacities are continually stretched as a means of reducing the propensity for automaticity in practice and performance (i.e., practice designed to improve performance). The applicability to real-world sport settings makes the challenge point framework a particularly relevant framework to assess the degree of challenge and potential for learning in developmental practice activities (i.e., play, practice and competition).  The emergence of skill acquisition through play is congruent with the constraints-led approach to skill acquisition in which environmental, individual and task constraints interact to procure skill acquisition (see Davids, Button, & Bennet, 2008). For example, a child that plays 3v3 or 4v4 street football with a group of equally skilled friends may not necessarily enter the game with the explicit intention to improve performance, yet once the game begins and pride is at stake, then it becomes very likely that he or she will engage fully within the activity. In this case, where the skill of team-mates and opposition is high, the unique interaction of constraints increase the functional task difficulty (i.e., difficulty relative to the performer) and challenge may be optimized. Clearly the converse may be true, whereby the player is not sufficiently challenged and he or she becomes apathetic, lazy and/or disengaged. Although it is likely that players with 18  intentions to improve will more likely seek out high challenge situations, high challenge is not necessarily a characteristic of practice and practice is not necessarily more optimally challenging than play.  Although Ericsson and colleagues (1993) used competition performance as a variable for comparing success; it was not thought to positively contribute to the development of expertise. Their categorization of competition as a work activity that it is time constrained, motivated by external rewards, lacking repeatability and opportunities for experimentation is not universally shared. Several researchers have proposed that experience of competition is a vital component of the talent development process (Abernethy, Farrow, & Berry, 2003; Singer & Janelle, 1999). In youth development circles, competition is often viewed as an extension of the practice session with the experience of playing against unfamiliar opposition, psychological preparation, and dealing with travel, fluctuations in score and crowds thought to facilitate the development of future experts (Cook et al., 2014; Holt & Dunn, 2004). Furthermore, reference to competition as a work activity may be a distinction that fails to translate across domains (i.e., from music to sport), since competition is generally rated as highly enjoyable and relevant activity in sports (Ward et al., 2007). Inconsistency also exists between the soccer expertise literature and the “real world” of youth development with respect to the relative importance of competition (e.g., Ward et al., 2007, Ford et al., 2009).  With respect to empirical research, comparisons across different skill levels rarely show differences with respect to the total number of hours that players engage in competitive match play during development (Ford et al., 2009; Ford & Williams, 2012; Haugaasen & Jordet, 2012). However, the total number and length of each game, during childhood and adolescence are often externally controlled by a league organization. As such, it is not surprising that there are very 19  little differences across skill level for time spent in competition. A “best v best” approach to games is continuously championed by coaches as being central to the development of young players. The idea behind this is that players are repeatedly placed in optimally challenging environments that equally tax players technically, tactically, physically and psychologically, forcing them to adapt and improve. Thus, like practice and play, it is unlikely that all competition is equal. Therefore analysis of the degree of relative challenge during developmental competition may add to the extant literature; informing as to the quality of the experiences (whether play, practice or competition) that define elite players. If a high degree of challenge is a good measure for activity quality, then it is reasonable to suggest that soccer activities, which elicit optimal challenge, will offer the best discriminatory validity with respect to later success at adult levels (or at least better validity when combined with assessments based only on practice (or play) amounts). By measuring perceptions of challenge from childhood to adulthood, we will be able to assess trends as to when (i.e., what age group) and what specific activities (e.g., practice, play, competition) are characterized by high degrees of challenge for the more elite players. As such, we hope to be able to better assess the role of soccer play and competition in elite development (beyond merely measuring quantity of hours). 1.11 Unique contributions to the sport expertise literature  Further to the contribution derived from each individual study, this thesis offers at least four novel contributions to the extant sport-expertise and talent development literature. These unique contributions relate to: i) dynamic changes in motivation across time and insights into childhood activities which best relate to future success through a prospective design; ii) insights into the key skills which distinguish across successful athletes in soccer and various outcomes, afforded through a multivariable measurement approach; iii) evaluations of highly elite samples of 20  athletes, both in the men and women’s game; and iv) the introduction and assessment of challenge, as a measure of practice quality across childhood development.    The prospective research design I have adopted in some of these studies has provided the opportunity to follow a cohort of elite youth soccer players over a five-year period from schoolboy academy player to professional youth players and then on to professional status in adulthood. This type of research is scarce, yet its importance is underlined as it can serve to give assessment of the variables that discriminate across skill at key transitions in the development of soccer expertise. It also helps ward against prematurely identifying factors associated with expertise in youth experts that may not necessarily translate to expertise in adulthood (e.g., Hornig et al., 2016). Criticisms abound of research where elite youth participants, particularly in soccer, are defined as experts (Collins & MacNamara, 2012; McCarthy & Collins, 2014; McCarthy et al., 2015). By following-up a developmentally elite cohort over a five-year period, I was able to assess the factors that discriminated between those that progressed from academy player (~ 14 yr) to youth (~ 16 yr) and adult professional (~ 19yr) status. Consequently, a better description was provided of the multidimensional and dynamic nature of youth development than previous studies in which expertise was based upon cross-sectional comparisons based on elite youth status only. Moreover, by collecting data from players that failed to progress to professional youth status we were able to obtain data normally excluded from an adult expert sample.  This is one of very few studies in which a multivariable approach is used to assess behavioral and psycho-social outcomes associated with engagement in different development activities. Although several studies have investigated relationships between developmental activities with individual skills (e.g., Roca, Williams, & Ford, 2012a; Williams, Ward, Bell-21  Walker, & Ford, 2012) and several propositions have been forwarded as to the expected relationships between variables (e.g., Côté et al., 2009, 2012), I directly examined the relationships between developmental activities associated with key skill and motivational aspects associated with soccer expertise. Given the multifaceted nature of skill in soccer, this type of approach is important at both an applied and theoretical level. A third strength of this thesis is that the participants involved in each study reflected a truly elite sample. One criticism of the sport expertise literature in general has been directed towards loose definitions of expertise, either by including youth athletes as experts or sub-elite athletes as experts (Baker, Wattie, & Schorer, 2015).  I recruited World Class international female players from a team ranked in the top 10 of the world, and adult professional male players from the UK, which represent ~.04% of registered soccer players. Thus, I had access to athletes participating at the upper echelons of sport, such that the research findings should be representative of “genuine” expertise. In respect to women’s soccer, currently there is a paucity of research on the development of female soccer experts (Gill, 2001; Gledhill & Harwood, 2014). Thus, findings from this study carry both significant theoretical and practical significance, given that propensity of existing research about development in soccer is based upon data derived from male participants.  Finally, in line with calls to examine the quality of activities engaged in by experts (e.g., (Ericsson & Pool, 2016; Ford et al., 2015) the inclusion of perceived ratings of challenge across the development time-span offers potential new insights into the development of expertise. The assessment of activity challenge could help provide researchers with more detail as to the role of soccer play or competition in developing skill.  22  In summary, understanding the developmental pathways engaged in by elite male and female soccer players, as well as the behavioral and motivational outcomes associated with these pathways is of considerable practical and theoretical merit. The combination of methods used, quality of experts and outcome measures employed are anticipated to add significantly to the extant sport-expertise and talent development literature.  1.12 Overall research objectives  The overall purpose of the four research studies presented within this thesis was to extend knowledge of talent development in soccer. Specifically, key variables including developmental activities (practice, play and competition), coach ratings of skill (technical, tactical, physical and creative), and self-determined motivation were selected for further examination. Selection of these variables was derived primarily from the DMSP and postulated outcomes associated with an early diversification/sampling pathway (including large volumes of deliberate play and multisport activity; Côté et al., 2007, 2009, 2012; Côté & Erickson, 2015; Côté & Hancock, 2014). These postulated outcomes formed the basis of research questions related to studies 1 - 3. Across these three studies I used a prospective design to assess the contribution of the early specialization and diversification/sampling pathways towards soccer expertise and outcomes associated with involvement in high volumes of play and practice.   For the first three studies, data were collected from elite youth, male soccer players in the UK at three-time separate time points approximately 2.5 yr apart (T1 = October 2011; T2 = February 2014; T3 = September 2016). At T1, players were approximately age 14 years and signed to the youth academies of professional soccer clubs in Scotland. At T2, players were approximately 16 years old and were categorized based upon success in receiving a professional youth contract. Finally, at T3, players were further subdivided based upon success transition to 23  playing adult professional football at first team level (~19 yr). In Study 1, soccer play and practice estimates were assessed in relations to their discriminability across skill based on attainment of youth-professional status (T2) and adult-professional status (T3). In Study 2, I investigated the discriminability of coach ratings of skill (provided at T1 and T2) at the two key transitions in the soccer development pathway towards expert status (i.e., T2, and T3). I also studied the relations between developmental soccer activities and coach ratings of technical, tactical, physical and creative skill in order to determine what childhood activities are most conducive to the development of specific skill sets. The purpose of Study 3 was to test Côté and colleagues (2012) postulating positive relationship between childhood (soccer) play and SDM. I also tested how motivations changed as a function of age and also expertise, through comparisons across time and also with a sub-elite sample. In Study 4, I assessed the developmental activities of adult elite (national team) and sub-elite (varsity) soccer players in Canada in order to determine which developmental pathway best defined success in this population. I also tested the applicability of the challenge point framework (Guadagnoli and Lee, 2004) to assessments of developmental activities engaged in by world class players and whether this variable adds to the discriminability of practice history data. Collectively, these 4 studies provide a comprehensive account of the developmental pathways engaged in by elite male and female soccer players.    24  Chapter 2: Early majority engagement pathway best defines transitions from youth to adult elite men’s soccer in the UK: A three time point prospective and retrospective study    2.1 Introduction Talent development has been described as a complex, non-linear and dynamic process with the attributes for success being multifaceted (Collins & MacNamara, 2012; Williams & Reilly, 2000). Thus far, researchers have attempted to identify commonalities in pathways and profiles of elite athletes to aid an appreciation of key variables that potentially facilitate the development process. These efforts have included descriptions of the developmental activities in which athletes engage (such as practice, play and competition; for a review see Ford, Coughlan, Hodges, & Williams, 2015).  The world of sport, and in particular soccer, offers researchers a valuable environment to facilitate the understanding of expert performance, primarily due to the vast worldwide participation base and the relatively small proportion of players that reach the professional (expert) level. Although the nature of sports expertise is multifaceted, it is relatively well established that extensive experience in the primary domain is largely related to future levels of skill and expertise (Ford, Hodges, & Williams, 2013). In the current study, we use a prospective design, with data collected across three-time points, to evaluate how the early developmental activity experiences (e.g., soccer practice and play) of elite-youth soccer players are related to successfully transitioning from youth academy player (~ 14-15 yr) to the attainment of a professional contract first at age 16-17 yr, and then on to the adult, professional status (i.e., age 19-20 yr).  25  The theory of deliberate practice (Ericsson, Krampe & Tesch-Römer, 1993) has been the prevailing theory upon which the study of expertise development in sports is based. Accordingly, there is a monotonic relationship between expertise and cumulative hours in deliberate practice activities, engaged with the primary intent of performance improvement. Although Ericsson originally made specific reference to the solitary nature of deliberate practice (see also Ericsson, 2014), a number of sports’ expertise researchers showed similar positive relationships between performance attainment and accumulated hours in both team and individual practice activities (e.g., Ford & Williams, 2012; Helsen, Starkes & Hodges, 1998; Hodges, Kerr, Starkes, Weir & Nananidou, 2004; Ward, Hodges, Starkes, & Williams, 2007; Starkes & Hodges, 1996; Zibung & Conzelmann, 2013). In recent years, the deliberate practice framework has been scrutinized by expertise researchers, particularly in sports (e.g., Côté, 1999; Côté, Baker, & Abernethy, 2007; Côté, Murphy-Mills, & Abernethy, 2012; Hambrick et al., 2014; Macnamara, Hambrick, & Oswald, 2014). Recent meta-analyses have shown that hours spent in practice activities have only accounted for 18% of the variance in sport expertise (Macnamara et al., 2014). Further, within groups of elite sports performers, accumulated practice only accounted for ~1% of the variance in skill (Macnamara et al., 2016). It has been argued that the requirement of deliberate practice can be circumvented by participation in other, potentially related sports, during childhood (5-12 yr) or by participation in play-type activities rather than formal, structured practice (e.g., street soccer/hockey, Côté et al., 2007, 2012). A strong view is that play-type activities and diversified sport experiences during childhood circumvent the need for deliberate practice and that they are necessary for the development of sport expertise (e.g., Côté et al., 2012; Côté & Hancock, 2014). 26  The Developmental Model of Sports Participation (DMSP; Côté, 1999; Côté, et al., 2007) offers a framework for describing and studying sport expertise, highlighting childhood (5-12 yr) as an important time period for long-term athlete development and continued engagement in sport (see Côté, et al., 2012). Accordingly, there are two primary pathways towards sports expertise: (i) early specialization, involving high volumes of domain specific deliberate practice in one sport from an early age and (ii) early diversification, involving participation in a variety of different sports and play activities during childhood and later specialization. We also advocate that a third pathway should be considered, that is a less extreme version of the “specialization” pathway. This third pathway is in line with what has been termed “the early engagement hypothesis” (Ford et al., 2009). Because children at an early age may spend the majority of their time within one sport, but not at the exclusion of any others (which is the extreme version of the early specialization route), this route likely captures expertise in many sports. In addition to skill impediments to playing at a high level after a late entry, political or social barriers often prevent later involvement. For example, in sports such as soccer and field hockey, although exclusive specialization in the main sport may not have been until adolescence (e.g., in German field hockey, this was ~15 yr for the Olympic team), athletes had started to play field hockey at the age of 5 yr (Güllich, 2014). In soccer, age of engagement for successful players is routinely about 4-5 yr, but the players continue to engage in other sports throughout their childhood into adolescence (e.g., Ward et al., 2007). Considerable debate exists as to which pathway is most beneficial for the attainment of sport expertise and related, positive youth development (e.g., Côté et al., 2012; Ford et al., 2013). Some researchers claim distinct benefits from sampling a variety of different sports and play experiences during childhood that are associated with future success, motor skill acquisition, 27  decreased risk of drop-out or injury (e.g., Baker, Côté, & Abernethy, 2003; Côté et al., 2007, Côté, Lidor, & Hackfort, 2009) and potentially increased intrinsic motivation (Côté et al., 2009, 2012). In a recent comparison of international athletes, those who had medaled at international competitions had spent more time engaging in sports other than their primary sport in comparison to non-medalists (Güllich, 2016). Others present evidence to show that early specialization in the primary domain frequently defines individuals who attain success, based on retrospectively collected developmental reports (e.g., Ford & Williams, 2012; Hodges et al., 2004; Ward, Hodges, Starkes, & Williams, 2007; Zibung & Conzelmann, 2013) and that time in other sports or in play-type activities fails to differentiate across skill groups or predict future motivation type (e.g., Hendry et al., 2014; Ward et al., 2007).  These apparently contrasting findings detailed above may be related to several factors. For example, sports such as gymnastics and ice skating, which have an early age for peak sport attainment, demand engagement in an early specialization pathway (Côté et al., 2005). Conversely, there is evidence of talent transfer programs which take adult athletes from one sport, and “transfer” them into another due to a high emphasis on specific anthropometric or physiological capacities (e.g., from 100 m sprinter to the winter Olympic sport of skeleton; see Bullock et al., 2009). In sports such as soccer, where the participation base is large and the opportunities to achieve expertise are low (~0.04 % of registered players reaching professional status, FIFA, 2007), there appears to be a greater necessity for early and large volumes of soccer specific activity during development (e.g., Ford & Williams, 2012; Haugaasen & Jordet, 2012; Zibung & Conzelmann, 2013).  Two of the primary elements of both the early specialization pathway and early diversification pathway; practice and play, are important components of the early engagement 28  pathway (Ford et al., 2009). Emphasis is placed not only on early engagement in the primary sport but also engagement in play-type activities (i.e., informal, sport-specific activities that are primarily engaged in for fun, rather than improvement). Evidence in support of this pathway has been primarily from studies of elite soccer players. Elite youth soccer players engaged in more soccer specific practice and more informal “play” activities during childhood than non-elite youth players (Ford et al., 2009). Follow-up of the elite players showed that only time in domain-specific play during childhood distinguished between players that later achieved a professional contract at age 16 yr from those that did not (Ford et al., 2009). Considering the implications of these ideas for formal versus informal practice time during childhood, additional verification of these findings is important. Distinctions between what has been considered play and practice have sometimes been blurred across studies. For example, elements of play can be included in practice and vice versa (e.g. free play during practice and individual practice during play time). Some authors have attempted to retrospectively measure play and practice amounts based on whether activities were engaged in for fun or improvement (e.g., Ford & Williams, 2012; Ward et al, 2007). In the DMSP, Côté and colleagues (Côté, 1999; Côté, et al., 2007, 2012) have argued that engagement in fun, play activities defines a positive, development route. However, retrospectively assessing time spent in activities that are primarily engaged in for fun or improvement is problematic, particularly when based on retrospective, survey methods (Côté et al., 2005). Hence, in the current paper, we distinguish formal soccer practice, defined as structured, coach or adult-led soccer activities, mainly directed to skill improvement (such as drills, small-sided games), from play, defined as unorganized, self-led soccer activities conducted without a coach (including fun games, general kick around or individual play/practice). These distinctions allow athletes to 29  recall activities spent in and outside of formal/organized practice. This is arguably easier, more reliable and potentially of more direct relevance to athletes and parents in helping determine “optimal” amounts of coach-structured practice than judging whether a past activity was engaged in for fun. However, a potential conceptual limitation of this approach is what is judged as “play” will likely include some activities that constitute practice. Being retained by (or brought into) a professional soccer academy at the age of 16 yr has been used as a benchmark of success in UK soccer and as evidence of adult ‘expertise’ (Ford et al., 2009; Ford & Williams, 2012). However, this method has been criticized due to its relatively early age of assessment and the finding that many of the players that achieve this milestone, do not go on to play first team, professional soccer (Swann, Moran, & Piggott, 2015). The transition from receiving a professional youth team player contract (e.g., Under 20/U20 yr team squad) to establishing playing time in the club’s first team(s) as an adult can be extremely difficult (e.g., Cook, Crust, Littlewood, Nesti, & Allen-Collinson, 2014).  In the following study, we tracked elite youth soccer players over a 5 year period to determine how well their developmental practice and play activities (collected between ages 13-15 yr) discriminated later attainment of a professional contract at age 17 yr, and later progression to adult, first-team soccer (~20 yr). This type of extended follow-up into adult-professional soccer has not been conducted. The advantage of this retrospective design is that we have data from individuals that are typically excluded from analysis of solely “expert” groups (i.e., those who drop out or are forced to leave). We additionally collect practice data from the youth professional players (at age 17 yr), such that practice activities can be tracked at two-time points to evaluate how well more recent estimates predict later adult success in comparison to childhood estimates. Because of the elite nature of our sample, we are able to make more refined, 30  within skill-class comparisons concerning success, rather than the more typical between skill class comparisons (i.e., expert-novice) where differences in practice activities are almost always demonstrated. Based upon previous research (e.g., Ford et al., 2009; Ford & Williams, 2012; Rees et al., 2016), we expected that future professional players would show early engagement in their sport, but not early specialization as defined by exclusivity of soccer, and more hours in practice and play during childhood than those not selected as youth (~17 yr) or adult professionals (~20 yr). Because sport-specific play hours had differentiated future successful players from non-successful (released) players in an earlier study with English soccer players at age 16 yr (Ford et al., 2009), we expected sport-specific play hours to discriminate the professional players from the non-professional players. We did not know whether this variable would distinguish at a more elite level, between the Youth-professional only and Adult-professional groups. With respect to early childhood sporting diversity, our predictions were mixed. In line with predictions from the DMSP, more sports engaged in childhood should engender general cognitive and motor competences which would be related to later success. However, in past work with elite soccer players, these groups have been characterized by relatively low amounts of sporting diversity (e.g., Ward et al., 2007). As such, there was also reason to suspect that diversity would be inversely related to successful skill/age-group progressions.  2.2 Methods 2.2.1 Participants Male, elite youth soccer players (N = 102, born in 1996/1997), were recruited from the youth academies of five professional soccer clubs competing at the highest level of youth soccer in Scotland. These players were followed up within the academy system up to 5 years after initial 31  data collection to determine successful age-group progressions onto youth-professional and adult-professional contracts. At T1 (Oct. 2011), players were aged 13-15 yr (M = 14.85 yr; SD = .63) but all were playing at the U15 age group, albeit some players were still eligible to play U14. This sample was drawn from a larger cohort of players (U13-U17 yr), reported in a previous paper (Hendry et al., 2014).  At T2 (May 2014), players were now 16-18 yr (M = 17.34 yr, SD = .69) and eligible to receive a full-time (youth) professional contract. Players that received such a contract we termed “Professional-youth” (n = 26) and those players which did not we termed the “Academy-only” group (n = 76). These latter players were subsequently deselected from their respective academies between T1 and T2 and no further data were collected on these players (although we know that no players from this group progressed to Adult-professional status at T3). At T3 (Oct. 2016), the remaining players were now 19-20 yr (M = 20.56 yr, SD = .61) and were further delineated based on whether they had been selected to play first-team, adult soccer in the UK, what we termed “Adult-professional” (n = 9). Therefore n =17 of the youth-professional sample were not successful at the adult level, termed “Youth-professional only”. No data were collected from youth-professional players that had entered the Academy system after T1 data collection. There were no between group age differences at T1, F<1 (i.e., Academy-only, Youth-professional only or Adult-professional).  2.2.2 Procedure At T1 and T2 separately, participants completed a demographic and retrospective developmental practice activity questionnaire. The T2 questionnaire focused primarily on the developmental activities engaged in by participants between T1 and T2 (~2.5 yr). Academy-only 32  players did not complete the questionnaire at T2 since they were no longer within the Academy system. At T3, only information about success as an adult professional was recorded.  Information regarding success attaining professional status at the youth and adult levels were collected from the academy directors and coaches from participating clubs (who were contacted individually at separate time periods, see Appendix C). At the initial data collection period, participants agreed to be contacted again for future follow up. To aid convergent validity, a sample of parents (T1, n = 15; T2 n = 4) provided estimates of soccer practice and play using the same practice history questionnaire as their sons. Similarly, coaches (n = 8) provided estimates of hours/week in a typical weeks’ organized practice session at T1 and T2 (see Hopwood, 2015 for recommendations regarding these validation methods). Three weeks before data collection, letters/emails were distributed to parents, players and coaches detailing consent procedures. Parents were given three weeks to object from their child participating in the study, otherwise passive consent was assumed. Players, coaches and parents provided written informed consent before completing the questionnaires. Ethical approval was granted by the Research Ethics Board at the University of British Columbia. 2.2.3 Measures 2.2.3.1 Practice Questionnaires  A developmental activity questionnaire adapted from previous research and the Participation History Questionnaire (PHQ, e.g., Ford, Low, McRobert & Williams, 2010, Hodges et al., 2004) was used to gain sport and soccer specific practice activity data. This retrospective, survey method remains one of the best available method for collecting practice history estimates from elite athletes (see, Hopwood, 2015). Basic demographic information 33  relating to participants’ date of birth, start age in soccer and the academy system were first ascertained.  Estimates of time spent in soccer activities across athlete’s careers were collected in table format. Operational definitions and examples of organized practice and play were provided and explained. Practice was defined as activities conducted with a coach/adult used mainly to improve skills. Play was defined as unorganized, self-led activities not conducted with a coach/teacher. Players recorded: (i) number of organized practice sessions/week, (ii) average length of each session and (iii) hours/week in soccer play. These estimates were for a typical week/training session and were solicited from 5 years of age to present in 2-year intervals (i.e. 5–6 yr, 7–8 yr, etc.). Significant breaks from soccer were also recorded. Linear interpolation was used to estimate data during intervening years to enable calculation of accumulated hours in practice and play. Accumulated hours in practice during childhood were calculated by multiplying the number of hours/session by the number of sessions/week. This figure was multiplied by the number of weeks of practice/year (~46 weeks in a typical season), subtracting reported weeks lost through illness/injury. This procedure was repeated for soccer play. In a separate section of the questionnaire, indices of sporting diversity were ascertained. Players first indicated the number and type of other sports engaged in during childhood (from 5 yr – 12 yr). In addition to providing information about start and end age in these sports, players were also asked to provide estimates of average practice hr/week for up to a maximum of 5 sports, which included soccer, for two-time periods (5-8 yr and 9-12 yr). Because separate estimates of hr/week in soccer practice were given within the same questionnaire, these data provided a measure of intra-person reliability (what we refer to as player-player reliability). However, we acknowledge that players may have recognized their previous answers, which may 34  have artificially increased player-player reliability results. We also used data from these questions to determine age of specialization or majority engagement in soccer.  The questionnaire used at T2 consisted of the same demographic and developmental soccer activity questions as T1 (from the first sections only), but differed in that data were only collected in yearly intervals from the end of T1 to T2.  2.2.4 Statistical analyses 2.2.4.1 Reliability and validity Intra-class correlations (ICCs) provided a measure of strength of the association between two measures and percent agreement (PAs) was also calculated to provide a measure of similarity. These were based on methods recommended by Atkinson and Nevill (1998) and Hopwood (2015). Player-player reliability of practice estimates were calculated based on comparisons of accumulated practice hours across the two sections of the questionnaire. Estimates derived from the soccer practice history tables, where data was collected in 2 yr intervals, were compared to estimates from a separate section where data was ascertained pertaining to 4 year time spans (5-8 yr, 9-12 yr). We did expect some variation here due to the fact that players were providing only one average value for longer time periods in this second section. Player-player reliability for practice and play during the overlapping time period between T1 and T2 was also calculated for the Professional-youth players (n=26) who had completed questionnaires at each time point. Player-parent estimates of accumulated hours in soccer activities (practice and play), and player-coach estimates of weekly practice were analyzed at T1 and T2.  35  2.2.4.2 Group differences We used a difference-based ANOVA approach, rather than prediction based, logistic regression analysis due primarily to differences in sample size between the groups and associated issues in testing against the null hypothesis based on expected probabilities. The data were checked for normality using the Shapiro-Wilk test. When the magnitude of skewness was less than 1, indicating only a tendency towards positive skewness (Bulmer, 1979), and there were no significant differences in homogeneity of variance between the groups, we used parametric methods for our analyses based upon the robustness of this technique to violations in normality (Glass, Peckham, & Sanders, 1972; Pallant, 2007). In cases where assumptions were not met, (number of other sports, accumulated hours in play and practice up until T2 for the Professional-youth groups), we performed a log transformation to normalize these data. This allowed us to continue with parametric based methods, particularly more powerful, pre-planned orthogonal group contrasts. The first contrast allowed us to compare Academy-only to the Professional-youth group. For the second contrast, the two professional groups (Youth-professional only and Adult-professional) were compared to each other (see Figure 2 for illustration of analyses). Cohen’s d and partial eta-squared (ηp2) are provided as measures of effect size for between group comparisons. Statistical analyses were conducted using IBM SPSS version 22. 2.2.4.3 Stage 1  Our primary analyses were designed to determine whether childhood practice and sport-specific play estimates (collected at T1), as well as sport-specific demographic information related to start age and sport involvement, were related to eventual expertise at T3 (Academy only, Youth-professional only, Adult-professional). We refer to this as our Stage 1 analysis and we have illustrated the two stages of our analysis in Figure 2. 36  We first used MANOVA to assess for overall differences between the groups for a) Soccer Milestones; including start age in soccer and start age in the academy system, and b) Soccer Activities; including accumulated hours in soccer play and practice. Significant group effects from the MANOVA were followed up with preplanned contrasts based on univariate ANOVAs. We also ran separate univariate analyses with preplanned group contrasts to compare the proportion of time accumulated in soccer play during childhood. Our rationale for including this relative measure, rather than just relying on absolute practice/play amounts, was based on the fact that this variable normalizes across overall practice volume and provides a standard assessment of play versus practice time. This allows consideration of whether formal versus informal soccer activity ratios also vary as a function of later skill attained. With respect to sporting diversity, two analyses were conducted. The number of sports participated in during childhood were compared across the 3 groups in a univariate ANOVA and a 3 Group X 2 Age period (5-8 yr, 9-12 yr), mixed design ANOVA was used to compare average estimates of hours/week in practice across all reported sports (other than soccer). Again, preplanned contrasts were used to test group effects. We provide only descriptive statistics pertaining to age of majority engagement in and specialization in soccer, due to fact that there was little variation in these measures.  2.2.4.4 Stage 2 We also studied differences between soccer activity estimates collected up until T2 between the Youth-professional only and Adult-professional groups, which were the only two groups that had practice data from both these time points. Following MANOVA, independent t-tests were conducted to test for group differences on each variable (practice and play), as well as proportion of time in play. 37  2.3 Results 2.3.1 Reliability and validity Player-coach reliability measures pertaining to soccer practice hours were high at T1 (PA = 91.44%, ICC = .90, p < .001) and T2 (PA = 94.12%, ICC = .94, p < .001). At T1, player-player estimates of time in practice (from estimates within the same questionnaire) were moderate (5-8 yr, PA= 69.8%, ICC =.50, p < .05; 9-12 yr, PA= 72.7%, ICC =.65, p < .05). When we compared across questionnaires, from T1 and T2 for the overlapping year the reliability indices were high (practice, PA = 94.80%, ICC =.92, p < .001; play, PA = 83.43%, ICC = .86, p < .001). Player-parent estimates of accumulated hours in T1 practice (PA = 62.31%, ICC = .60, p = .01) and play (PA = 58%, ICC = .62, p = .01) were moderate at T1 and high at T2, for both practice (PA = 82.46%, ICC = .85, p = .01) and play (PA = 74.0%, ICC = .70, p = .01). 2.3.2 Between group comparisons 2.3.2.1 Stage 1 Average data showing start ages in soccer and in the academy system as a function of group are displayed in Table 1. The MANOVA analysis based on these start ages showed general group differences accounting for ~10% of the variance, F(4, 196) = 3.73, p < .01, Wilks’ λ = .81, ηp2 = .10. Pre-planned comparisons showed that Academy-only players started later in the academy-system than the professional groups (p < .01, d = 1.04) but there were no differences between Adult- and Youth-professional only groups (p =.64, d = .07). The groups did not differ in terms of initial start age playing soccer.  With respect to accumulated hours in childhood soccer activities (play and practice), also shown in Table 1, overall group differences were observed at the p = .05 level, F(4,196) = 2.34, Wilks’ λ = 91, ηp2 = .05. For illustration, we have provided Figures for practice and play 38  comparing accumulated hours across the ages and three groups (see Error! Reference source not found.a&b). The professional groups (Youth and Adult) had accumulated ~300 more hours in practice during childhood than the Academy-only players. Preplanned contrasts confirmed this effect statistically (p = .01, d = .60) but there were no practice differences between the two professional groups. Although play estimates were higher for the Adult-professional group (~700 hours more than both the Academy-only and Youth-professional only, see Table 1 and Error! Reference source not found.b), there was significant variation between athletes in each of the groups such that there were no significant group effects at contrast 1 (p = .30, d = .02) or contrast 2 (p = .14, d = .71). There was also no difference in the relative proportion of time in play versus practice during childhood (between 51% and 59% of overall time spent in play versus practice for all groups).  For the log transformed, number of sports engaged until age 12 yr, there were no significant differences between groups for either contrast (overall ANOVA, F(2, 99) = 2.05, p = .13,  ηp2 = .04; contrast 1, p = .07, d = .05; contrast 2, p = .78, d = .02 ). The non-transformed data are displayed at the end of Table 1, showing general engagement in ~4-5 other sports. Estimates of average hours/week in other sports (summed across all reported sports) from 5-8 years and from 9-12 years are also shown in Table 1, along with hours/week in soccer activities reported for these same time periods, for comparison only. A mixed design ANOVA revealed no group differences in hours/week in other sports across both time periods, F(2, 99) = 1.56, p = .23, ηp2 = .06; contrast 1; p = .09, d = .07; contrast 2; p =.34, d = .03. Participants engaged in fewer other sports from 5-8 yr than 9-12 yr, F(1,100) = 13.64, p < .001, ηp2 = .15, but this did not interact with group, F(2,99) = 1.37, p = .21, ηp2 = .04. Less than 10% of the total sample specialized in soccer exclusively from childhood 39  through to adolescence. No Adult-professional players specialized in soccer and only 2/17 (12%) of the Youth-professional only group specialized. Apart from 4 athletes (<5% of the sample, of which only 1 progressed to Youth-professional), players spent the majority of their time in soccer during childhood (than in other sports). Players spent ~6 times more hours in weekly soccer practice (~3 hr) from age 5-8 yr than the next most popular sport (~.5 hr), and almost 5 times as many hours in soccer practice (~7 hr) as the next most popular sport from age 9-12 yr (~1.5 hr). The type of sports engaged by participants was quite varied, ranging from golf to boxing.  2.3.2.2 Stage 2 We calculated practice and play estimates across the careers of the players (up until T2) for the professional groups. These data are shown in the middle of Table 1. Based on an overall MANOVA, the groups were not differentiated with respect to accumulated hours in all career soccer activities, F(3, 22) = 2.02, p = .14, Wilks’ λ = .78, ηp2 = .22. However, separate analyses on practice and play showed that although the two groups did not differ with respect to practice hours, t(24) = .14, p = .89, d = .06, they were differentiated with respect to sport-specific play, t(24)= 2.37, p = .03, d = 1.00. The Adult-professional group reported more time in play and they also spent relatively more time in play than practice than the Youth-professional only group, t(24) = 2.13, p < .05, d = 0.93.    2.4 Discussion Based on retrospective follow-up of elite-youth elite soccer players in the UK, we evaluated the relative importance of developmental soccer activities engaged in during childhood and adolescence and their relationships with successful transition from youth academy player to professional status in youth and then adulthood. This research was conducted in reference to 40  models of youth-sport development, particularly the DMSP and postulates emanating from this model with respect to practice, play and specialization. We were interested in the role of informal, self-led “play” activities in childhood (within the primary sport) in comparison to more prescribed, organized practice, for determining later soccer success. Analysis of variables related to start age, majority engagement in soccer, and assessment of sporting diversity during childhood allowed additional insight into the role of early sport engagement in later success at the adult level in soccer. Professional-youth players (~17 yr), accumulated more hours in organized practice, but not play, during childhood than Academy-only players. For athletes that attained Adult-professional status (~20 yr), although small in numbers, this group had accumulated more hours in play (up until T2) than the Youth-professional only group. In addition, the Adult-professional group engaged in high volumes of domain specific practice from an early age (i.e., majority engagement), gained early entry into an elite academy and participated in a moderate number of other sports during childhood. In this sense, Adult-professionals potentially received benefits from both the broad base “sampling/play” approach advocated by Côté and colleagues (2007, 2012) as well as the sport-specific performance and learning benefits specified in the early engagement hypothesis (e.g., Ford et al., 2009; Ford & Williams, 2012; Roca, Ford, McRobert, & Williams, 2013; Roca, Williams, & Ford, 2012). The retrospective follow-up  research design fills a gap in the extant literature, helping researchers and practitioners identify the consistencies and dissimilarities in the developmental approaches engaged in by expert performers across key time periods. This approach enabled the inclusion of data from unsuccessful groups of players which are typically not captured when practice estimates are only gained from current elite players. 41  Notably, few participants met criteria pertaining to either a specialization/practice or a sampling/play pathway, as outlined in the DMSP. Rather, participants engaged in high volumes of soccer practice from an early age through adolescence but also participated in several sports other than soccer and partook in high volumes of sport-specific play (particularly after age 10 yr), findings more consistent with the early engagement hypothesis and supported by other soccer expertise research (e.g., Hornig, Aust, & Güllich, 2016; Rees et al., 2016; Zibung & Conzelmann, 2013). Consequently, we recommend that a third pathway be considered within the DMSP framework, which is based on early and majority engagement in a primary sport, but not exclusive engagement as implied by the specialization route. In fact, only ~9 % of the sample truly specialized in one sport based upon this exclusive definition outlined within the DMSP, and none of those players progressed to Adult-professional status. Professional players (Adult & Youth) participated in several different sports during childhood, speaking to a diversified involvement among all players. However, the reported average hours/week in these other sports were ~1-3 hours, compared to ~4-8 for soccer (based on similar 5-8 yr and 9-12 yr time periods respectively). Across all groups, the type of sporting activity varied considerably across participants. Despite our data being limited to soccer, we argue that a blended, early-engagement approach, including elements from both pathways outlined in the DMSP, is likely applicable to other sports where early engagement is a necessity to remain competitive (often due to the popularity of that sport for a culture and hence a large participation base).  Several conceptual mechanisms have been forwarded to explain the importance of play in the development of expertise related to benefits of implicit learning (Masters, Poolton, Maxwell, & Raab, 2008), non-linear pedagogies (e.g., Chow, Davids, Renshaw, & Button, 2013), and the development of creativity (Memmert, Baker, & Bertsch, 2010; for a review see: Côté & 42  Erickson, 2015). From these data we are unable to specify what types of activities that players engaged in during sport-specific play. It is of course possible that play time at these youth elite levels merely affords the athlete a way to accrue extra practice and as such it is the volume, rather than the type that is the critical variable. When looking at the relative proportions and total activity amounts in Table 1, it can be seen that this is indeed a possibility as the Adult-professional players spent ~59% of their time in play activities and organized practice hours did not distinguish the groups.  Notably, our definition of play was based on distinguishing formal (structured/externally-directed) practice hours from more informal (unstructured/self-directed) activities. We did not isolate activities primarily engaged in for fun, from those engaged in for improvement, although fun games/kick around were provided as examples of “play” activities in the questionnaires. This differs from Ford et al. (2009), who specified “fun” as a defining characteristic of play, along with the self-directed nature of the activity (based on DMSP, Côté et al., 2007, 2012). Despite differences in terminology, our practice and play hours were similar to those obtained from English soccer players (Ford et al., 2009), suggesting that the addition of “fun” did not impact estimates of play and that the distinction between self and coach led hours might be driving these differences in activity. Given the age of the children and primary time span under investigation, it is unlikely that self-initiated activities in early childhood (i.e., before 12 yr) are not engaged in primarily for fun. The types of activities that make up play will most likely be a product of the constraints on the child, such as the availability of a yard or nearby park, a flat(ish) surface, siblings/neighborhood peers, school playground conditions and climate.  Early recruitment into a professional soccer academy (within one’s country of origin) is an important factor in achieving success at the youth level, which effectively acts as a precursor 43  to adult professional status (Le Gall, Carling, Williams, & Reilly, 2010; Meylan, Cronin, Oliver, & Hughes, 2010; Zibung & Conzelmann, 2013). This apparent advantage for players entering the UK academy system early (when comparing Academy-only and professional groups) could be viewed positively in that the developmental soccer philosophy of the club and/or country is successful in developing and producing players. Conversely, this could be viewed as an organizational or social bias towards players that have been within the system for a prolonged period and exhibit most strongly the behaviors valued within the system (Cushion & Jones, 2012). Based on the current data, we are unable to ascertain the degree to which the players’ expertise was a cause or consequence of this extended practice and exposure. It may be that the “best” players were selected into the Academies early because they were the most “talented” and that future success reflects this initial talent. Likely there is some interaction between both factors (i.e., initial early prowess/evidence of skill and extended exposure to quality practice; see Boccia, Rainoldi, & Brustio, 2017). In looking at the distributions of birthdates (not included in the results), there was some indication that selection into the Academies and later retention within the professional system was aided by factors related to early birthdate (including physical maturity), referred to as the relative age effect (for discussions of these effects in sport see Cobley, Baker, Wattie, & McKenna, 2009; Helsen et al., 2012; Votteler & Höner, 2017). Compared to an expected 25% based on birthdate distributions across four quartiles, 47% of the Professional-youth players were born in the first quarter of the selection year (Jan-Mar), and this was 33% for the Academy-only and Adult-professional players. This might suggest a relative age effect towards more physically mature players at the professional youth level, but diminishing early age benefits in terms of success as an adult (for similar age-related pattern in German soccer see Votteler & Höner, 2017).  44  In addition to potential strengths of conducting a prospective study, there are some methodological limitations with our study. While the athletes’ developmental profile aligned with those reported in other studies (~5000 hr, Ford et al., 2012, 2009; Hornig et al., 2016; Zibung & Conzelmann, 2013), the developmental pathway of players who progressed to a professional status might still just reflect unique features of the cohort. This is further compounded by the relatively small number of players that progressed to professional status (both at the youth and adult level), which causes issues for both power and external validity. Despite these concerns, we would argue that the elite status of our group, the retrospective follow-up design and the natural attrition associated with elite youth development in soccer helps to underscore the validity of the methods and subsequent conclusions derived from these data. Although only less than 10% of participants made it to the adult professional level, data suggest that less than 20% of current European first team squads are comprised of academy players across multiple age groups (Poli, Ravenel & Besson, 2015). Thus, to have 9% from our sample at U15 yr progress to professional adult status is likely quite reflective of what is typical in elite professional soccer where adult players are brought in to clubs from other academy systems across their own country as well as from different countries. There are also potential issues associated with retrospective recall of practice activities, even though our questionnaires have been shown to provide valid estimates of developmental activites (Hodges, Huys, & Starkes, 2007; Hopwood, 2015) and steps were taken to validate these estimates in the current study. In comparison to many studies relying on this retrospective technique, players in the current study were still children at the time when estimates were collected. Therefore, these individuals were questioned about ages that were in close proximity to their current age (at both T1 and T2). Although activity estimates were collected via 45  questionnaires, they were administered in an environment that was facilitated by the study-team in order to overcome potential issues from collecting such data from children. Both parent and coach estimates corroborated estimates given by the children (being moderate to high), although the strength and similarities were higher for formal practice hours rather than play hours and for more recent estimates. Player-player estimates from different sections of one questionnaire and across different time points were all moderate to high in terms of similarity and strengths. Moreover, there was no reason to suspect group differences in recollection/estimates, given that the sample was relatively homogenous in terms of age and experience at T1.  One of the strengths of our study was that we determined adult success at ~20 yr, based on match play in the first team of a professional soccer club in the UK. In prior work, age 16 yr, when players receive professional youth contracts, is taken as an indicator of “adult” success. Although there is the possibility that some players will still progress to achieve adult professional status after our cut-off date (Oct., 2016), we know that no players from the original Academy-only group progressed to Adult-professional status at T3. Although continued follow-up of all players from the original sample (n = 102) at age 25-26 yr would be one way to better check adult success, it is unlikely that any more, or more than 1 or 2 would make it into this elite, adult sample. A more likely scenario is that some of the original sample would no longer be playing first team professional soccer, thus decreasing our sample size from n = 9, causing issues for further analyses. In future work, one way to potentially increase sample size and get a better understanding of pathways to success, would be to measure practice history profiles of new players at each stage of follow up as well as to find a way to track players which get deselected. In summary, players that successfully transition to Adult-professional status gained early entry into a soccer academy, engaged in high volumes of soccer specific practice and play 46  activities throughout their youth careers (defined by majority engagement in soccer from childhood), and participated in several sports other than soccer during childhood. Thus, successful athletes are best characterized by an early (majority) engagement pathway. In this hybrid of pathways, players may be receiving the skill related benefits associated with deliberate practice (Ericsson et al., 1993), and early engagement (Ford et al., 2009), as well as the reductions in injury/burnout associated with the early diversification pathway (Côté et al., 2007, 2012). Notably, accumulated hours in soccer play did not distinguish Academy-only players from those that received a professional youth contract at ~17 yr. However, when comparing players who further progressed to the adult professional level at ~20 yr, play hours accumulated in later childhood (>10 yr) did discriminate across groups (both in absolute terms and proportionally relative to practice). In future work, it will be important to consider how these data generalize to other youth Academy sport systems, where elite athletes are identified at a young age, yet only a restricted number progress into adult elite systems (both in team sports such as rugby or hockey and also individual sports such as tennis). It would also be of interest to identify and compare adult elite athletes who have progressed through traditional (academy) and non-traditional systems to determine what types of activities are potentially compensating for academy-related practice hours and/or show positive transfer.  With respect to models of skill development, the question as to whether emphasis should be placed on early diversified sports experiences and self-led play, outside of more externally-regulated, structured activities for later adult success continues to be debated. When combined with early majority engagement in a primary sport, there may be some potential ancillary benefits of engaging in other sports, but the data in our study do not show any benefits with 47  respect to differentiating adult from youth professionals. Given that players engaged in high volumes of soccer activities from an early age, it appears prudent to advise against participating in other sports at the expense of soccer specific activities, at least if adult success is the ultimate goal. Whether this generalizes to other sports, where participation rates are lower, such as rugby, hockey or even women’s soccer remains to be tested. Unstructured, self-led “play” hours in later childhood distinguished across groups at the highest level, suggesting that there is something about these types of activities that predicts success. However, the Adult-professionals amassed approximately 600 mores hours in soccer activities in general compared to the Youth-professionals. Thus, there is rationale to argue that play hours simply add to practice volume in a general sense and the nature of the activity (i.e., self- or externally-led or engaged in for fun or improvement) is of less importance. This may be particularly true for this population whose formal practice hours are highly regulated and hence less likely to show variation between players. Moreover, doing more, including engagement in activities outside of formal, structured practice may just reflect more intrinsically motivated players or players who show greater competence. As such, more hours in play might reflect greater initial prowess and/or high motivation in addition to helping to develop these skills and positive forms of motivation. In future work, there will be a need to better understand the athletes reasons for engagement in self-led play and the specific activities that comprise this category as a function of age of development, perhaps through a combination of diary, questionnaire and interview methods.        48  Table 1. Soccer milestones and soccer activity average estimates (and SDs) of accumulated hours during childhood (i.e., until T1) for the Academy-only and the professional (pro) groups (groups determined at T2) and for accumulated hours up until T2 for the Professional-youth group, latterly subdivided into Youth-professional only and Adult-professional (at T3). ____________________________________________________________________________________________________________      Academy-only Professional-youth          Youth-Pro only  Adult-Pro         (n = 76)        (n = 26)     (n = 17)     (n = 9)     ____________________________________________________________________________________________________________ Soccer milestones (yr) Start age in soccer 5.42 (1.63)  5.08 (1.35)   5.18 (1.51)  4.89 (1.65)   Start age in Academy 11.39 (2.09)  9.53 (2.18)   9.59 (2.20)  9.44 (2.24)    Soccer activities Accumulated hr until T1  Practice estimates 1221.73 (523.06) 1529.00 (493.76)  1518.65 (517.64) 1548.56 (474.68)   Play estimates  1736.20 (1221.79) 1945.77 (952.83)  1701.47 (791.47) 2407.22 (1103.09)  % time in play      54.98 (14.87)     54.40 (12.94)      51.84 (12.70)      59.22 (12.71)  Total practice + play 2960.30 (1495.85) 3474.77 (1148.95)  3220.17 (1017.82) 3955.78 (1286.55)  Accumulated hr until T2  Practice estimates     2710.86 (829.10)  2737.10 (205.03) 2661.28 (281.79)   Play estimates     3127.17 (1426.64)  2688.57 (318.67) 3955.71 (437.97)  % time in play         52.26 (13.05)      48.56 (13.53)     59.26 (9.04) Total practice + play    5838.03 (1684.61)  5425.67 (1604.99) 6616.99 (1634.74)       49  Sport diversity indices Hr/week 5-8yr: Soccer  2.88 (2.18)    4.39 (3.18)    4.63 (3.82)    4.06 (2.49) Other sports 1.25 (1.07)      .97 (1.87)      1.06 (1.19)    1.10 (1.87) 9-12yr: Soccer  6.38 (3.75)    8.34 (5.37)    9.23 (5.48)    7.13 (5.33)  Other sports 3.83 (3.80)    3.06 (3.17)    3.23 (2.30)    3.50 (3.35) Total # other sports (range) 5.42 (1-13)    4.27 (1-8)    4.18 (1-8)    4.44 (2-7)  __________________________________________________________________________________________________________       50    T1 (~15 yr)   T2 (~17 yr)  T3 (~20 yr) Youth-elite (Academy) Youth-professional Adult-professional    Variables:       Groups: Stage 1 a) Start milestones      Academy-only   b) Soccer activities (T1)         contrast1   Youth-professional   c) Sport diversity indices     Adult-professional     Variables  Groups: Stage 2     Soccer activities Youth-professional (T1-T2)  Adult-professional  Figure 2. Schematic to show the two stages of analysis designed to discern group based differences based on variables related to soccer milestones, soccer activity amounts and sport diversity. At stage 1, two orthogonal preplanned contrasts were conducted to compare 1) Academy-only to the two professional groups and then 2) the two professional groups to each other.     contrast2      51  a)   b)   Figure 3a & b. Average numbers of hours accumulated (and between-subjects SDs) in a) soccer practice and b) play as a function of skill and age. 050010001500200025003000350040005 6 7 8 9 10 11 12 13Hours in soccer practiceYearsAdult-ProYouth-Pro onlyAcademy-only050010001500200025003000350040005 6 7 8 9 10 11 12 13Hours in soccer activityYears     52  Chapter 3: Coach ratings of skills and their relations to practice, play and successful transitions from youth-elite to adult-professional status in soccer   3.1 Introduction Current models of youth development designed to identify and nurture prospective soccer players have been criticized due to their overall inefficiency and lack of predictive utility (e.g., Barreiros, Côté, & Fonseca, 2014). In this paper, we evaluate elements of high level skill development models through a prospective research design. We test whether ratings of soccer skills (technical, tactical, physical and creative), collected from elite-youth players and their coaches, differ across future skill groupings at two key transitions in elite soccer; from youth-academy to youth-professional, and then on to adult-professional. We also assess the relationships between these skills and engagement in childhood soccer activities (soccer specific play and practice).  3.1.1 Conceptual models of high level sports skill development The Developmental Model of Sport Participation (DMSP, see Côté, Murphy-Mills, & Abernethy, 2012) is a prominent model of sport-expertise development (Bruner et al., 2010). The DMSP emerged as a counterpoint to deliberate practice theory (Ericsson, Krampe, & Tesch-Römer, 1993). According to this latter theory, there is a monotonic relationship between skill attainment and hours accumulated in highly effortful (cognitive and physical), relevant and purposeful practice, designed with the intention of performance improvement (see also Ericsson & Pool, 2016). Practice activities are proposed to be engaged primarily for improvement, rather than for their inherent enjoyment. As such, early involvement in domain-specific deliberate practice activities is considered necessary for later skill attainment. The DMSP departs from      53  early involvement in deliberate practice activities (termed the early specialization/practice pathway) by suggesting a second, alternative pathway to high levels of skill attainment. This second pathway, termed the early diversification/play pathway, consists of multi-sport involvement and participation in high volumes of deliberate play activities in both the primary sport and other sports from an early age, followed by specialization during adolescence. Deliberate play activities are assumed to be self or peer-led, highly enjoyable, unstructured games, using rules adapted from the adult form that are typically monitored by the athletes (Côté et al., 2007). Empirical evidence has been provided showing significant support for the important role of practice in later skill attainment (for reviews see Baker & Young, 2014; Ford, Coughlan, Hodges, & Williams, 2015). There is also evidence that high level skills are promoted by this second diversification/play pathway (e.g., Baker & Côté, 2003; Berry, Abernethy, & Côté, 2008; Hornig, Aust, & Güllich, 2014; Soberlak & Côté, 2003). A third pathway exists, which provides a more nuanced version of the above two pathways and seems to perhaps provide a better description of development in elite, professional soccer. The “early engagement hypothesis” recognizes the importance of early engagement in sport-specific practice and minimal diversification (e.g., Ericsson et al., 1993), as well as the importance of early sport-specific play in developing expertise (Ford et al., 2012, 2009; Ford & Williams, 2012; Haugaasen & Jordet, 2012; Zibung & Conzelmann, 2013). Thus, it is the combination of early engagement in the sport, along with both sport-specific practice and play, which leads to later success at adult levels.  3.1.2 Developmental sport activities and their relations to perceptual-cognitive skills In several theoretical and empirical papers, early involvement in diversified and play-type activities have been proposed to facilitate the development of psychological skills including;      54  motivation (e.g., Côté, Lidor, & Hackfort, 2009; Côté, Murphy-Mills, & Abernethy, 2012), social skills (e.g., Côté & Erickson, 2015), decision making (e.g., Baker, Côté, & Abernethy, 2003; Berry et al., 2008; Roca, Williams, & Ford, 2012) and creativity. There has been some evidence to support these ideas, whereby positive relationships have been demonstrated between engagement in both non-sport specific and sport-specific play activities and perceptual-cognitive skills, provided that the sports share similar underlying structures (e.g., participation in related invasion games, (Baker et al., 2003; Berry et al., 2008; Causer & Ford, 2014).  The accumulation of high volumes of soccer-specific play has been related to performance on laboratory-based assessments of perceptual cognitive skills; including tactical decision making and anticipation (e.g., Roca, Williams, & Ford, 2012; Williams, Ward, Bell-Walker, & Ford, 2012). In these studies, skilled performers also accumulated more hours in soccer practice compared to less skilled counterparts. The authors suggest that the combination of practice and play offer opportunities to innovate, improvise and respond strategically in game-related contexts, which mirror the same underlying structures involved in match play (e.g., Williams et al., 2012). Similar ideas regarding play has been forwarded by Côté and colleagues, albeit they suggest that these same motoric and perceptual-cognitive skills can be developed predominantly through engagement in other sports and play activities, without any of the potential drawbacks associated with sport-specific practice (e.g., increased injury, psychological burnout, see Côté & Erickson, 2015; Côté et al., 2012).  Creativity, characterized by the generation of several solutions to a given problem which can be denoted as surprising, rare and or original, is viewed as a highly desirable characteristic in soccer (Memmert, 2015; Memmert et al., 2010). As a higher order component of expertise, Ericsson and Lehman, (1999) hypothesize that creativity can only be attained once an individual      55  has mastered the skills within their domain through sustained practice. As a counterpoint to this view, optimal creative development is thought to be developed through initial predominant involvement in deliberate play and diversified experiences during childhood, and later increased specialization during adolescence (Santos, Memmert, Sampaio, & Leite, 2016). Deliberate play activities are thought to afford more opportunities to experiment with new ideas, movements and techniques, more than structured deliberate practice activities (Bowers, Green, Hemme, & Chalip, 2014; Côté et al., 2012; Memmert et al., 2010). Some empirical support has been provided in support of these claims, showing relations between hours accumulated in deliberate play activities and creativity in team sports, including soccer (Koslowsky & Botelho, 2010; Memmert et al., 2010). Relatedly, Bowers and colleagues (2014) directly compared participation in unstructured (play-type) versus structured (practice-type) sport activities. Results showed that participating in more unstructured activities during childhood was associated with higher scores of general, non-sport specific creativity.   Despite these few correlational studies showing support for play as a precursor to later creative play, some authors have argued that the benefits of engagement in play are overstated (MacNamara, Collins, & Giblin, 2015). Although these authors acknowledge that there may be some psycho-social benefits to play, they argue that cognitive and motor skill development is best served via effective practice design, instruction and feedback from expert practitioners (see also Ericsson et al., 1993). In keeping with this viewpoint, published reports suggest that tactical and technical skills are related to hours accumulated in soccer-specific practice (Huijgen et al., 2009; Kannekens et al., 2009). In a cross-sectional comparison of academy and non-academy soccer players in the UK, perceptual-cognitive expertise in soccer (i.e., memory and decision making), in children as young as 8 yr, were attributed to hours in high quality sport-specific      56  practice (Ward & Williams, 2003). However, no data were gathered on sport-specific play in this latter study.  3.1.3 Relations between various skills and adult-success (professional status) in soccer Several authors have investigated the predictive capabilities of individual soccer skills at the youth level to determine future professional status; including technical (e.g., Höner, Votteler, Schmid, Schultz, & Roth, 2014; Huijgen, Elferink-Gemser, Post, & Visscher, 2009), tactical (e.g., Kannekens, Elferink-Gemser, Post, & Visscher, 2009) and psychological (e.g., Zibung, & Conzelmann, 2013) skills. Although assessments of skills have generally been based on specific (objective) tests, decisions regarding selection to professional status are typically based upon the subjective opinions of coaching staff (Cushion, Ford, & Williams, 2012). Further, the external validity of testing procedures used to isolate specific soccer skills from the game have been criticized as they rarely reflect complexity and contextual factors involved in match play (for reviews, see Ali, 2011; Unnithan et al., 2012). To date, few researchers have made comparisons between multiple skills to assess which one(s) differentiate at the youth level, and later impact selection to the adult, professional level (for exceptions, see Huijgen et al., 2009; Vaeyens et al., 2006). Such comparison allows for conclusions about which skills are important (or at least valued) within an advanced group of players for later success.  In this study, we assessed relationships between soccer-specific practice and play activities during childhood and coach evaluations of technical, tactical, physical and creative skills, which are considered fundamental for advancement in the sport (Höner et al., 2014; Huijgen et al., 2009;  Kannekens et al., 2009; le Gall, Carling, Williams, & Reilly, 2010; Memmert, 2015; Williams & Reilly, 2000). We studied elite-youth soccer players in the UK at three-time points over a 5 year period to determine how well coach skill ratings at age ~13 yr      57  (time 1, T1) related to attainment of youth-professional status at ~16 yr (T2) and progression to adult-professional status at ~19 yr (T3). We focused our analyses on these skill evaluations, their ability to distinguish across groups and their relations to accumulated practice and play amounts. Practice and play activity data have been presented elsewhere (Hendry & Hodges, 2018). Consistent with deliberate practice theory, we predicted that early exposure to high amounts of soccer practice would be related to all assessed skills. Yet, there was reason to expect that increased time in soccer-specific play activities in childhood would be positively related to ratings of tactical and creative skill, arguably differentiating successful from non-successful athletes at the highest levels (e.g., Memmert et al., 2010; Roca et al., 2012: Williams et al., 2012). We expected that the future professional groups would be rated as more skillful in general, particularly the adult-professional group. Our predictions about group differences in individual skills were somewhat exploratory. There was reason to expect that tactical/creative skills would better distinguish future adult professional players (see Memmert, 2015). Due to a decline in maturational/relative age advantages as players progress to adult levels (especially within select groups of elite athletes to start), we did not expect physical skills to distinguish across the groups (e.g., Helsen et al., 2012). Time 2 ratings (at age ~16 yr) were expected to better distinguish across the groups of professional players than T2 ratings, due to their proximity to final skill achievement at the adult level. Over time, the ratings were expected to increase for the groups which successfully progressed to adult level, but either not change or decrease for the players who made it to the youth-professional only.       58  3.2 Methods 3.2.1 Participants Altogether, 102 male, elite youth players participated from the youth academies of five professional soccer clubs in Scotland. They were born in 1996/1997 (Mage =14.85 yr., SD = .63; range 13-15 yr). Only signed players from these professional clubs, which were competing at the highest level of youth soccer in Scotland, were recruited to participate. As part of an earlier study on motivation with a larger age-range of athletes, these 102 youth players had completed questionnaires pertaining to childhood soccer practice and motivation for playing (Hendry et al., 2014). At this time, the players and the coaches also provided evaluations about their soccer-related skills (T1 = Oct. 2011). We followed up these individuals over two further time points, which were ~2.5 years apart. The second time point was in May 2014 (T2), when some of the players had been offered a full-time professional contract at age 16 yr (n = 26; Mage = 17.34yr., SD = .69;  range 16-18 yr). Coach and player skill evaluations were further gathered at this time. Players that did not progress to T2 we termed the “academy-only” group (n = 76) and no further data were collected from this deselected group, whereas players that progressed were termed youth-professional players.  The third-time point was in October 2016 (T3), which corresponded to the time when the youth-professional players would have been offered adult professional contracts to play first team soccer in the UK at age 19 yr (n = 9; Mage = 20.56yr., SD = .61; range 19-20 yr). As such, the youth-professional players were further subdivided into the adult-professional group and “youth-professional only” group (n=17).       59  3.2.2 Procedure At T1(all players) and T2 (professional-youth players only), participants completed a practice history questionnaire (see details below), providing estimates of hours spent in soccer practice and soccer play in childhood (5-12 yr) and up until their current age where relevant (see Appendix A). The T2 questionnaire (see Appendix B) was a truncated version of the T1 questionnaire, focusing on soccer activities between T1 and T2 (~2.5 yr). Coaching staff, (including age group coaches and academy directors, N = 8) provided ratings of technical, tactical, creative and physical skills for each player at T1. At T2, coaches (n = 6) from the original cohort repeated skill ratings on the selected players. Players also provided this information at T1. All coaches were fully licenced to UEFA standards and ratings were made relative to average standards of academy players at a similar age.  At T3, only information concerning selection to adult professional, first-team soccer was recorded, as provided by academy directors and coaches. In accordance with procedures outlined in study 1, parents provided passive consent for their sons to complete the practice history questionnaires before players gave written informed consent. During initial data collection (T1), participants agreed to be contacted again for future follow up. All procedures adhered to the guidelines of the lead university ethics’ board. 3.2.3 Measures 3.2.3.1 Practice Questionnaires  An adapted version of the Participation History Questionnaire (PHQ, e.g., Ford, Low, McRobert, & Williams, 2010; Ward, Hodges, Williams & Starkes, 2007) was used to ascertain career practice and play estimates. This method of collecting practice history data through prompted recall is regarded as one of the best available methods for obtaining developmental      60  activity histories from elite and developmental athletes (see, Hopwood, 2015). In the questionnaire, operational definitions and examples of organised practice and play were provided and explained to all participants by the research team. Practice was operationally defined as organised soccer-specific activities conducted with a coach/adult engaged in with the primary intention to improve skills/performance (e.g., group soccer practice). Play was operationally defined as unorganised soccer-specific activities that were self/peer-led and were not conducted with a coach/teacher (e.g., street soccer). Further examples of each activity type were given by the research team. Players recorded: (i) number of organised practice sessions/week; (ii) average duration of each session; and (iii) hours/week in soccer play. The estimates were for a typical week/training session and solicited from 5 years of age to present age in 2-year intervals (e.g., 5–6 yr, 7–8 yr). This method of ascertaining data in 2 year intervals is mostly to aid in efficiency in data collection, especially important for young children. Significant breaks from soccer were recorded and linear interpolation was used to estimate play and practice data during intervening years. Accumulated hours in each activity were calculated by multiplying hr/session, by number of sessions/wk, by weeks of activity/yr (~46 wk in a season), subtracting reported weeks lost through illness/injury.  Reliability and validity of the soccer activity estimates were established through procedures detailed in Hendry and Hodges (2018). These pertained to player-player (within the questionnaire and across T1 and T2), player-parent (T1, n = 15; T2 n = 4) and player-coach (n=8) reliability based on the strength and similarity of the practice and play estimates (see recommendations by Atkinson & Nevill, 1998; Hopwood, 2015). All relationships were moderate to high and increased with proximity to current age.       61   3.2.3.2 Skill ratings Coaches used a 5-point scale (1 = poor to 5 = excellent) to rate each player relative to other players at an age and skill-appropriate level at T1 and T2. The use of subjective coach ratings to assess skill have been used in previous expertise research as a valid and reliable method of skill in the absence of more robust or valid measures of soccer skills (e.g., Ali, 2011; Unnithan et al., 2012). Operational definitions of skills were provided and explained by the research team to aid interpretation by all participants.  Tactical skills were defined by the player’s ability to make fast and accurate decisions in relation to the ball, team-mates and opposition (e.g., Elferink-Gemser et al., 2004). Technical skills related to the basic motor aspects of skill including; passing, dribbling, shooting or kicking (e.g., (McMorris, 2004). Physical skills referred to the conditioning necessary to play effectively, such as endurance, speed and strength (e.g., Baker et al., 2003). Creative skills were defined by a person’s overall flair and originality in making decisions and displaying unusual skills (e.g., Memmert et al., 2010). At T1 only, players were asked to rate their own skills relative to others of a similar age and skill. Although we expected the ratings for the players to be higher overall than the coaches, this allowed for some validity check. 3.2.3 Statistical analyses Although the skill ratings data were not normally distributed based on the Shapiro-Wilks test, the magnitude of the skewness for each variable (T1 and T2) was less than 1, (Bulmer, 1979), indicating only a tendency towards positive skewness. Further, there were no significant differences in homogeneity of variance between the groups. Taking these factors into consideration and the fact that the sample size was greater than 40, we chose to use ANOVA      62  methods for our analyses due to the robustness of this technique to violations in normality, especially when the other conditions detailed above are met (Glass et al., 1972; Pallant, 2007). 3.2.3.1 Player and coach skill ratings  To test for differences between player and coach ratings of skill at T1, we ran a 2 Role (player/coach) x 4 Skill-type, mixed ANOVA, with follow up Tukey HSD post-hoc tests. Intra-class correlations (ICCs) were also calculated to index the strength of the relationship between the player and coach ratings as well as percent agreement (PA, based on division of the smaller by the largest value for each pair, multiplied by 100) to indicate similarity between the ratings. Similar analyses were also conducted on the T1 and T2 ratings of the same players as a further index of reliability, although we did expect ratings to change over time. 3.2.3.2 Group differences in skill ratings MANOVA analyses were used to compare the three “future” groups (academy-only, youth-pro. only or adult pro.) for coach ratings of the 4 skills at both T1 and T2. Separate ANOVAs based on high powered, orthogonal, pre-planned contrasts at T1 allowed us to compare; 1) Academy-only players with future professional players and 2) the two professional groups to each other (youth-pro. only & adult-pro.). For T2 ratings, follow up independent t-tests were used to distinguish across the two professional groups for each of the skills. To determine if and how the ratings had changed over time, exploratory RM ANOVAs were conducted on each skill comparing across Group (youth-pro. only, adult-pro.) and Time (T1, T2). Here we used Bonferroni adjusted p values (p = .05/4 = .013) to interpret these data. Partial eta squared (ηp2) are provided as measures of effect size for ANOVAs involving more than 2 means and Cohen’s d for pairwise comparisons.       63  3.2.3.3 Relations between skill ratings and soccer activities The hours accumulated in soccer practice and play in childhood were correlated with all skills using Pearson’s correlations. In order to normalize for differences in activity amounts, we also assessed relations based on relative amounts in play versus practice (i.e, % time in play). Separate correlations were run on coach skill ratings collected at T1 and T2 (career estimates were also correlated with ratings at T2). Separate analyses were also run on the academy-only (T1 ratings) and professional players (for T1 and T2 ratings). Due to differences in “n” between groups, we focus our reporting on rs >.30, deemed a medium effect size (Cohen, 1992). 3.3 Results 3.3.1 Player and coach skill ratings At T1, coaches rated the players on average as moderately skilled (M = 3.4, range = 2-5), in reference to other “academy” players at the same age (see Table 2). Technical skill received the highest rating (M = 3.53) and tactical skill the lowest (M = 3.26). These ratings were comparable to those given by the players, although as expected, mean ratings were higher among the players (M = 3.70). This “role” effect was statistically significant, F(1, 2.33) = 34.19, p < .001, ηp2 = .11. Although there was a main effect for skill-type, F(1, 2.33) = 15.53, p = .01, ηp2 = .12, with lower ratings for tactical than technical and physical skills (ps <.05), there was no Role x Skill-type interaction, F < 1.  Player-coach ICCs (whole sample) were moderate (.35-.50), yet all were statistically significant and importantly all PAs were high (88.5 – 99.27%). Within subgroup ICCs were generally stronger, especially for the future professional-group (technical, r = .61, p = .01, 95% CI [.29, .81]; creative, r = .74, p = .01, 95% CI [.49, .88]; tactical, r = .44, p = .04, 95%, CI [.06,.71]; and physical skills, r = 37, p = .12, 95% CI [-.38, .72]) as compared to the academy-     64  only players (technical, r = .36, p = .02, 95% CI [.02, .58]; creative, r = .50, p = .01, 95% CI [.24, .68]; tactical, r = .34, p = .03, 95% CI [.02, .60]; and physical, r = .39, p = .02, 95% CI [.06, .60]). When comparing the ratings of the same players across T1 and T2, ICCs were moderate (.38 - .78), but the PAs were high (90.1 - 98.0%), reflecting change in skills, but some persistency in these initial differences. 3.3.2 Group differences in skill ratings Figure 4 (left side) presents the skill ratings at T1 for the academy-only group and ratings at T1 and T2 for the future professional players. On the right of the figure, professional groups have been divided into youth-pro. only and adult-pro. for T1 and T2 ratings.  A MANOVA comparing the three different groups, based on T1 skill ratings, was significant, F(8, 178) = 2.84, p = .01, Wilks’ λ = .79, ηp2 = .11. Univariate ANOVAs based on the two pre-planned contrasts showed significant differences between 1) academy-only and professional players for tactical (p = .04, d = 0.76), technical (p = .02, d = .50) and physical skills (p = .03, d = .61), but not creative skill (p = .33). Professional players were rated higher than academy-only players. 2) Comparing across professional groups, the only significant difference was for creative skill, (p = .02, d = 1.08). Contrary to expectations the youth-pro. only group was rated higher in creative skill than the adult-pro. group (p = .02). A second MANOVA comparing across the two professional groups for T2 skill ratings yielded a significant group effect, F(4, 21) = 4.98, p = .01, Wilks’ λ = .51, ηp2 = ..49. Separate independent t-tests only showed significant differences between the youth and adult-professional groups for tactical (p = .01, d = 1.45) and a trend for technical skills (p = .06, d = .08).  Exploratory RM ANOVAs were conducted on each of the skills to determine changes over time (T1 - T2) in ratings for the two professional groups (based on Bonferroni adjusted p      65  values). Significant Group x Time interactions were noted for tactical, F(1, 24) = 14.71, p < .001, ηp2 = .38; technical, F(1, 24) = 12.77, p < .01, ηp2 = 32; and creative skill, F(1, 24) = 22.16, p < .01, ηp2 = .48 but not for physical skill, F(1,24) = 2.22, p = .15, ηp2 = .09. As illustrated on the right of Figure 4, tactical, technical and creative skill ratings decreased from T1-T2 for the youth-Pro only group (ps = or < .01). For the adult-pro group, skill ratings increased for tactical and creative skills across the same period (ps = .01). No other time-related effects were significant. 3.3.3 Relations between skill ratings and soccer activities As presented in Table 3, there were no significant correlations between accumulated hours in childhood soccer activities (5-12 yr) and T1 skill ratings. However, within the youth-pro. group there were medium sized correlations between practice and technical and creative skills, (rs >.30). For play, all correlations were low even within a group (rs <.17). For % hrs in play versus practice, the correlations were mostly negative (r = -.22 to -.27, for creative, technical and tactical skills), although none were significant or >.30.  The relationships between hours in childhood soccer activities with T2 ratings of skill for the professional groups are displayed in Table 4. Although for the whole sample, correlations were moderate and not statistically significant, within the youth-professional only group there were significant correlations between practice hours and technical (r =.71, p = .01, 95% CI [.50, .84]) and creative skill (r =.62, p = .01, 95% CI [.36, .79]). For the adult-professional group, there was a positive correlation between childhood practice and physical skill (r = .64, p = .05, 95% CI [-.04, .86]). A surprising negative, though non-significant relation, was shown between practice and technical skill (r= -.54).      66  Consistent with T1 ratings, there were no significant associations between childhood play and skills ratings at T2. The only moderate, positive relation, was for physical skills (r =.30, 95% CI [-.03, .57] whole sample) and a moderate, negative relation for technical skill (r =-.33, 95% CI [-.01, .65], youth-professional only). For the youth-professional only players, accumulating proportionately more hours in soccer play was negatively related to tactical (r = -.55, p = .04, 95% CI [-.82, -.09]) and technical skill ratings (r = -.52, p = .04, 95% CI [-.80, -.05]).  Accumulated hours in career practice (start age to T2) were correlated with T2 ratings of technical (r = .50, p = .01, 95% CI [.12, .67]), tactical (r = .49, p = .01, 95% CI [.19, .71], and creative skills (r = .43, p = .03, 95% CI [.12, .67]), for the whole sample of professional players. Medium to large correlations were evidenced within the youth-pro. only group between career practice and tactical (r = .69, p = .01, 95% CI [.31, .88]), technical (r = .76, p = .01, 95% CI [.44, .91]) and creative (r = .60, p = .03, 95% CI [.25, .81]), skills. There was also a large correlation between career practice and physical skill within the adult-pro. group (r = .75, p = .02, 95% CI [.25, .92]). Career soccer play did not significantly correlate with any measure of skill, although within the adult-pro. Group, there were medium to high positive correlations between play and tactical and physical skills.  3.4 Discussion We report relationships between developmental soccer activities and ratings of technical, tactical, physical and creative skills, as well as the ability of these skills to distinguish success in elite soccer (i.e., attainment of a professional contract at adult and/or youth levels). We expected that skill would discriminate across the two transitions, such that the adult-professional athletes would receive higher ratings than the youth-professional athletes, who would in turn receive higher ratings than the academy-only players (adding to their general validity). We also expected      67  that the skill ratings would be related to childhood practice and potentially play, particularly with respect to creative skill for the latter and with respect to technical and tactical skill for the former. Although T1 ratings discriminated academy-only players from later professional players, showing their general predictive validity, they were not sensitive to individual differences in hours of accumulated soccer practice or play (see also Hodges & Starkes, 1996). T2 skill ratings discriminated the two professional groups, with higher ratings for the adult-professional group in comparison to the youth-professional only group. For the professional athletes, skill ratings were positively related to practice (and negatively related to the proportion of play vs. practice). The higher within-group correlations for the youth-professional players were somewhat unexpected, given that we would expect players to become more alike and harder to distinguish as they progress through elite ranks. However, hours in soccer play did not show expected correlations with ratings of tactical and creative skill (or any skill) at any time point.  Overall, there were few moderate relations between skill ratings provided at T1 and hours accumulated in sport specific play and practice during childhood. In keeping with deliberate practice theory, only sports-specific practice showed any relation to ratings of skills at T1. However, these correlations were small to moderate and mostly within the professional groups (for technical and creative skills). Childhood soccer practice was related to technical and tactical skill ratings provided at T2, particularly for the youth-professional only group. These results suggest that the effects of sport-specific practice on the development (and discriminability) of skills takes time to emerge. Within this group of already “elite” athletes, the relations between practice amounts and evaluations of skills are not strong, probably because of the relative      68  homogeneity in the practice profiles of these players at least in comparison to samples where elite and non-elite groups are compared. Sport-specific play experiences during childhood (5-12 yr) are thought to comprise conditions that facilitate the development of motor, perceptual-cognitive and creative development (e.g., Côté et al., 2012; Ford et al., 2012). We were unable to show support for this proposal at the within group level (c.f., Bowers et al., 2014; Memmert et al 2010). Accumulated hours in soccer practice, particularly career practice estimates, better discriminated across coach skill ratings. These findings align with the assertions made by MacNamara et al. (2015), that motor skill development may be best served via effective practice design, instruction and feedback from expert practitioners. While we do not discount the importance of sport-specific play in developing soccer expertise (e.g., Ford et al., 2009; Ford et al., 2012; Hendry & Hodges, 2018), these data suggest that the refinement of high level skill commensurate with mastery, and a pre-requisite for creativity (e.g., Ericsson and Lehman, 1999), are developed primarily from engagement in sport-specific practice designed and implemented by trained practitioners. Consequently, for performance attainment and skill development these data do not support ideas that childhood deliberate play activities should take precedence over practice (cf., Côté & Hancock, 2014). While there may be additive benefits from participation in self and coach-led practice and play activities from an early age, potentially as a result of the volume of sport-specific activity and the variations in the types of practice which is good for sport-skill learning (see Guadagnoli & Lee, 2004; Handford, Davids, Bennett, & Button, 1997; Hendry & Hodges, 2013; Williams & Hodges, 2005), accumulated play hours or greater amounts of play rather than practice did not significantly relate to any of the skill ratings, except in a negative fashion.       69  Previous soccer-related studies have shown similar relationships between accumulated sport specific practice amounts and assessments of technical and tactical skills (e.g., Huijgen et al., 2009; Kannekens et al., 2009). However, hours in soccer-specific play were shown to explain more of the variance in laboratory-based assessments of tactical skills, including decision making and anticipation, (e.g., Roca et al., 2012, Williams et al, 2012). Although in the study by Roca et al., both sport specific practice and play were significantly related to overall tactical accuracy.  Reflecting the dynamic nature of youth development, rather than enduring individual differences in skills or abilities from an early age, skill ratings significantly changed over time. For future adult-professionals, tactical and creative skill ratings increased, whereas for the youth-professional only group, tactical, technical and creative ratings decreased. Given that the groups were not differentiated at T1, it may be that the period between 13-16 yr is a significant period for skill refinement and discrimination, where evidence of strong(er) perceptual-cognitive skills start to show. These skills may have been less important for younger athletes. Moreover, differences in physical maturity start to even out following peak height velocity at ~13 yr (Neinstein & Kaufman, 2002), potentially allowing for greater discriminability in factors other than speed and strength. The fact that T1 ratings were not good predictors of success at the adult level underlines difficulties associated with attempting to earmark potential future experts using pre-adolescent data and an individual differences approach to talent identification at an early age (e.g., Barreiros et al., 2014). Differences in physical skills were only noted when comparing the academy-only to the youth professional groups, but there were no differences between youth and adult professional groups (either at T1 or T2). Thus, for the transition to youth-professional, physical skills matter      70  for selection (as do tactical and technical skills), such that concerns over selection bias towards more physically capable players in adolescence appear valid (e.g., le Gall, Carling, Williams, & Reilly, 2010). A minimum standard of physical competency is likely necessary to play professional soccer, but in terms of distinguishing players at the highest level, physical skills appear to be of less importance for the transition to adult professional status.  As to whether skill measurements (i.e., an objective test vs. coach ratings) contributed to the discrepancies in findings is unclear. The fact that something can be measured objectively, does not make it a more valid measurement of skill (such as a reaction time test, cognitive recall test), nor necessarily more discriminatory (for critiques, see Ali, 2011; Unnithan et al., 2012). Despite their subjectivity, coach evaluations are key determinants in future decisions about successful progression to professional youth and adult status (e.g., Cushion, Ford, & Williams, 2012). They are by their very nature discriminatory, accounting for the complexity in skills inherent in soccer match-play (e.g., Williams & Reilly, 2000). We are cognizant of the fact that a greater range in coach ratings of players would have aided discriminability, but the restricted range is most likely a function of the elite-nature of the sample. Moreover, only 8 coaches provided ratings of the players and as such no inter-rater reliability measures were collected. However, player ratings of skill showed good agreement with coach ratings (i.e., high percentage agreement), even though players rated themselves more highly than coaches (cf.,  Fogarty & Else, 2005). Moreover, the skill ratings were based on different cohorts of athletes, across different clubs and were provided by different coaches. The fact that these ratings had discriminatory power is therefore significant when considering these sources of variance. We also looked at the PAs and ICCs for the coaches across time. Although we expected change here, the PAs remained high.       71  There are several limitations with the current paper. Retrospective recall methods can be prone to recall error and bias (e.g., Hodges, Huys, & Starkes, 2007). However, in employing a prospective design, we were able to collect data at intervals close to the recorded time-period, rather than requiring adults to reflect on childhood practice amounts. T2 data collection was recorded at a different time of year in comparison to T1 and T3. The reasons for these differences were mostly practical (when we could get access to the players). It may the case that skill ratings vary as a function of the time in the season and the reason they become more reliable predictors of skill at T2 is because the coaches had spent a whole season with the players, rather than a few months. In future, it will be important to validate our current measures of skill with other measures (e.g., the Tactical Skills Inventory for Sport (TACSIS; Elferink-Gemser et al., 2004) and to move away from single item ratings of each skill. For example, with respect to  tactical skill, questions could be asked about its various facets including positioning and decision making on and off the ball.  One final discussion point concerns the operational definitions of the various skills and soccer activities. Although our operational definitions of the 4 skills were based on previous research, they were still relatively simple. While the simplicity was important, due to the age of the players, it may have been that more precise definitions might have contributed to greater discriminability, particularly for the coaches. With respect to definitions of play and practice, sport specific play was distinguished from practice based upon distinctions between unstructured, self-directed activities and structured/coach-led activities respectively. We did not isolate activities based on the reason for engagement even though this was mentioned in our definition of play and practice (i.e., mostly engaged in for fun vs. improvement; cf., Ford et al., 2009). Of course, self-led activities can consist of any activity that could in principle be      72  considered play or practice. Coaches in this study reported that ~50% of practice time involved some sort of semi-structured or unstructured “free-play”. Therefore, to argue for one or another type of practice during childhood may not be fruitful since activities within the various “practice” or “play” definitions are potentially so varied. In future research, it may be useful to investigate the underlying perceptual-cognitive and physical-motor components of each activity type (related to play and practice activities) to better relate the types of activities to the development of specific sports skills. In conclusion, what should be emphasized from our data is that high quality, structured, sport-specific practice, was associated with  the development of skills more so than unstructured, sport-specific play activities, which are not necessarily designed in a way to maximize skill development. While soccer-specific play activities, in concert with structured practice likely aid in the development of future adult experts in soccer (e.g., Ford et al., 2009; Hendry & Hodges, revision submitted), our current data do not show any relations between hours in play and skill evaluations. We also show that ratings of skill are dynamic and change over time, emphasizing the importance of not labelling children too early as being “technically gifted” or “creative”. Although much has been made about the importance of developing creative players, there was no evidence in the current sample, that ratings for this skill were related to future success as determined from transitions from academy to youth or adult professional status.        73  Table 2. Average coach and player ratings and within group SDs at T1 for technical, tactical, physical, and creative skills. Comparisons are made across the Academy-only (Acad) with the future Professional groups (Pro) as well as within this latter group dependent on whether they were youth-Professional only (YPro) or adult-Professional (APro).   ____________________________________________________________________________________________________________  Technical   Tactical   Physical   Creative  Acad Pro YPro APro Acad Pro YPro APro Acad Pro YPro APro  Acad  Pro YPro APro   ____________________________________________________________________________________________________________  Ratings:  Coach Means 3.37 3.96 4.12  3.67 3.13 3.58 3.59 3.56 3.34 3.85 3.76 4.00 3.36   3.70 4.00 3.11  SDs  .77  .77 .78 .71 .89 .86 .87 .71 1.00 .61 .70 .50 .88 .88 .70 .50   Player Means 3.80 4.04 4.10 4.00 3.57 3.88  3.94 3.86 3.71 3.68 3.56 3.43 3.67   3.83 3.81 3.57  SDs  .72 .64 .68 1.0 .60 .58 .57  .69 .87  .85 .96 .78 .94   .76 .63 .98  ____________________________________________________________________________________________________________       74  Table 3. Pearson r correlations between T1 skill ratings for all players assessed at T1 as well as separately for the Academy-only (Acad) group and players selected to play professional-youth at T2 (Pro) and Accumulated (Accum) practice /play during childhood as well as % of overall time in play relative to practice (Play %).  __________________________________________________________________________________________ Technical   Tactical   Physical   Creative All Acad    Pro  All Acad    Pro  All Acad Pro  All Acad Pro   ____________________________________________________________________________________________________________ Correlations: Childhood (5-12 yr).     Accum practice .11 -.04   .33  -.06 -.20   .15  -.07  -.16   .01   .05 -.10   .36     Accum play  .06 -.11   .04  -.12 -.14   -.17  -.06 -.11   .04  -.11 -.16   .01     Play %  .04 -.03   -.23  .20  -.08   -.27  .16 -.03  .04  .10 -.16   -.22   ____________________________________________________________________________________________________________ Correlations >.30 have been italicized (medium effect size, Cohen, 1992).       75  Table 4. Pearson r correlations between T2 skill ratings for all professional players (Pro) and for the subdivided, professional-youth only (YPro) and adult-professional (APro) groups and accumulated (Accum) practice /play during childhood (and Play % as a function of play+practice) and across the player’s career. ____________________________________________________________________________________________________________ Technical   Tactical   Physical   Creative Pro YPro APro  Pro YPro APro  Pro YPro APro  Pro YPro APro  ____________________________________________________________________________________________________________ Correlations: Childhood (5-12 yr).     Accum practice .31  .71** -.56  .23 .37 .01  .29 .25 .64  .39 .62** -.10    Accum play  .16 -.04 .11  .01 -.32 -.18  .30 .04 .23  .06 -.04 -.02    Play (%)  -09 -.52* .40  -.17 -.55* -.11  .09 -.03 -.32  -.22  -.45  -.02 Correlations: Whole career (5 yr -T2).    Accum practice .50* .76** .04  .49* .69** .45  .21 .16 .75*  .43* .60* .10    Accum play  .20 -.12 .38  .17 -.18 .07  .34 -.09 .56  .05 -.20 .23 ____________________________________________________________________________________________________________ * = significant correlation at p<.05 level, ** = significant correlation at p<.01 level. Correlations >.30 have been italicized (medium effect size, Cohen, 1992).       76     Figure 4. Mean skill ratings (tactical, technical, physical and creative) for the academy-only group (T1 only) and all professional (pro.) youth players (T1 & T2, left-side). The professional youth groups are subdivided on the right side into the youth-professional only, and adult-professional groups (T1 & T2).   123456Skill rating (1-5)TacticalTechnicalPhysicalCreative     77  Chapter 4: Factors associated with self-determined motivation in youth soccer players: Comparisons across age and skill in a combined prospective and cross-sectional study  4.1 Introduction A multitude of psychological characteristics potentially influence the pathway towards expertise in sports (e.g., Jordet, 2015). From these characteristics, motivation is considered an essential characteristic of expertise, since high levels of motivation are considered necessary to sustain time and effort in activities aimed at improving performance. The importance of motivation is underlined when considering that these activities may not always be high in enjoyment or immediate reward (e.g., Baker & Young, 2014;  Ericsson, Krampe, & Tesch-Römer, 1993). Of importance to this study, are emerging ideas and evidence that either purport or show relationships between developmental activities (practice and play) and motivation (e.g., Côté , Murphy-Mills, & Abernethy, 2012; Côté, Lidor, & Hackfort, 2009; Hendry et al., 2014; Vink, Raudsepp, & Kais, 2015). In the current study, we track a group of elite youth soccer players to prospectively measure changes in self-determined motivation (SDM, e.g., Deci & Ryan, 2002) over time. In addition to comparing these changes to those based on cross-sectional, age and skill-group comparisons, we evaluate relationships between self-determined motivation and accumulated hours in various developmental soccer activities.  Numerous talent development programs appear to be selecting aspiring experts at increasingly younger ages, with a view to optimizing the volume and quality of practice accumulated during development (Côté, Coakley & Bruner, 2011). Yet, the overall efficacy of this early selection approach and its psycho-social impact on players has been questioned (e.g., Côté & Erickson, 2015). There is evidence that “deliberate play” activities (i.e., unorganized,      78  self-led, sporting activities that are not conducted with a coach/teacher) during childhood can contribute to the emergence of adult expertise and foster positive forms of motivation (Baker, Côté, & Abernethy, 2003; Berry, Abernethy, & Côté, 2008). These findings are encapsulated within the Developmental Model of Sport Participation (DMSP;  Côté et al., 2007; Côté, 1999). The DMSP is a theoretical framework consisting of two primary pathways towards sports expertise; one based on early specialization and deliberate practice in one sport from an early age and a second involving participation in a variety of different sports and play-based sporting activities during childhood and later specialization. According to proponents of this framework, the advantage of the second pathway is that expertise can be attained without the negative consequences associated with early specialization, such as increased incidence of burnout, drop-out, injury and a general decline in well-being (e.g.,  Côté et al., 2007;  Côté, 2009). The largely volitional and enjoyable nature of deliberate play in childhood is thought to develop intrinsic and self-determined forms of motivation in adulthood sport participation (e.g., Côté et al., 2007, 2012; Côté, 2009).  The early specialization pathway dictates single sport involvement from an early age, as a means of maximizing the accumulation of hours in highly specific practice activities so that expertise can be achieved at a relatively young age. This pathway towards expertise is based on ideas emanating from the theory of deliberate practice (Ericsson et al., 1993), which reports a monotonic relationship between deliberate practice activities, engaged with the sole intent of improvement, and performance. Deliberate practice theory has been criticized, with recent meta analyses indicating that hours in deliberate practice accounted for less of the variance across and within skill levels than previously thought (e.g., Hambrick et al., 2014; Macnamara, Hambrick, & Oswald, 2014). Despite this criticism, a considerable body of evidence in sports supports the      79  idea that skill and deliberate (or purposeful) practice are positively related and hence high volumes of deliberate practice are needed to succeed (see Baker & Young, 2014; Ford et al., 2015). As learners must invest maximal cognitive and physical effort over an extended period of time in deliberate practice, motivation is central to this theory (Ericsson & Towne, 2010). Different types of motivation are required to engage in deliberate practice activities since they are often described as not always being inherently enjoyable (e.g., Ericsson et al.,  1993; Coughlan, Williams, McRobert, & Ford, 2014). Further, the reasons for engaging in deliberate practice may change from engaging in practice for enjoyment in practice itself (e.g., intrinsic motivation), to enjoyment from the rewards of practice (e.g,. improved performance, Ward et al., 2003).  The complex nature of motivation involved in  practice engagement is encompassed within Self-Determination Theory (SDT; Deci & Ryan, 1985; Ryan & Deci, 2017). SDT is a meta-theoretical framework which offers a nuanced, multidimensional account of motivation. At the forefront of this theory is the idea that humans have an innate tendency to seek growth and embrace challenges which results in engagement in an activity for interest and enjoyment (i.e.,  intrinsic motivation, Vansteenkiste, Lens, & Deci, 2006). According to Côté and colleagues, the largely volitional and enjoyable nature of deliberate play in childhood is thought to develop intrinsic and self-determined forms of motivation in adulthood sport participation (e.g., Côté et al., 2007, 2012; Côté, 2009). This stands in contrast to deliberate practice, which is often externally controlled and may not necessarily be intrinsically rewarding (Ericsson, et al., 1993).  Central to SDT is Organismic Integration Theory (OIT; Deci & Ryan, 1985, 2000; Ryan & Deci, 2017). The OIT places motivation along a continuum of self-determination, in which initial engagement in an activity for contingent (or externally rewarding) reasons can become      80  internalized over time so that the behavior becomes progressively integrated into one’s sense of self (i.e., more self-determined). There are three broad types of motivation, namely, intrinsic, extrinsic, and amotivation, which are underpinned by six behavioral regulations. As the only intrinsic motivation, Intrinsic regulation occurs when an individual performs for enjoyment or interest. Next on the continuum is extrinsic motivation, consisting of four behavioral regulations. As the most self-determined motivation, integrated regulation reflects a full assimilation of the values and beliefs from the activity into a sense of self. The individual participates in sport because they identify themselves as an athlete and live their life in accordance with becoming a better athlete (Taylor, 2015). Identified regulation signifies sport engagement because the benefits of sport involvement are highly valued. Participating in sport to avoid feelings of shame or guilt associated with non-participation is referred to as introjected regulation. These feelings may occur when an athlete participates to appease family members or feelings of contingent self-worth. External regulation, which signifies sport involvement to seek rewards (e.g., trophies or medals) or avoid punishment (scolding from parents/ coaches) is the least self-determined extrinsic motivation. Amotivation denotes a complete lack of motivation. Behavioral regulations can be encompassed within two higher order themes; autonomous (including intrinsic, integrated and identified regulations) and controlled motivation (including introjected and external regulations. Generally speaking, autonomous forms are associated with positive outcomes, whereas controlled motivation are largely related to negative outcome (Ryan & Deci, 2017).  Organismic Integration Theory (OIT) offers some means of understanding the complexity of motives for athletes and may aid our understanding of the relationships between early sport activities in developing SDM.  According to this theory, changes in SDM are moderated by several factors such as external rewards, age and skill. For example, in a meta-analytic review of      81  SDM in educational contexts, the use of external rewards was shown to undermine autonomous motivation (Deci, Koestner, & Ryan, 2001). Although external rewards typify the attainment of professional status in many sports, in particular men’s soccer, changes in SDM over time, as the lure of professional rewards become more salient, has not to date been investigated in longitudinal-type investigations. Age-related declines in SDM have been shown in non-elite, physical education settings during early adolescence, perhaps related to competing interests at this age (12-14 yr; e.g., Barkoukis, Taylor, Chanal, & Ntoumanis, 2014; Otis, Grouzet, & Pelletier, 2005). However, higher performing students did not show this decline. Rather, a positive association was seen between students’ performance and autonomous (or self-determined) motivation (Barkoukis et al., 2014). Based on these factors, there is reason to suspect that SDM would change over time, potentially as a function of age and/or skill, becoming less autonomous with age (around adolescence) and then later more autonomous as skill is achieved. In high-level, youth sports, where the lure of external rewards increase with age, there is also reason to suspect that motivations would become less rather than more autonomous.  According to Taylor (2015), controlled forms of motivation related to performance improvement, achieving status positions and winning competitions, become increasingly important through the transitions towards adult expertise. Aspects of controlled motivation, such as introjected regulation, appear to facilitate perseverance and resilience, which are needed when practice or competition become demanding and/or monotonous (Gillet, Berjot, Vallerand, Amoura, & Rosnet, 2012; Gillet, Berjot, & Gobancé, 2009; Hardy et al., 2016).  In a cross-sectional comparison of elite youth soccer players across several age groups, the oldest group of soccer players (i.e., under 17 yr age group, U17) was shown to be less      82  autonomously motivated than the younger age-group (U13 and U15; Hendry et al., 2014). Also, older players (U17) had lower behavioral regulation scores for integrated and identified regulations, suggesting a diminished value of soccer and a reduced assimilation between soccer and their sense of self. For the older, U17 players only, accumulated hours in Academy practice were negatively related to global measures of SDM and positively related to controlled motivation. There were no associations between accumulated hours in childhood, play-type activities and motivation for any of the age-groups (cf., Côté, 2009). Since this previous study was based on cross-sectional comparisons across different age-groups and there were no skill-group comparisons, questions remained about the reasons for these differences. Specifically, whether these differences would be replicated across time in the same participants as they transition from U15-U17 yr (ruling out a potential cohort effect) and whether these age related effects were specific to elite-youth athletes, rather than sport-matched, non-elite peers. In the current study, we followed up elite-soccer players who had progressed from U13 to U15 (yr) and from U15 to U17 (yr), soccer-Academy age groups. We compared the current U15 and U17 elite-age groups with age-matched, non-elite soccer groups to assess whether any age-related differences in SDM were indicative of general developmental trends in sports, unrelated to the elite-Academy setting. We expected to see a general reduction in autonomous motivation with age (from U15 to U17 yr, but not from U13-U15) and an increase in controlled motivation, yet we were unsure the extent to which these declines would covary with skill. Although there was reason to suspect declines in measures of SDM in adolescence (e.g., Barkoukis et al., 2014; Otis et al., 2005), the nature of external rewards associated with professional contracts as the elite-youth players progress from U15-U17 years might also lead to the prediction that age group differences will be specific to elite groups.       83  A second reason why age-group differences or declines in measures of SDM might be observed in older groups of youth-elite soccer players is related to the quantity and demands of practice. Therefore, we also evaluated whether engagement in recent soccer practice and play activities was related to current measures of motivation and any changes in motivation over this time period (see, Côté et al., 2012). We expected that more time spent in formal practice across the intervening years would be negatively related to autonomous, and positively related to controlled motivation. Based on earlier research (Hendry et al., 2014), we did not expect relations between play and motivations, at least for the elite sample. For the non-elite group, childhood play may be an important variable in promoting longer term self-determined motivation, because the relative amounts of play vs. practice are expected to be larger and other factors related to extrinsic rewards are less likely to moderate any potential relationships.  4.2 Methods 4.2.1 Participants We collected data from 62 male, youth soccer player (n= 31, elite players from five professional youth Academies in Scotland; n = 32 non-elite players from Western Canada). The elite players completed practice and motivation questionnaires at T1 (Oct. 2011; see Hendry et al., 2014) and T2 (Jan. 2014). Elite players, participating in the highest tier of Scottish youth soccer, had transitioned through their respective professional soccer academies from U13 (12-13 yr) & U15 (14-15 yr) at T1 to U15 (n = 15) & U17 (n = 17; 16-17 yr) in the longitudinal follow-up (T2). Data from the non-elite group were collected from U15 (n = 16) and U17 (n = 16) soccer age-groups playing in competitive, yet recreational, youth leagues in Western Canada (Dec. 2015). Both the elite and non-elite groups, whilst different to each other, had accumulated      84  a similar number of soccer activity hours (including match play) as detailed in previous studies of soccer players participating in the UK (~ 3000-5500 hr; e.g., Ford & Williams, 2012; see also Table 5). T1 motivation scores from elite players only, reported previously (Hendry et al., 2014) were included within the current study as a means of assessing change in motivation from T1 to T2 within the same sample of players. The ~2.5 year gap between data collection points corresponded to age-related differences based on cross-sectional comparisons observed in previous work. All parents provided passive consent for their sons to complete the questionnaires before players gave written informed consent. All procedures were approved by the lead institution’s local ethics’ committee. 4.2.2 Procedures Initial recruitment was made via email correspondence with participating clubs. At T1, elite players completed questionnaire 1 (Q1) which included a soccer-specific, practice history questionnaire and the Behavioral Regulation in Sport Questionnaire (BRSQ, Lonsdale, Hodge, & Rose, 2008). The data was collected in small groups supervised by the first author, such that clarification and assistance could be provided when needed. At T2, elite players completed questionnaire 2 (Q2), which included a truncated version of the soccer activity questionnaire focusing on the developmental activities engaged in between T1 and T2 (~ 2.5 yr period), as well as the full BRSQ. To aid convergent validity, a sample of parents (T1, n = 6; T2, n = 4) provided career estimates of soccer practice and play using the same questionnaire. Also, coaches (T1, n = 6; T2, n = 4) provided estimates of the number and content of a typical weeks’ organized practice session for their respective age groups (see Hopwood, 2015 for recommendations regarding these validation methods).      85  Non-elite players completed Q1 only and followed the same procedures as the elite group at T1. Again, players’ coaches (n = 5) provided estimates of the number and content of a typical weeks’ organized practice session and a sample of parents (n = 4) provided career estimates of hours in soccer activities. Participating clubs were contacted via email at T1 and T2 and follow-up emails and meetings were made with the individual team managers or coaches.  4.2.3 Measures  4.2.3.1 Retrospective questionnaires The soccer-specific practice questionnaire was adapted from what has been referred to as the “Participation History Questionnaire” (PHQ) and previous research related to testing of deliberate practice theory (initially based on methods used by Ericsson et al., 1993). This questionnaire and similar versions have received validation with respect to their ability to provide estimates that differentiate across elite and less elite samples, matching of estimates across current weekly practice amounts, diary estimates and estimated yearly amounts, matching of estimates across coach, parent and athlete samples as well as validation from triangulation of retrospective methods with age-related, cross-sectional samples (e.g., Ford et al, 2007, 2010; Helsen et al., 1998; Hodges & Starkes, 1996; Hodges et al., 2004; Ward et al., 2007). This retrospective method remains the best available method for collecting practice histories from elite athletes (Hopwood, 2015).  Basic demographic information pertaining to start age in soccer activities, typical current weekly practice amounts in soccer (for reliability purposes), total number of other sports engaged in outside of school, and the number of years in the Academy system were collected in Q1 and Q2. Operational definitions and examples of organized practice and play were provided. Practice was defined as activities conducted with a coach/adult used mainly to improve skills (i.e., formal      86  practice). Play was defined as unorganized, self-led activities that are not conducted with a coach/teacher (i.e., informal, self-led soccer activities). Players provided estimates of: i) number of organized practice sessions/week; ii) average duration of each session; and iii) hours/week in soccer play, during a typical week. These data were solicited from 5 years of age to the present time in 2-year intervals (i.e. 5-6 yr, 7-8 yr …15-16 yr).Significant breaks from soccer were recorded. Linear interpolation methods were used to estimate accumulated practice/play hours during intervening years, which involved estimating practice from the missing years (e.g., 6-7 yrs), based on the average of hours reported from 5-6 yr and 7-8 yrs (e.g., Ward et al., 2007).  Accumulated hours in practice were calculated by multiplying hours per session by the number of sessions/week. This number was multiplied by 46 weeks practice/year for elite players and 36 weeks for non-elite players (these were the average reported season lengths for participating players), subtracting weeks lost through illness or injury for individual players. This procedure was repeated for soccer play. We calculated accumulated hours in soccer practice and play during childhood (5-12 yr) and across careers (5–current yr). Questionnaire 2 (Q2) was a truncated version of Q1. Although it consisted of the same demographic and developmental soccer activity questions as in Q1, it differed in that data were collected every year for the 2.5 year period spanning T1 to T2.  In order to assess reliability and validity, intra-class correlations (ICCs) and percent agreement (PA, based on division of the smaller by the largest value for each pair, multiplied by 100) were calculated for; i) the player-player estimates within the same questionnaire, ii) player-player weekly estimates from Q1 (last yr) and Q2 (first yr), iii) coach-player weekly estimates of soccer practice and iv) parent-player estimates of accumulated hours spent in developmental soccer activities (i.e., both practice and play). These give an indication of the strength of the      87  relations and similarity between estimates respectively and this combined analyses has been recommended as the most comprehensive for assessment of validity and reliability of activity estimates (Atkinson & Nevill, 1998b; Hopwood, 2015).  At T1 (elite group only), the strength and similarity of player-player estimates of time spent in weekly soccer activities (from different sections of the questionnaire) were deemed moderate to high and increased for more recent estimates (n = 31; PA range = 68.10% - 83.43%, ICC range = 0.46 - .91, ps <.05). In comparing the time-period during which estimates from Q1 and Q2 overlapped, the strength and similarity was again high for estimates of play (PA = 83.5 %, ICC = .87) and practice (PA = 93.1%, ICC = .91). Also, there was a high correlation (ICC = 0.92) and degree of similarity (PA = 91.3%) between coach and player estimates of weekly practice hours. Parent-player estimates (based on accumulated hours) were also moderately correlated for both practice (PA = 59%, ICC = .58) and play (PA = 56%, ICC = .60). Similar reliability was established at T2 for the elite players, although no within questionnaire, player-player estimates were obtained. There was a high correlation (ICC = 0.94) and similarity (PA = 92.70%) between player and coach weekly practice estimates and between player-parent estimates for both practice (PA = 80.1%, ICC = 0.82) and play (PA = 75.6%, ICC = .73).  For the non-elite players, player and coach estimates of practice fell within the high range (PA = 82.0 %, ICC = .84), as did player and parent estimates of accumulated hours in play (PA = 70%, ICC = 0.76) and practice (PA = 85%, ICC = 0.90).  4.2.3.2 Motivation The 24 item, BRSQ uses four item subscales to measure each of the 6 behavioral regulations from SDT and provides overall indices of motivation (see Table 6). Participants respond to the following stem; “I participate in soccer because…” before responding to each      88  item using a 7-point Likert scale where 1 = Not at all true, 4 = somewhat true and 7 = very true. The items for each subscale were aggregated to provide an overall (average) score for each behavioral regulation. Global indices of SDM (SDI) and autonomous and controlled motivation were calculated by applying a coefficient to the behavioral regulations, see Table 6 (Hodge & Lonsdale, 2011). Scale reliability of each behavioral regulation score was determined using Cronbach’s α = .70, which given the low number of items used to measure each subscale was deemed acceptable (Cortina, 1993). Motivation change scores were calculated for the elite players that had completed the BRSQ at T1 & T2. To ameliorate potential for Type 1 error, we focus primarily on composite scores of SDI (overall self-determined motivation score) and autonomous and controlled motivation, given that these measures were most related to our predicted effects due to age and or skill development.  4.2.4 Statistical analyses The data were checked for normality using the Shapiro-Wilk test. When the magnitude of skewness was less than 1, indicating only a tendency towards positive skewness (Bulmer, 1979), and there were no significant differences in homogeneity of variance between the groups, we used parametric methods for our analyses based upon the robustness of this technique to violations in normality (Glass et al., 1972; Pallant, 2007). In cases where assumptions were not met, which was the case for accumulated soccer activity estimates, non-parametric tests were used to assess relationships (i.e., Spearman’s correlation coefficient). Confidence intervals (95%) around mean differences for significant pairwise comparisons and for Pearson’s correlations are provided.       89  4.2.4.1 Soccer development and demographics Independent t-tests were used to evaluate differences between the elite and non-elite players with respect to various soccer-related demographics including: start age in soccer; start age in soccer practice; number of other sports; and hours per week and accumulated hours in play and practice.  4.2.4.2 Motivation comparison across age and skill As part of the prospective assessment of motivation for the elite group, we ran a 2 (Current Age category; U15yr, U17yr) x 2 (Time; T1, T2) repeated measures ANOVA for the primary dependent variables, SDI, autonomous and controlled motivation. To determine whether any potential age-related differences in motivation were specific to the elite group, we conducted separate 2 (Skill level; Elite, Non-elite) x 2 (Current Age category; U15, U17) between-subjects’ ANOVAs for the same indices of motivation as noted above and used Tukey HSD post hoc procedures to evaluate any potential interactions.  4.2.4.3 Soccer activity relationships with motivation Spearman correlations were conducted to assess the relationships between T2 indices of motivation and accumulated hours in soccer activities during childhood (5-12 yr, for the elite and non-elite players) as well as in the more “recent” 2.5 years (T1-T2; elites only). In order to potentially explain any change in motivation across time, we also analyzed the relationship between change in indices of motivation (from T1 to T2; elites only) and recent practice over this same time period. Correlations where rs >.30, were considered to reflect a moderate effect size (Cohen, 1988).  4.3 Results 4.3.1 Soccer development and demographics       90  Table 5 shows the mean, soccer-related practice data and inferential statistics comparing the elite and non-elite groups. The elite players engaged in more soccer practice and play/week, accumulated more hours in soccer practice and soccer play and engaged in general soccer activities earlier and participated in fewer sports when compared with the non-elite group (p’s < .05). The groups did not differ with respect to when they first participated in soccer practice.   4.3.2 Motivation comparisons across age, time and skill Indices of motivation and data for all the behavioral regulations for T1 and T2 are shown in Table 6. For the elite groups across time, the current U15 group showed little change from T1 (U13) to T2 for autonomous motivation, whereas controlled motivation decreased. However, from T1 (U15) to T2 for the current, elite, U17 group, autonomous motivation showed a small decrease, whereas controlled motivation increased. These data are additionally illustrated in Figure 5. These descriptive comparisons were confirmed statistically by Age X Time interactions for SDI, F (1, 29) = 7.85, p = .01, ηp2 = .21 and controlled motivation, F (1, 29) = 5.79, p = .02, ηp2 = .21 (none of the main effects were significant). Post hoc analyses showed that for SDI, the U17s had significantly lower SDI scores than the U15s at T2 only (p < .01, Mdifference = 5.38, 95% CI [1.38, 9.37]). There was also a significant decline across time for the current U17s (p <.05, Mdifference = 4.59, 95% CI [.06, 9.11]) but the increase in SDI for the U15 group (from U13 yr), was not significant. Post hoc analysis of controlled motivation also showed significant age group differences at T2, with the U17s scoring higher than the U15s (p < .05, Mdifference = 4.23, 95% CI [0.72, 7.74]) but the increase in controlled motivation scores over time for U17s was not significant. For autonomous motivation, the younger players (U15) scored higher than the older players (U17), F (1, 29) = 10.00, p = .02, ηp2 = .26, Mdifference = 1.75, 95% CI [0.68, 2.82] and      91  there was a tendency for a reduction in autonomous motivation with time, F (1, 29) = 4.05, p = .05, ηp2 = 13, Mdifference = 0.70, 95% CI [0.50, 0.88], but there was no interaction. When comparing the non-elite players to the elite players, for motivation indices, the elite groups generally scored higher than the non-elite groups (see Table 6). Separate 2 (Skill; Elite, Non-elite) x 2 (Age category; U15, U17) between groups ANOVAs supported this skill main effect for SDI, F (1, 61) = 13.81, p < .001, ηp2 = .19, Mdifference = 4.04, 95% CI [3.65, 4.43] and autonomous, extrinsic motivation, F (1, 61) = 19.88, p < .001, ηp2 = .25, Mdifference = 2.59, 95% CI [2.39, 2.79]. For SDI, a Skill X Age interaction was significant at p =.05 level, F (1, 61) = 4.29, ηp2 = .06. For elite players, the U15 group scored significantly higher than the U17 group (p < .05, Mdifference = 3.90, 95% CI [3.09, 4.71]), and it also scored higher in comparison to the non-elite, U15 group (p < .01, Mdifference = 5.34, 95% CI [2.66, 8.02]) and U17 group (p < .01, Mdifference = 5.53, 95% CI [2.59, 8.64]). The U17 elite players were not different to the non-elite U17 players. There were no age main effects for any of the indices. For controlled motivation, the Skill X Age group interaction approached significance, F (1, 61) = 3.65, p = .07, ηp2 = .06. Inspection of the means showed that when comparing across skill, the U15 elite players had lower scores than the U15 non-elite (p < .05, Mdifference = 2.51, 95% CI [.03, 5.34]). This was not the case for the U17 players, where scores were higher for the elite group, but there were no significant differences.   4.3.3 Soccer activity relationships with motivation For the elite players, neither childhood soccer practice nor play were significantly correlated with T2 indices of motivation (rss <.30). Hours in organized soccer practice in the more recent 2.5 years were, however, negatively correlated with SDI (rs = -.59, p = .005, 95% CI [-.77, -.30]) and autonomous motivation (rs = -.52, p = .009, 95% CI [-.74, -.21]). Controlled      92  motivation was moderately, positively correlated with recent soccer practice (rs = .36, p = .04, 95% CI [-.63, -.01]). Practice hours were not significantly related to motivation change scores (from T1-T2) for any of the indices (rs < .30). Recent hours spent in soccer play did not correlate with any of the composite measures of motivation, either for the whole sample, or for the two age groups separately.  For the non-elite players, there was a moderate, negative correlation between childhood practice and autonomous motivation (rs = -.35, p = .04, 95% CI [-.62, -.01]). This relationship was also observed for “recent” practice (rs = -.48, p = .03, 95% CI [-.71, -.16]). For SDI, there was also a negative, moderate relation with recent practice (rs = -.40, p = .05, 95% CI [-.66, -.06]). As with the elite players, childhood play did not correlate with SDI, autonomous or controlled motivation in the non-elite group.  4.4 Discussion Based on a prospective follow-up of elite youth soccer players and cross-sectional comparisons with age-matched, non-elite players, we tested whether measures of self-determined motivation differed as a function of age and the player’s skill and whether they were related to early practice and play experiences. Declines in SDM over time within the older (current U17) elite players were consistent with previous cross-sectional work (Hendry et al., 2014). Within the present study, older players exhibited a less self-determined profile at T2, including lower SDI, lower autonomous and higher controlled motivation scores, than younger elite players. These findings suggest that differences in SDM across age groups were not cohort specific and are indicative of a more consistent trend within elite youth soccer. The inclusion of age matched (U15, U17), non-elite soccer players provided opportunity to assess whether age related differences (or changes) in motivation were specific to these elite athletes. Elite players scored higher for SDI and      93  autonomous motivation than the non-elites. A Skill X Age interaction for SDI showed that the younger elite (U15) participants scored significantly higher than their U17 elite counterparts and higher than both non-elite age groups, but no differences were seen across age for the non-elites. Thus, although we have data consistent with age-related differences and declines in SDM in elite athletes they were not observed for non-elite athletes. Therefore, rather than age alone being a reason for change in SDM over time, especially during adolescence as detailed in studies conducted in physical education settings (Barkoukis et al., 2014; Otis et al., 2005), differences in SDM are related to both age and skill (in elite/professional pathways in soccer). These data lead us to suspect that elite sport in general encourages or requires more SDM, which drops off around 16 years of age (U17), to levels commensurate with non-elite athletes.  The higher controlled motivation scores in the older elite players might be due to several factors. First, the proximity to the external rewards associated with professionalism (e.g., money, status) may have contributed to an increase in controlled motivation. This is consistent with meta-analytic data from education showing a shift towards more controlled forms of motivation once external rewards are introduced to previously self-determined and intrinsically rewarding activities (Deci et al., 2001). Second, the time demands placed upon elite youth athletes are vast and require an element of sacrifice from engaging in non-soccer related activities (e.g., Cook, Crust, Littlewood, Nesti, & Allen-Collinson, 2014). Not only may this result in a sense of conflict from trying to balance sport and other activities, but this may also result in a diminished sense of autonomy over their overall training schedule, which again can undermine soccer-related SDM (Pelletier, Fortier, Vallerand, & Brière, 2001). Although not measured within the present study, the overarching impact of the social environment within the UK Academy setting requires further consideration (Deci & Ryan, 2002). Published reports have described a tendency      94  for the motivational-climate to become more controlling with age (Partington et al., 2014), potentially impacting basic psychological needs of autonomy (Ryan & Deci 2017).   Despite the aforementioned shifts in motivation occurring over time, practically speaking, the change scores in motivation were small, suggesting that the nature of the motivation remained relatively stable over this 2.5 yr period (see Table 6). For the elite group, indices of autonomous motivation remained consistently high, while controlled motivation, despite increasing over time, remained relatively low (see also Zuber, Zibung, & Conzelmann, 2014). While the elite players exhibited a largely self-determined profile, the gradual shift towards less self-determined and more controlled motivation within the older elite players hints at the emergence of co-existing forms of motivation. High scores for both autonomous and controlled motivation characterised elite fencers and runners who, despite outperforming their less elite peers, reported being more physically and emotionally exhausted (Gillet et al., 2009, 2012). Related, introjected regulation, which is a key constituent of controlled motivation, is linked to positive outcomes such as persistence, interest, and satisfaction (Briere, Vallerand, Blais, & Pelletier, 1995; Pelletier et al., 2001). These outcomes are viewed as being essential to cope with the rigours of elite sport (see Jordet, 2015). An absence of a purely self-determined motivational profile is also consistent with qualitative research conducted with super elite athletes (multiple gold winners at Olympic and World Championships; Hardy et al., 2017) and coach reports of former youth players that had gone on to play elite, adult soccer (e.g., Cook et al.,  2014). It appears that older elite players are motivated for an innate desire for self-improvement as well as a contingent sense of self-worth attached to outperforming others (e.g., team-mates, opposition).  A secondary aim of this study was to test Côté and colleagues postulate that engaging in childhood play would foster later intrinsic and self-determined motivation (Côté et al., 2012). We      95  evaluated this postulate within an elite and a competitive, yet non-elite sample. Overall, these data did not support this postulate. There were no statistically significant (or moderately sized) relationships between indices of motivation and estimates of childhood soccer play across both samples. However, within the non-elite group, accumulated childhood practice hours were negatively related to autonomous motivation. This finding is partially in line with Côté and colleagues assertion that early practice activities may have negative psycho-social outcomes. This result is somewhat attenuated by the fact that non-elite players amassed less than half the total of childhood practice hours compared to elite players. Therefore, it is not simply the amount of soccer practice that is a concern for motivation, but perhaps it is the amount of practice invested as a function of success, or relative amounts of soccer practice (compared to other sports or play). Recent practice amounts (practice over the last 2 years) were positively related to controlled motivation within the elite group and negatively associated with autonomous motivation. However, change scores in motivation were not significantly associated with recent practice amounts. This suggests that factors other than practice and play amounts were responsible for SDI change across the age groups, such as the proximity to rewards associated with professional status.  Despite strengths associated with the overall research design, there are of course some methodological challenges. Retrospective recall techniques, although prone to bias (Hopwood, 2015) still remain the best method of ascertaining estimates of practice histories (Hodges, Huys, & Starkes, 2007). Because participants in the current study were still children when estimates were collected, and thus their recall would be less “retrospective” than data based on adult samples, we anticipate less of a validity issue with this method. Further, a small sample of      96  parents and coaches provided practice estimates which provided convergent validity for child estimates of soccer activity hours and the within and between questionnaire estimates for the elite players were strong and similar. We do acknowledge that the samples were relatively small (n =32/skill group), creating issues for statistical power and generalization. Yet, the high level of our elite sample, allied to the prospective nature of the study, and the natural attrition associated with elite soccer transitions, adds validity to our choice of sample and subsequent conclusions. Limits are also associated with the non-elite group given that these soccer players were from Canada, yet the elite players were from the UK. However, the non-elite players were playing a relatively high level of competitive soccer in Canada and had participated in practice volumes similar to those noted in studies of UK-based recreational, yet competitive soccer players (e.g., Ford & Williams, 2012).  In conclusion, we have provided evidence that motivations in youth, elite soccer are dynamic and dependent on age and skill. Shifts along the OIT continuum towards less self-determined and more controlled motivation with time (and age) in elite players is likely related to the increasing competitive demands of elite youth soccer and proximity to external rewards associated with professional status (e.g.,  Deci et al., 2001). It does not appear to be related to an increase in hours spent in soccer activities. However, regardless of age, elite youth players were generally more autonomously motivated than the non-elite athletes. Although it is possible that childhood play activities promote enjoyment (all players participated in high volumes of childhood soccer play), there was no evidence that this early enjoyment persists in its influence with respect to enhanced SDM. The prospective and cross-sectional study design allowed us to determine changes within the same population as well as within different cohorts (age and/or skill-matched). Despite      97  attrition associated with prospective follow-up, and reliability issues associated with retrospective recall techniques, this design provided the opportunity to validate previous age-related data based on cross-sectional comparisons, which speaks to changing motivations in elite youth soccer athletes with age (from U15 to U17 levels). Across two different cohorts, age-group differences in SDM were observed in the elite players, as well as declines in SDM as the same groups progressed from U15 to U17. These differences appeared to be influenced, at least in part, by the psychosocial environment surrounding involvement in an elite practice environment. We suspect that the findings would generalize to other competitive situations where the necessity of high volumes of practice are required and external rewards such as government funding and professionalization are introduced to an extent that they are fundamental towards achieving elite level, adult sport status.         98  Table 5. Means, SDs and 95% confidence intervals corresponding to accumulated and weekly hours in practice and play (during childhood and across the player’s careers) for the elite and non-elite groups, as well as start age in soccer activities and number of sports participated in childhood. Statistical analyses are also presented based on independent t-tests (df = 61). Cohen’s d is given as a measure of effect size. Soccer activity and age  Elite   Non-elite t  Cohen’s d 95% CI (mean differences)               Lower  Upper __________________________________________________________________________________________________________ Childhood (5-12 yr; hrs):  Accumulated soccer practice  1834 (824)  886 (367) 6.35**  1.55  629.25  1266.75 Accumulated soccer play  2259 (1156)  888 (608) 4.21**  1.04  909.45  1832.55   Career (5 yr – current yr; hrs): Accumulated soccer practice  2741 (1083)  1403 (466) 3.53*  0.86  955.34  1720.73  Accumulated soccer play  2724 (887)  1224 (814) 9.81**  2.42  895.58  1746.42 Current weekly soccer practice 8.29 (2.34)  3.07 (.49) 16.16** 3.69      4.37        6.06 Current weekly soccer play  3.91 (2.50)  2.14 (1.95) 3.24**  0.79       .06         2.89  Recent soccer activities (last 2.5 yr; hrs): Accumulated soccer practice  907 (212.62) Accumulated soccer play  465 (324.30)  Soccer Milestones: Start age soccer (yr)   4.55 (1.21)  5.24 (1.26) 2.46*  0.56       .07        1.31 Start age soccer practice (yr)  5.80 (1.98)  6.44 (1.81) 1.49    .03     -.03        1.51 Number of other sports  2.61 (1.35)  4.44 (1.21) 6.29**  1.43    1.18        2.47 ___________________________________________________________________________________________________________ *p<.01, **.001     99  Table 6. Mean (and SD) self-determined motivation scores of the current U15 & U17 elite and non-elite soccer players at time 1 (T1) and time 2 (T2).            Elite                      Non-elite             U15       U17    U15  U17      T1 (U13) T2 (U15) T1 (U15) T2 (U17)  T2  T2 Motivation indices SDI (Max = 25) (2 x IM + 1 x IG + 1x ID +  16.00 (4.10) 18.41 (3.08) 16.42 (5.96) 13.03 (5.19)  13.08 (4.25) 12.89 (4.85) (-1) x IJ + (-2) x EX)  Autonomous EM (Max = 28)  26.88 (1.00) 26.63 (1.34) 26.03 (1.92) 24.88 (1.60)  23.52 (2.94) 23.98 (2.83) (2x IM + 1 x IG + 1 x ID) Controlled EM (Max = 21)  10.88 (4.28) 7.63 (4.29) 9.55 (6.38) 11.86 (7.37)  10.44 (2.51) 10.19 (3.41) (-1x IJ + (-2) x EX)   Behavioral Regulations (Max = 7) Intrinsic (IM)    6.98 (.75) 6.94 (.14) 6.89 (.30) 6.67 (.40)  6.59 (.46) 6.54 (.63) Integrated (IG)   6.76 (.52) 6.62 (.48) 6.35 (.75) 6.07 (.68)  5.36 (1.12) 4.94 (1.07) Identified (ID)    6.15 (.82) 6.14 (.94) 5.89 (.98) 5.54 (.93)  4.97 (1.26) 5.06 (1.00) Introjected (IJ)    3.74 (1.83) 2.55 (1.35) 2.90 (1.66) 3.75 (2.32)  3.17 (1.08) 3.01 (1.15) External (EX)    1.70 (.78) 1.32 (.51) 1.86 (1.65) 2.18 (1.51)  2.05 (.66) 2.08 (.87) Amotivation    1.38 (.38) 1.04 (.17) 1.36 (.56) 1.86 (1.52)  1.63 (.87) 1.36 (.56) ____________________________________________________________________________________________________________ SDI = Self Determination Index; IM = Intrinsic motivation; EM = extrinsic motivation     100     Figure 5. Group means (and SD bars) for global self-determined motivation (SDI) and controlled extrinsic motivation (EM) as a function of time (time 1, T1 or time 2, T2) and current (T2) age group (U15 & U17 yr) for the Elite players.0510152025T1 T2 T1 T2Motivational index scoreSDI Controlled EM U17 U15     101  Chapter 5: Influence of developmental activities and perceptions of challenge in the development of expertise in women’s soccer.  5.1 Introduction  Women’s soccer is a rapidly growing in popularity with approximately 30 million participants worldwide, making it one of the most popular sports for females in the world (FIFA, 2015). Paradoxically, however, there remains a paucity of research focusing on the developmental experiences of elite female soccer players which could be used as a benchmark to guide future generations of elite players (Gledhill & Harwood, 2014). Our knowledge to date is largely derived from research on elite and sub-elite male players (e.g., Ford, Ward, Hodges, & Williams, 2009; Ford & Williams, 2012; Hendry, Crocker & Hodges, 2014), yet it is unclear whether these findings are generalizable to female players (Gill, 2001). In this paper, we present data on the career practice histories of elite (National) and sub-elite (Varsity) women soccer players in Canada, with respect to soccer play, practice, and competition, as well as involvement in other sports. Our aim is to assess the extent to which extant models of sport development, developed based on data mostly from male players, capture the experiences of world class and highly-skilled female soccer players. In addition, we assess the discriminability of measures of practice quality, namely challenge perceptions, across two groups of highly skilled, female soccer players.   While studies of female soccer experts are scarce, there has been qualitative research directed to uncovering some of the key developmental markers of elite (National), youth, female soccer players in England (Gledhill & Harwood, 2014). One developmental activity viewed as beneficial to later success was early competition experience with male players (i.e., 10- 12 yr). Competitive, co-recreation opportunities tended to cease from 12-14 yr, at which point elite      102  players participated in organized female-only teams, even though they continued to engage in informal soccer activities (“play”) with boys. Unfortunately, the qualitative nature of this research led to the use of a small sample size, reducing the generalizability of the findings. Moreover, it is unlikely that developmental trends in the UK are mirrored in North American (i.e., Canada and the USA), where female soccer participation, starting at the youth level, has been a popular, high participation sport for a more extended period of time. Opportunities to play organized soccer for young girls is a relatively new development in much of Europe and the UK where the men’s game has dominated for many years (Williams, 2007).  5.1.1      Developmental practice pathways for elite sport  Several researchers have used deliberate practice theory (Ericsson, Krampe, & Tesch-Römer, 1993) and the Developmental Model of Sports Participation to examine the developmental pathways for elite athletes (DMSP; Côté, Baker, & Abernethy, 2007; Côté, Murphy-Mills, & Abernethy, 2012; Côté, 1999; for a soccer-specfic review, see Haugaasen & Jordet, 2012). The DMSP proposes that two pathways exist on the road to expert performance. One pathway is based upon high volumes of sport specific practice from an early age (~5 yr), through adolescence and into adulthood, termed the early specialization pathway. The second pathway, is based on early diversification where high volumes of play and multisport activity are engaged in during childhood with specialization occurring later during adolescence. Distinctions have been made between deliberate practice activities (i.e., structured, coach-led practice activities engaged in with the primary intention of improvement; Ericsson, Krampe, & Tesch-Römer, 1993), and what has been termed deliberate play (i.e., unstructured, peer-led sport activities, engaged in for the primary purpose of fun and enjoyment; Côté, 1999; Côté & Erickson, 2015; Côté et al., 2012).       103   Although the DMSP provides a broad theoretical framework to account for athlete development, the pathways outlined do not align well with published research involving soccer players (e.g., Ford et al., 2009; Ford & Williams, 2012; Haugaasen & Jordet, 2012; Hendry & Hodges, 2018). In general, elite male soccer players follow what has been termed an early, engagement pathway (Ford et al., 2009; Hendry & Hodges, 2018). According to this pathway, early development is characterised by majority engagement in one sport, but not at the exclusion of other sports, and involves high volumes of both structured practice and unstructured play activities. It is likely that this pathway best defines sports where participation rates are high and competition to succeed is equally high (e.g., soccer, basketball, and ice-hockey, although no data exist on the latter two sports). However, talent development opportunities are likely culturally and contextually dependent (Gledhill & Harwood, 2015).  While the women’s game has shown continual growth over the last few decades, participation rates are still low relative to men’s game (FIFA, 2007). As such, it is unknown whether the early engagement hypothesis that generally describes the developmental pathway of elite male soccer players (Ford et al., 2012, 2009) would define success in women’s soccer. Furthermore, since women’s soccer does not require peak performance at an age as young as gymnastics (Law, Côté, & Ericsson, 2008), this further questions whether an early specialization pathway would best describe how women players reach the elite level in soccer. However, a multi-sport, early diversification pathway, could best capture success, as supported by studies of elite level athletes across several Olympic sports (e.g., Güllich, 2016). We were uncertain as to the role that soccer play activities would have in the development of female elite players. It is possible that social norms and gender expectations could partially inhibit soccer play in girls,      104  such as kicking a ball around in the park or at recess in school (Clark & Paechter, 2007; Williams, 2007; cf., Gledhill & Harwood, 2014).    A number of claims have recently been made about the potential benefits of early play experiences and the diversification pathway when compared with the early practice and specialization pathway (e.g., Côté & Erickson, 2015; Côté, Murphy-Mills, & Abernethy, 2012). It has been postulated that early unstructured play experiences leads to more intrinsic forms of motivation when compared with early, structured, practice experiences (Côté & Erickson, 2015; Côté et al., 2012). However, this claim has not been supported in recent work with male, youth-elite soccer players (Hendry, Crocker, & Hodges, 2014; Hendry, Williams, Crocker & Hodges, in prep). Similarly, there have been suggestions that involvement in more play than practice experiences early in childhood will facilitate motor skill acquisition “naturally” and that this will benefit future sport expertise attainment through means of implicit learning (e.g., Côté & Hancock. 2014). However, there have been questions asked about the recommendation of play over practice, where opportunities for instruction, feedback, and guidance are low in the former (MacNamara, Collins, & Giblin, 2015). In studies where high volumes of play have discriminated across different skill groups in sport, play was always in addition to significant involvement in practice within their primary sport from an early age and as yet it is unclear whether the benefits of play are more related to increased volume rather than the development of specific skills (e.g., Baker et al., 2003; Ford et al., 2012, 2009; Haugaasen & Jordet, 2012; Hendry & Hodges, 2018; Hendry, Williams & Hodges, 2018).  In this paper, we evaluate practice histories of adult, elite women soccer players in Canada, playing at either the National Team or Varsity (Canadian University) level. Our aim was to evaluate practice history profiles with respect to current models of sport-skill development,      105  particularly early sport experiences with respect to multi-sport or specialized involvement and practice and play. Comparisons are made both across the two skill levels (Varsity and National), as well as with existing literature. While we might expect National female soccer players to engage in an early majority engagement pathway, similar to male soccer experts, there is reason to suspect that differences in participation rates overall and fewer opportunities for high level competition as adults might promote a more diversified, later engagement pathway. 5.1.2 Challenge perceptions  As a secondary research aim, we evaluated the discriminatory ability/sensitivity of a new index of developmental practice activities related to the concept of challenge. A significant concern in the sport expertise research relates to the predictive validity of developmental activity measures (Hodges, Huys, & Starkes, 2007; Hopwood, 2015). This concern stems, at least in part, from difficulties in attempting to recall and distinguish activities along distinctions of enjoyment- intention (i.e., play for pleasure/fun versus practice for performance improvement), as well as to a lesser degree, organization (i.e., coach-led practice, individual practice or play and peer-led practice). With these difficulties in mind, it may be a good time to determine other ways of establishing “quality developmental activities” that are more easily recalled by an athlete and which encompass all types of developmental sport activity. One possibility is based upon the challenge point framework which has been applied to help understand the best mentally-demanding conditions for motor learning (Guadagnoli & Lee, 2004).   According to the Challenge Point Framework, there is a theoretical optimal challenge point that emerges when the constant degree of difficulty inherent in the task, termed nominal task difficulty, is equal to or slightly higher than the skill level of the learner relative to the task, termed functional task difficulty (Guadagnoli & Lee, 2004). At this challenge point, the learner      106  is processing an optimal amount of information which maximises the potential for learning. Challenge that is either too high or too low will not facilitate learning and improvement to the same extent. The optimal challenge point can shift so that as the skill of the learner improves so does the degree of task challenge. In many respects, the challenge point framework is analogous to many of the components of Ericsson’s (1993) notion of deliberate practice in which expertise is achieved by continuously progressing practice away from automaticity and stretching the limits of the learner’s current capacities (i.e., practice designed to improve performance). The applicability of the challenge point framework to developmental practice activities in sports, make it a key consideration when attempting to find novel methods of capturing high quality experiences, whether in play, practice or competition.   One of the major differences between deliberate practice and play is based upon whether an individual is primarily engaged in an activity to improve performance (Côté, 1999; Côté & Erickson, 2015; Ericsson et al., 1993). As such, intention to improve performance separates practice from play, wherein the latter, the primary intention is fun. However, skill acquisition can emerge as a by-product from engagement in play, irrespective of any specific intention and in soccer, positive associations have been demonstrated between soccer specific play amounts and later soccer expertise (Ford et al., 2009; Ford & Williams, 2012; Hornig, Aust, & Güllich, 2016). Therefore, it may be that it is the functional difficulty or challenge experienced, whether in play, practice or competition, that is as, or more, important than the specific intention behind engagement in the behaviour. An assessment of the degree of challenge associated with each developmental activity at different milestones would provide an indication of the quality of the activity and best relate to or predict later attainment of expertise. Although it is likely that players with intentions to improve will more likely seek out high challenge situations, high      107  challenge is not necessarily a characteristic of practice, and play activities are not necessarily less challenging than those encountered in practice.   Several researchers have proposed that experience of competition is a vital component of the talent development process (Abernethy, Farrow, & Berry, 2003; Singer & Janelle, 1999). In youth development circles, competition is often viewed as an extension of the practice session with the experience of playing against unfamiliar opposition, psychological preparation, and dealing with travel and other variables viewed as beneficial to athlete development (Cook, Crust, Littlewood, Nesti, & Allen-Collinson, 2014; Holt & Dunn, 2004). This idea runs counter to original conceptualizations of competition as “work” when considered within a deliberate practice framework (Ericsson et al., 1993). However, the idea that competition, like work, is motivated by external rewards and lacks opportunities for experimentation and feedback is not universally shared. Although time spent in competition has not shown to be a discriminatory variable in sport domains, the total number and duration of games are often externally controlled by leagues or organizations and this likely contributes to the lack of difference across skill groups. It is likely that it is the type of competition that is important in developing skill. A “best vs. best” approach to games is continuously championed by coaches as being central to the development of young players. The idea behind this is that players are repeatedly placed in optimally challenging environments that equally tax their technical, tactical, physical, and psychological capacities, forcing them to adapt and improve. Therefore, analysis of the degree of relative challenge during developmental competition may add to the extant literature; informing as to the quality of the experiences (whether play, practice or competition) which define National players.       108   If the challenging nature of the activity is a good measure for activity quality, then soccer activities that are judged to be of moderate to high challenge, will differentiate across skill and potentially provide greater discriminability than assessments based upon accumulated soccer activity hours. Therefore, in addition to retrospectively assessing accumulated hours in general soccer activities, in the following study we assessed perceptions of challenge involved in various developmental soccer activities for elite (National) and sub-elite (Varsity) female soccer players. By measuring perceptions of challenge from childhood to adulthood, we assess trends as to when (i.e., what age group) and what specific activities (e.g., practice, play, competition) were characterised by high degrees of challenge. We anticipated that the challenge in practice and competition would increase linearly with age and that the more elite players (National) would have engaged in more optimally challenging developmental activities during childhood (5-12 yr) and adolescence (12-18 yr).  5.2 Methods 5.2.1 Participants  Participants were female, Canadian soccer players (N = 45), consisting of 21 National and 24 Varsity players. National players (M age = 28.26, SD = 3.95 yr) were participating at the international level of performance, ranked within the top 10 teams in the world. Varsity level athletes (M age = 19.60, SD = 4.31 yr) were currently competing at the highest level of soccer in the university system in Canada. No Varsity players had played or were expected to play adult-National team soccer, albeit 9 had represented Canada at various youth levels. Although all participants were adults, the National players were ~8 yr older than the Varsity players at the time of data collection, t(43)= 7.21, p < .001, d = 2.17. Participants provided written informed      109  consent before participating. All procedures adhered to the University’s REB guidelines and participants were given a small honorarium ($10 gift card) for participation. 5.2.2 Procedures  Contact was initially made with representatives from the National and Varsity teams via email correspondence before players were approached. After a briefing on the purpose of the study and informed consent, participants completed a series of questionnaires to provide information pertaining to: soccer milestones; developmental soccer activities; and developmental activity challenge (see Appendix D). These were completed in a room with supervision from members of the research team. The questionnaire took approximately 45 min to complete. 5.2.3 Questionnaires  Participation History Questionnaire (PHQ): In the past, the PHQ has been shown provide valid and reliable estimates of the developmental practice activities engaged in by athletes (e.g., Ford, Low, McRobert, & Williams, 2010). This type of retrospective survey is regarded as one of the best available methods for obtaining data on the developmental activity histories of elite athletes (see Hopwood, 2015). The PHQ consists of three sections. In section one, basic demographic information pertaining to start age in soccer (not in an organized league), supervised soccer practice, soccer competition, and participation in an elite development program is solicited. Further questions with respect to start age in co-ed soccer (i.e., playing on boy’s teams or any teams where boys and girls play together) were included in this section of the questionnaire.   The second component of the PHQ elicited information relating to estimates of hours accumulated in developmental soccer activities. Four activities were listed based upon previous research (e.g., Côté, Ericsson, & Law, 2005; Ford et al., 2009). These included ‘match-play’      110  (organized competition usually between two teams supervised by adult/s and engaged in with the intention of winning), ‘coach-led practice’ (organized group practice supervised by coach/adult engaged in with the intention of performance improvement), ‘individual practice’ (practice alone engaged in with the intention of performance improvement), and ‘soccer play’ (play-type games with rules supervised by oneself or peers and engaged in with the intention of fun and enjoyment, such as street or playground soccer). Athletes also recorded information pertaining to participation in other sports. For each component, players recorded: (i) number of organized practice sessions/week; (ii) average duration of each session; (iii) and months per year. Estimates for a typical week/training session were solicited from age 5 yrs to present in 2-year intervals (i.e., 5–6 yr, 7–8 yr). To aid retention, players completed the retrospective questionnaire in reverse chronological order. Linear interpolation methods were used to estimate values in intervening years. Accumulated hours in all soccer and other sport activities were calculated by multiplying the number of hours/session by the number of sessions/week and months/yr. Significant breaks through illness/injury were recorded and subtracted from yearly estimates. From these estimates, accumulated hours in soccer and other sport activities were calculated for childhood (5-12 yr) and adolescence (13-18 yr).   Perceptions of challenge: For each age-category (e.g., U19, U17, U15) participants were asked to recall the challenge associated with each developmental soccer activity (i.e., match-play, individual practice). Challenge was operationally defined as an activity that “continually tests your abilities, that is, is demanding and/or stimulating.” Participants were asked to respond using a 5 point scale where, 0 = Not at all challenging/easy, 1 = Some /low challenge, 2 = Moderate challenge, 3 = High challenge, 4 = Too much challenge/extremely challenging.       111  5.2.4 Statistical analyses  The data were checked for normality using the Shapiro-Wilk test. When the magnitude of skewness was less than 1, indicating only a tendency towards positive skewness (Bulmer, 1979), and there were no significant differences in homogeneity of variance between the groups, we used parametric methods (Glass, Peckham, & Sanders, 1972; Pallant, 2007). In cases where assumptions were not met (i.e., # of other sports, estimates of hours in soccer and multisport activities), we performed a log10 transformation to normalize these data. 95% confidence intervals around the mean difference are reported for independent t-test and mean pairwise comparisons.   5.2.4.1 Developmental milestones and activities  MANOVA was used to assess skill group differences (National, Varsity) for soccer milestones including start age in structured and unstructured soccer activities, competition, and entry into an elite development practice setting (i.e., academy). Follow up t-tests with Bonferroni adjustments were used to isolate the primary dependent variables responsible for any effects (Huberty & Morris, 1989). Separate analyses were conducted to determine the number of players and the duration for which co-ed soccer was engaged. For this analysis, we subtracted start age from end age. We compared number of activities, other than soccer, participated in during childhood and adolescence in a 2 Skill X 2 Age Period ANOVA with follow-up Bonferroni comparisons for significant effects.   For soccer activity estimates, a MANOVA was again used to compare across Skill level (National, Varsity) and Age period (childhood, adolescence), for practice, play and competition. Follow up ANOVA’s with repeated measures on the last factor, in addition to adjusted pairwise comparisons were used to ascertain differences in developmental activity hours across the      112  different ages. To compare estimates of hours in soccer practice versus hours in other sports we also ran separate 2 (Skill) x 2 (Activity; soccer practice, other sports) x 2 (Age period; childhood, adolescence) ANOVAs with repeated measures on the last 2 factors.   5.2.4.1 Reliability  To assess player-player reliability, two separate sections of the questionnaire asked for estimates of hr/week in current practice. In the first section, players provided an overall estimate of hr/week in soccer practice, whereas in the second, players provided separate entries for number of sessions/week and hours/session. In accordance with recommendations made by Atkinson and Nevill, (1998) and Hopwood (2015) we assessed the strength and similarity of estimates using intra-class correlation (ICC) and percent agreement (PA) scores respectively.    5.2.4.3 Challenge perceptions  To assess skill group differences in challenge ratings, MANOVA was used to compare across the 2 skill groups and 2 age periods for all soccer activities (practice, play, and competition). Mean challenge ratings were calculated by aggregating the mean individual challenge score for each activity across each year, for childhood and adolescence respectively. Follow up ANOVAs were conducted for each activity. Using the challenge ratings we calculated accumulated hours in challenging soccer activities by aggregating the total number of hours in activities rated as either highly or extremely challenging (i.e., 3 and 4 on the perceived challenge scale) and compared these across skill in an independent t-test.    For any significant ANOVA effects (p < 0.05), post-hoc pairwise comparisons were applied with Bonerroni corrections. Greenhouse Geisser corrections were applied to correct for sphericity violations across all analyses. Partial eta-squared values are reported for significant ANOVAs and Cohen’s d as a measure of effect size for pairwise comparisons.       113  5.3 Results 5.3.1 Developmental activity milestones  The current age and ages at which the National and Varsity players reached soccer milestones are presented in Table 7. With respect to childhood milestones, MANOVA indicated that the two groups did not differ overall; F(5,38) <1, Wilks’ λ = .86, ηp2 = .16. Secondary analyses on the various dependent measures showed that age of participation in structured soccer practice, t(43) = 2.43, p = .02, d = .78, Mdifference = 1.23 yr, 95% CI [.02, 2.23] and start age in an Academy environment, t(44) = 2.58, p = .02, d = .89, Mdifference = 3.28 yr, 95% CI [1.02, 5.54], showed that these ages were surprisingly later for the National group.   A similar number of National (n = 15) and Varsity players (n = 14) had played co-ed soccer as a child and they did not differ in their average number of years played, t(27) = 1.21, p =.23, d = .04, Mdifference = .58 yr, 95% CI [-1.16, 4.72]. With respect to sporting diversity, National players partook in fewer sports than the Varsity players, F(1,44) = 7.43, p =.01, ηp2 = .14, Mdifference = 2.03 sports, 95% CI [0.64, 3.42] and this did not depend on age period, F(1,44) = 1.05, p = .31, ηp2 = .02 (see Table 7). 5.3.2 Developmental activity hours  The mean accumulated hours in soccer activities across the developmental timespan are presented in Figure 6 (5 – 18 yr, that is, U6-U19 yr age-groups). Soccer activity estimates were first log-transformed before analysis (Tabachnick & Fidell, 2007). Since ~38% of participants did not participate in individual practice from U6 to U11 yr and only approximately 20% from U12 – U16 yr, we combined individual and coach-led practice estimates to get a combined practice measure. The 2 Skill x 2 Age-period (childhood, 5-12 yr; adolescence 13-19 yr) MANOVA conducted on accumulated estimates across the three soccer activities yielded      114  significant skill, F(3,42) = 6.79, p = .01, ηp2 = .36, Wilks’ λ = .65, and age period, F(3, 42) = 12.87, p < .001, ηp2 = .51, Wilks’ λ = .49 effects.  There was no interaction, F<1, ηp2 = .06, Wilks’ λ = .94.    Follow-up RM ANOVA’s showed skill-group differences in estimates of hours accumulated in soccer play, F(1, 44) = 13.62, p = .01, ηp2 = .26, but not in practice or competition (Fs<1). National players engaged in more soccer play compared to Varsity players, Mdifference = 519.26 hr, 95% CI [220.86, 817.59]. There were significant differences across age periods (childhood and adolescence) for accumulated hours in both soccer practice, F(1, 44) = 23.09, p = <.001, ηp2 = .37, Mdifference = 1045.84 hr, 95% CI [768.50, 1323.17] and competition, F(1, 44) = 38.09, p = <.001, ηp2 = .49, Mdifference = 204.93 hr, 95% CI [132.67, 276.35], but not soccer play (F<1). More hours were accumulated in adolescence than childhood.   We compared the log-transformed accumulated hours in other sports with accumulated hours in soccer practice using a 3-way (Skill X Age-period X Activity) repeated measures ANOVA. We have plotted these data in Figure 7. There were no skill-related effects only significant main effects for activity, F(1,44) = 232.62, p < .001, ηp2 = .88, and age period, F(1, 32) = 16.22, p < .001, ηp2 = .33. Players engaged in more soccer practice hours compared to other sports (Mdifference = 3238.48 hr, 95% CI [2997.66, 3478.34]) and engaged in more activity in adolescence compared to childhood (Mdifference = 2502.95 hr, 95% CI [2152.37, 2853.53]). The only significant interaction was for Age-period and Activity, F(1, 44) = 7.42, p = .01, ηp2 = .19. Bonferroni adjusted pairwise comparisons indicated that players engaged in more soccer practice hours compared to other sports during both childhood (p < .001, d = 1.65, Mdifference = 256.20 hr, 95% CI [141.47, 370.53]) and adolescence (p < .001, d = 26.98, Mdifference = 2157.38 hr, 95% CI [2151.17, 2242.83]) and more soccer practice during adolescence than during childhood (p <      115  .001, d = 5.74, Mdifference = 2168.34 hr, 95% CI [1305.40, 3031.28]). There were no differences across age periods for hours in other sports (p = .25, d = 1.17).  5.3.3 Within questionnaire player-player reliability  For National players, the strength and similarity of estimates of current hours in soccer practice activities were high (ICC = .85, PA = 90.75), whereas for the Varsity players the strength of relationship between variables was moderate (ICC = .54) but estimates were highly similar (PA = 88.75). Based on inspection of within group standard deviations, estimates of current hours per week in soccer practice were consistent in both the National (M = 10.43, SD = 1.16) and Varsity (M = 10.66, SD = 1.45) players 5.3.4 Childhood and adolescent challenge ratings  We first compared the three soccer-related activities with respect to the ratings of challenge using a 2 Skill X 2 Age-Period MANOVA. These data are displayed in Table 8. We have also illustrated the change in ratings across the various ages in Figure 8a and b. The highest ratings of challenge were given for competition (i.e., match play) and the lowest ratings for play. There was a significant skill effect showing that the National players generally engaged in activities that were rated as more challenging than the Varsity players, F(3,42) = 8.36, p = .01, ηp2 = .37, Wilks’ λ = .63. There was also a significant age-period effect, F(3,42) = 25.37, p < .001, ηp2 = .64, Wilks’ λ = .36, but no interaction, F<1, ηp2 = .05, Wilks’ λ = .95.    Follow-up RM ANOVA’s showed skill-group differences in challenge perceptions for soccer play, F(1, 44) = 15.77, p = .01, ηp2 = .26, Mdifference = 1.14, 95% CI [.51, 1.76] and a trend for these differences for competition, F(1, 44) = 3.31, p = .07, ηp2 = .07, Mdifference = .94, 95% CI [-.04, 1.92]. The National players rated play (and somewhat competition) as more challenging than Varsity players. However, there were no differences for ratings of practice, F(1, 44) = 1.33,      116  p = 26, ηp2 = .03. There were significant differences in perceived challenge across age periods (childhood and adolescence) for competition, F(1, 44) = 71.63, p <.001, ηp2 = .62, Mdifference = .21, 95% CI [.14, .28] and soccer practice, F(1, 44) = 15.77, p = <.001, ηp2 = .26, Mdifference = .42, 95% CI [.24, .59], but not play, F(1, 44) = 1.45, p = 23, ηp2 = .03. As expected, practice and competition were rated as more challenging in adolescence compared to childhood.   As detailed at the bottom of Table 8, we calculated accumulated hours in practice for each age-period which were rated as being either challenging or highly challenging (i.e., 3 or 4 rating). These data were compared in a 2 Skill X 2 Age-period MANOVA, but there were no significant effects involving skill (i.e., skill, F(3,42) = 1.37, p = .26, ηp2 = .09, Wilks’ λ = .91; Skill X Age-group, F<1, ηp2 = .07, Wilks’ λ = .93). An age-period effect showed that all players accumulated more hours in challenging soccer activity during adolescence, F(3,42) = 29.58, p < .001, ηp2 = .68, Wilks’ λ = .31.   5.4 Discussion We provide a descriptive, cross sectional comparison of the developmental activities engaged in by National and Varsity female soccer players, allowing us to better understand pathways to elite performance in reference to extant models of sport-skill expertise. We also adopt challenge based measures (based on the challenge point hypothesis, Guadagnoli & Lee, 2004) to help determine developmental soccer activities deemed to be of “high” quality (see recommendations; Ford et al., 2015). We anticipated that National level female soccer players would invest large volumes of time in soccer practice, but we were unsure as to the extent to which players would focus primarily on soccer practice compared to other sports as well as their engagement in soccer play activity (i.e., between and within sport diversity). Furthermore, we hypothesized that the      117  National players would engage in more challenging soccer activities during childhood and into adolescence than Varsity players.   Overall, both the National and Varsity players engaged in higher volumes of practice than play in soccer and more hours in competition than play during both childhood and adolescence. Moreover, the National players had amassed more hours in play than the Varsity players, but their estimates were similar for practice and competition. This greater engagement in activities outside of coach-led, organized practice might reflect more overall “practice” volume in general for the National players (as confirmed by the MANOVA on overall hours in soccer activities; cf., Hendry & Hodges, 2018). Alternatively, it may be that play offers secondary benefits to aspiring athletes, such as opportunities to be more creative, which are perhaps not realized in a more structured coaching environment (Chow, Davids, Renshaw, & Button, 2013; Côté & Erickson, 2015; Hendry & Hodges, 2013; Memmert, 2015).   Regardless of these observed differences in “play”, as a function of skill, the estimates of hours in play could be considered relatively low and point towards a more coach-driven approach to women’s soccer than has typically been reported in male players. Although the female players had amassed a similar number of hours in soccer practice as compared to elite males by age 16 yr (~ 3000 hr; Hendry & Hodges, 2018, Ford et al., 2012), the low volume of soccer play hours contrasts to the high volumes of soccer specific play engaged in during childhood by elite male players (Ford et al., 2012, 2009; Ford & Williams, 2012; Hendry & Hodges, 2018). For example, academy-based, elite youth soccer players in Scotland had accumulated ~6 times more hours in soccer play in comparison to the estimates provided by our National women players by age 16 yr (Male = ~3000 hr, Female = ~ 500 hr; Hendry & Hodges, 2018). Furthermore, a sample of Canadian recreational, yet competitive, male players had      118  participated in more than double the amount of play compared to National women players before the age of 16 yr (~1200 hr; Hendry & Hodges, 2018).  It is unclear why play volumes are relatively low in elite female soccer players compared to males. It may be that more opportunities to engage in play activities in childhood, at least among this current sample, were low, perhaps with boys tending to dominate the playground at recess. Some researchers have remarked on the negative socio-cultural expectations that may exist for females engaging in soccer play outside of formalized practice (Williams, 2007), although this may be more of a European phenomenon (rather than Canadian/USA), where traditionally soccer has been viewed as a “man’s game” (Clark & Paechter, 2007; Pfister, 2015). Finally, these play hour differences across the sexes might be reflective of early differences in inherent interest/motivation to play soccer.   In all existing developmental models of sports-skill development, childhood is deemed a crucial period in the attainment of adult-expertise, perhaps because skills are acquired during these potentially sensitive periods of development (Côté et al., 2012; Ford et al., 2009). The early specialization pathway dictates that in order to become an elite athlete in adulthood, a player must engage in high volumes of domain (soccer specific) practice early on. Although start age is not a critical component of deliberate practice theory (Ericsson et al., 1993), on which this pathway is based, in order to be able to compete with peers in terms of practice hours, early accrual of relevant practice is seen as an important component of later success. In fact, over 90% of the female players had begun participation in soccer activities from an early age (~5-6 yr). Different from previous published data based on elite male players (Hendry & Hodges, 2018), women National players had started participation in structured soccer activities and specialized “academy” practice later than the Varsity players (based on follow-up t-tests). While this might      119  be suggestive of benefits associated with starting coach-led, organized soccer training later rather than earlier in childhood, these differences might just reflect age differences across the two groups. The younger players where mostly made up of Varsity players in this study who were almost a decade younger than their National level counterparts. It is possible that in the ensuing decade the increased access to organized and systematic training academies may merely have created a situation where players engaged much earlier than was the case with previous generations. These differences in access to structured practice activity between groups may have been compensated for by the National players by the reported increased volumes in soccer play.   Overall, these data are relatively consistent with the early majority engagement pathway which defines soccer success in male elite players (Ford et al., 2009; 2012; Hendry & Hodges, 2018). National and Varsity players engaged primarily in soccer practice from an early age and, while they participated in a variety of other sports, they participated predominantly in soccer practice compared to other sports even in childhood. Although the hours accumulated in soccer play were low in comparison to male players at similar levels, it is noted that in keeping with the early engagement hypothesis, National players participated in more soccer play and rated soccer play as more challenging than Varsity players. No players from either group specialized exclusively in soccer.  In addition to measuring practice amounts, we determined ratings of challenge to assess how activity quality contributed to expertise development. Our descriptive comparisons across groups indicated that competition was rated as the most challenging activity and play as least challenging. The National players rated the developmental activities, across both childhood and adolescence, as more challenging than ratings given by the Varsity players, particularly play and to some extent, competition. Previously, researchers have failed to show discriminability of      120  competition in relation to developing expertise, at least when investigating competition quantity (e.g., Ford et al., 2009; 2012). Although Ericsson and colleagues (Ericsson et al., 1993) viewed competition as a work activity that contributed little to expertise attainment, these data suggest that high quality competition during development (as operationalised through challenge) could be an important discriminatory variable (Abernethy et al., 2003; Cook et al., 2014; Holt & Dunn, 2004; Singer & Janelle, 1999).   The retrospective recall technique adopted is still regarded as one of the best methods we have available for collecting practice history data (e.g., Hodges, Huys, & Starkes, 2007; Hopwood, 2015). The PHQ has been designed to optimally aid recall, including soliciting data from current years and working backwards to avoid inflation bias and using coach and team prompts to facilitate recall of practice amounts (Low, Williams, McRobert, & Ford, 2013). Although we were unable to collect data from parents and coaches to test for reliability, mostly because of the varied backgrounds and locations of the players, we were able to show within group consistencies, at least with respect to current practice hours. However, differences in the current age of participants, and as a result, potentially differences in increased access to soccer infrastructure, may have influenced estimates of activity quantity as a function of skill. It is worth noting that the estimates were generally similar across the samples, at least for practice. Furthermore because we only compared activity estimates through childhood and adolescence, differences in current age were partially controlled. Future work may benefit from corroborating activity estimates with player training diaries and records kept from national governing bodies and youth development clubs.   The current study adds to the literature in two ways. First, this is one of the first studies to describe and detail the developmental pathways engaged in by world class female soccer players.      121  Second, attempts were made to measure the quality of developmental activities, based on assessment of challenge (Guadagnoli & Lee, 2004). The fact that challenge distinguished across skill-groups points to the validity and potential usefulness of this measure in helping chart developmental practice activities which contribute to elite development in sport. As per the recommendations made by Ericsson and colleagues (Ericsson et al., 1993; Ericsson & Pool, 2016; Ericsson, 2014), in order to best determine activities which distinguish across skill groups, measures need to be ascertained that evaluate the quality as well as quantity of practice. This might be in terms of the intention, how the activity has been structured or the degree of challenge (or potentially all of the above). However, challenge quality did not distinguish across skill groups with respect to hours in challenging soccer related activities. Moreover, these challenge ratings have not been validated and although they are provided for multiple activities at each age category, they are only a single-item rating of challenge. It will be important in the future to validate this scale in field or lab. work, perhaps using self-report and/or psychophysiological measurement.   In summary, we have presented data showing that world-class (National) and sub-elite (Varsity) female soccer players in Canada show developmental profiles consistent with the early majority engagement pathway reported in male elite soccer players. National team players participated in greater volumes and more challenging soccer play than less elite, Varsity players. The differences across skill-groups with respect to challenge offers an interesting platform for researchers to pursue in future with respect to both current and retrospective assessment of developmental activity quality. There is a need to validate and extend this research, most obviously with current, female youth players in order to make stronger conclusions about      122  pathways which are most conducive to success, particularly when based on a prospective, longitudinal investigation.        123  Table 7. Mean ages (SD) for soccer milestones for National and Varsity women soccer players and number of other sports participated in childhood (5-12 yr) and adolescence (13-19 yr) ______________________________________________________________________________      All   National   Varsity  (n = 21)        (n =24)      ________________________________________________ Soccer milestones  Start age in soccer activities  4.95 (1.64)   5.43 (2.06)  4.50 (.96)  Start age in soccer practice  5.65 (1.85)   6.28 (2.19)  5.05 (1.21)*  Start age in soccer academy           14.03 (3.90)            15.42 (2.98)           12.14 (4.31)*  Start co-ed soccerǂ   7.37 (4.42)   8.00 (5.52)  6.80 (3.19)  End co-ed participation           10.69 (4.9)            11.51 (5.92)  9.73 (3.75) # Other sports  Childhood    3.80 (2.50)  2.77 (2.05)  4.75 (2.36)  Adolescence    4.28 (2.51)  3.64 (2.59)  4.88 (2.32) ______________________________________________________________________________ * significant difference between means at p<.05. ǂ, for this analysis, National (n = 15), Varsity (n = 14).     124  Table 8. Mean (SD) challenge rating and accumulated hours in “high challenge” (ratings of 3 and 4) soccer activities (competition, practice, play), during childhood and adolescence for the National and Varsity women athletes. ____________________________________________________________________________________________________________       National      Varsity     Childhood  Adolescence   Childhood  Adolescence Challenge ratings (0-4)   Practice   1.60 (.83)  1.76 (.61)   2.24 (.74)  2.38   (.61)  Play      .88 (.81)  1.10 (.98)      .46 (.70)      .69   (.90)  Competition   2.15 (.78)  3.32 (.73)   1.85 (.88)  2.93 (2.72)  Accumulated hrs. Practice    221.94 (285.39) 274.86 (338.86)  189.07 (362.29) 270.52 (347.34) Play     74.74 (113.38)   84.33 (228.80)   38.66   (42.43)    36.88   (25.54) Competition   218.97 (180.97)  661.63 (522.53)  212.91 (138.85) 589.49 (138.85)  ____________________________________________________________________________________________________________           125  a)            b) =  Figure 6a & b. Mean (and SD bars) for accumulated hours in soccer activities (practice, play and competition) by National (a) and Varsity (b) soccer players from the under 6 yr age-group (U6) to under 19 yr (U19).  010002000300040005000600070008000U6 U7 U8 U9 U10 U11 U12 U13 U14 U15 U16 U17 U18 U19Accumulated hoursAge group (yr)PracticePlayCompetition010002000300040005000600070008000U6 U7 U8 U9 U10 U11 U12 U13 U14 U15 U16 U17 U18 U19Accumulated hoursAge group (yr)    126   Figure 7. Mean accumulated hours (and SD bars) in soccer practice and practice in other sports as a function of age period (childhood or adolescence) and skill (National, Varsity).     0500100015002000250030003500400045005000Childhood soccer Childhood othersportAdolescencesoccerAdolescence othersportsAccumulatd hoursActivity type and Age periodNationalVarsity    127  a)  b)   Figure 8 a & b: Mean (and SD bars) for ratings of challenge across age-group and soccer activity (practice, play, competition) for a) National and b) Varsity players.  0123456U6 U7 U8 U9 U10 U11 U12 U13 U14 U15 U16 U17 U18 U19Challenge rating (range = 0-4)Age group (yr)PracticePlayCompetition0123456U6 U7 U8 U9 U10 U11 U12 U13 U14 U15 U16 U17 U18 U19Challenge rating (range = 0-4)Age group (yr)PracticePlayCompetition    128  Chapter 6: General discussion  In this thesis, I presented a multi-method approach based on the highest level of elite, youth soccer in the UK and adult women’s soccer in Canada. This approach involved a combination of cross-sectional comparisons based on retrospectively collected practice histories as well as some prospective follow-up of soccer-related developmental activities and progression through the youth to adult professional ranks. In general, these findings provide support for the early engagement hypothesis (Ford, Ward, Hodges, & Williams, 2009), characterized by primary involvement in high levels of soccer specific practice and play throughout the developmental  time-span (5 – 18yr). With respect to conceptual contributions, the data from my thesis cast doubts on the overall utility of the Developmental Model of Sport Participation (DMSP; Côté, 1999; Côté, Lidor, & Hackfort, 2009; Côté, Murphy-Mills, & Abernethy, 2012) and its related postulates, at least when considering soccer development and potentially other large participation sports. The importance of structured, coach-led soccer specific practice is outlined in relation to the development of skills associated with achieving success along the soccer development pathway. I also present data supporting the importance of a long-term approach to talent identification and selection, based upon technical skills and tactical awareness. Finally, I offer some methodological and conceptual insight as to how we might be able to learn about developmental practice activities which are most related to elite-adult status, based on the concept of challenge      129  6.1 Synthesis of study findings 6.1.1 Men’s soccer  6.1.1.1 Early majority engagement    Elite level players followed an early majority engagement pathway, in which they invested a majority of time in soccer specific practice and play from an early age, in comparison to multi-sport activities. The prospective design employed within Study 1 provided the opportunity to identify which developmental soccer activities distinguished across skill at two key transitions in the development of elite-level soccer players. Those players that progressed from an elite academy, affiliated with a professional club, and onto professional contract at approximately 16 years of age amassed more hours in soccer practice during childhood than those players that did not. Contrary to previous studies (e.g., Ford, Ward, Hodges, & Williams, 2009; Ford & Williams, 2012), hours in soccer play did not differentiate across the groups statistically. However, these successful youth professionals did engage in approximately 600 more hours in soccer specific play during childhood in comparison to their peers who did not transition from youth-academy levels. Despite failing to reach significance, due primarily to the large variability with these data, this difference is sizeable especially when considered against the backdrop of previous data showing how hours outside of coach-led practice in play distinguish across skill. At the adult professional level, hours in soccer play did differentiate those that progressed to adult professional status versus those that did not. Also across their careers, the adult professionals amassed a greater proportion of time in play relative to practice (as expressed as a percentage) than those that failed to make the transition from youth-professional status. Overall, these findings generally align with the early engagement hypothesis     130  (Ford et al., 2009), and point towards the importance of soccer play as being a vital ingredient in the long term development of professional adult soccer players.  Professional male players (youth and adult) were also recruited into a professional youth academy at an earlier age and participated in fewer multi-sport activities than those players that did not progress beyond their respective schoolboy academies. Closer inspection of the male data, across all skill levels, revealed that only a very small percentage (< 10%) of participants truly specialized in soccer. Importantly, no professional male players, and none of the women players had specialized in just soccer. Athletes reported participating in several (~4) other sports in addition to soccer. In this regard, the players did not meet the criteria of specialization outlined within the DMSP (e.g., single sport involvement, high volumes of deliberate practice from an early age; Côté, 1999; Côté, Lidor, & Hackfort, 2009; Côté, Murphy-Mills, & Abernethy, 2012) and engaged in more multi-sport activity than those players outlined within the early engagement hypothesis (Ford et al., 2009). While this might lead one to advocate towards an early diversification/sampling approach espoused with the DMSP, estimates of weekly hours in soccer practice activity were considerably greater than estimates of hours in other sports activities. In fact, players reported spending approximately six times as many hours in soccer activities than their second most popular sport during childhood. Based upon these data a blended approach to athlete development involving early majority engagement in soccer practice and play activities coupled with some limited participation in other sports appears to be the most fruitful pathway towards soccer expertise.       131   6.1.1.2 Development of soccer-specific skills In Study 2, I investigated how well coach ratings of technical, tactical, physical and creative skills, discerned at two-time points, discriminated across future skill groups with respect to successful transitions from Academy-only to youth professional and eventually to adult professional status. These skills are generally considered as the most important elements for success in soccer (e.g., Williams & Reilly, 2000). I also assessed relationships between these various skills and developmental soccer activities (i.e., practice, play and competition). As might be expected, those selected as youth and adult professionals were rated as being more skillful than their less successful counterparts.  At the youth professional transition, ratings of technical, tactical and physical skills were higher for those players that successfully attained professional status in contrast to those that did not. Within the group of professional youth players, the determining factor in successfully transitioning to adult professional status was related to tactical skill, and to lesser extent technical skill (p = .06).  Childhood is postulated to be a particularly important time-period for the development of perceptual-cognitive and perceptual-motor skills (e.g., Côté & Hancock, 2014; Côté et al., 2009, 2012). According to these authors, involvement in childhood deliberate play (e.g., unstructured or peer-led, highly enjoyable games, using rules adapted from the adult form that are typically monitored by the athletes), offers advantageous perceptual and motor conditions which can facilitate skill acquisition to a similar or even greater extent to those provided within more structured practice environments (e.g., Côté, Coakley, & Bruner, 2011; Côté et al., 2012; Côté & Erickson, 2015; Côté & Hancock, 2014). Within a group of highly skilled youth soccer players, results from Study 2 did not align with these assertions. There were no significant correlations between involvements in childhood soccer practice or play amounts with the outlined skill     132  ratings as assessed at time 1 by expert soccer coaches. However, when viewed across a longer time frame, career practice amounts, but not play, were significantly related to later (time 2) scores of tactical, technical and creative skills. This latter finding suggests that the relationship between variables takes time to develop. As for practice being predominantly related to skill, it is likely practice offers a higher quality learning environment than play. Coaches reported using a mixture of highly task relevant, containing variable practice with plentiful decision-making opportunities that would lead to maximum transfer to soccer performance. In addition to effective instruction and feedback provided by expert level coaches, these conditions are likely to positively influence skill development in soccer.   6.1.1.3 Soccer activities and Self-Determined Motivation (SDM) In the third study, I investigated the dynamic nature of self-determined motivation (SDM) in elite youth soccer and studied relationships between developmental soccer activities and SDM (e.g., Côté & Erickson, 2015; Côté & Hancock, 2014; Côté et al., 2009, 2012). These research goals were predicated upon postulates derived from the DMSP suggesting a positive relationship between high volumes of deliberate play activities during childhood and intrinsic regulation and self-determined motivation (e.g., Côté & Erickson, 2015; Côté et al., 2012). In an earlier study  I conducted (not part of the PhD thesis) older (under 17 yr), elite male soccer players scored lower on measures of autonomous motivation than younger (under 13 yr & 15 yr) age group players (Hendry, Crocker, & Hodges, 2014). Within this “older group,” controlled motivation (i.e., lower SDM) was positively related to practice hours accumulated within an elite soccer academy. In Study 3, SDM was compared cross sectionally across both elite and recreational U15 yr and U17 yr male soccer players, and prospectively across two time-points within the elite group of participants. There were no significant relationships, in any direction, between childhood play     133  and indices of SDM. However, recent hours in practice (practice that took place over the last 2.5 yr of analysis; T1-T2) was negatively related to indices of SDM and autonomous motivation in both elite and non-elite groups. Hours spent in recent practice were also positively related to controlled motivation.  In general, these findings support previous findings and question the accuracy of the proposal made by Côté and colleagues (Côté, Baker, & Abernethy, 2007; Côté et al., 2009, 2012; Côté & Erickson, 2015; Côté & Hancock, 2014). No evidence that early play experiences fostered self-determined forms of motivation was reported, nor did involvement in childhood play provide players with any sort of protective mechanism against some of the negative motivational effects (i.e., controlled motivation) associated with highly structured practice. Based upon the high scores of intrinsic motivation (6.72 from a max score of 7) from the elite academy players it is likely that they derive a great sense of enjoyment from engaging soccer practice from an early age. It is also likely that the high level of youth coaching expertise provided by the elite academies offers highly relevant, engaging and enjoyable practice sessions. Analysis of practice microstructure of practice (not reported in this thesis), indicated that the players were engaged in approximately 50% of their practice time playing form activities (e.g., small sided games, 2v2, opposed practice, see Low et al., 2010). These types of activities are far removed from the overly prescriptive, externally controlled, coach-led drill and grid type activities typically associated with traditional forms of practice which are likely to be less enjoyable and intrinsically motivating.  Overall, the pattern of results indicated that changes in motivation were related to both age and skill. That is controlled motivation was generally higher in the older than younger players and higher in the elite versus non-elite players. While these effects might paint an overall     134  negative picture of the elite soccer Academy environment with respect to SDM, it is noteworthy that autonomous motivation scores were still generally higher in the elite players compared to the non-elites. In general, the elite athletes exhibited a co-existing pattern of motivation, characterized by high scores of both controlled and autonomous forms of motivation. Similar patterns of co-existing motivations have been shown in previous studies of elite level athletes using SDM (Briere, Vallerand, Blais, & Pelletier, 1995; Gillet, Berjot, & Gobancé, 2009; Gillet, Berjot, Vallerand, Amoura, & Rosnet, 2012; Pelletier, Fortier, Vallerand, & Brière, 2001). Based upon well-established findings from educational research (Deci, Koestner, & Ryan, 2001), we suggest that the external rewards associated with the attainment of professional status (e.g., money, status, fame) likely influenced the general shift towards more controlled motivation within the older elite level athletes.  6.1.2 Women’s soccer  6.1.2.1 Early engagement pathway In the last Study of my thesis, group differences in the quantity and quality of the developmental activities engaged in by world-class (National team) and highly skilled (Varsity) female soccer players in North America were investigated. An attempt was also made to answer calls made by prominent researchers to measure practice quality in addition to practice quantity (e.g., Ericsson, & Pool, 2016; Ford, Coughlan, Hodges, & Williams, 2015). To achieve this aim, players provided subjective ratings of the degree of challenge experienced during developmental, soccer-related activities. This work was based upon the challenge point framework (Guadagnoli & Lee, 2004) which posits that optimal learning occurs when the difficulty of the task is equal or slightly greater than the skill level of the performer relative to the task.      135  The general pattern of developmental activities engaged in by elite (world-class) and sub-elite female soccer players was similar to that noted by elite male players (Ford et al., 2009; Ford & Williams, 2012). Female National players engaged in a greater volume (and more challenging) soccer play than Varsity players.  However, contrary to males, female players amassed the majority of their soccer development within structured, coach-led, practice environments, as opposed to unstructured, soccer play type activities. Furthermore, within both groups, female players amassed less time in soccer play than in soccer competition. Again, this is different from data derived from male players, whereby competition accumulates the least amount of activity hours. Given the relatively low volume of soccer play engaged in by female players, it is likely that engagement in more unstructured play like activities would benefit overall development, even if this engagement does little more than increase overall soccer activity hours. In comparing across a skill class, the groups accumulated similar volumes of practice and competition. However, the elite group participated in more soccer play than the sub-elite players. Overall, these data are consistent with the early engagement hypothesis (Ford et al., 2009).   With respect to soccer milestones, world class female soccer players began structured soccer practice later than sub-elite players and entered an elite soccer academy later than sub-elite players. Similarly, the Varsity players participated in a greater number of other sports (~5) than elite National players (~4), albeit both groups participated in a large variety of multisport activities. Although, these findings suggest benefits towards a later start date in soccer practice activities, there were no significant differences in total practice hours. Given the older average age of the elite (world-class) players in comparison to the Varsity group, differences in academy start age may be attributed to fewer available developmental “academy-type” opportunities to older players in comparison to the younger, sub-elite players.     136  6.1.2.2 Perceptions of challenge Measures of perceived challenge yielded several novel findings. Across all participants, competition was viewed as the most challenging developmental activity. There was a trend for differences across the skill groups with respect to this measure. This was somewhat unexpected since previous research measuring soccer activity quantity has failed to show any relationship between competition and expertise (e.g., Ford et al., 2012, 2009; Ford & Williams, 2012; Haugaasen & Jordet, 2012; Ward & Williams, 2003). This lack of association between variables is likely accounted for by the externally controlled duration and frequency of match play in youth soccer, with governing bodies instituting maximum hours in competition. Further investigation is required to establish the specific nature of perceived challenge across all developmental activities and how this may be beneficial to development. This could be accomplished through analysis of the frequency, variability and difficulty in the number of technical skills or decision points encountered within and across each type of activity (practice, play and competition).  As expected, soccer practice and competitions were rated as being more challenging during adolescence in comparison to childhood. One would expect that as players are progressively funneled through the soccer development system, they will encounter higher quality practice and competition environments.  Soccer play was rated equally low in challenge across both time periods. Play was perceived as being less challenging than practice and competition across both skill groups. However, world-class players rated soccer play as being significantly more challenging than the sub-elite players. The inclusion of the challenge variable, offers an interesting starting point for future research into the conditions contributing to perceived challenge and its application to other contexts and related outcomes. This type of     137  research offers opportunity to tease out the differences in perceived challenge across activity types, as well as further analyze the specific technical, tactical and physical challenges faced by players in their development. Time use analysis (Ford, Yates & Williams, 2010) and experience sampling methods (Csikszentmihalyi, 2014) offer suitable methods to collect these data in situ and provide a more reliable measure less prone to memory recall and bias.  6.2 Conceptual contributions  The results from this thesis offer several contributions to knowledge and understanding of developmental pathways to success in soccer. These include support for an early majority engagement pathway, analysis and scrutiny of the DMSP and related pathways, knowledge of relationships between soccer skills and soccer activities at a within-group level, evidence showing the dynamic nature of SDM and initial evidence showing that measures of challenge and the challenge point framework in general, is applicable to the study of long term practice conditions which promote sport expertise. In addition to these contributions, I have used a wide range of methods and populations to study elite soccer development, adding to the general validity, including prospective research designs, samples of truly elite-level participants including world-class elite female players and adult professional male players in UK soccer, and a multi-variable measurement approach.  6.2.1 Early majority engagement in soccer  The pattern of results from the male players provides general support for the early engagement hypothesis (Ford et al., 2009). Male players that progressed to professional status engaged in high volumes of soccer specific play and practice from an early age. Although hours in soccer play did not differentiate between those players that received a professional contract at age 16 yr versus those that did not, play hours were approximately 600 hr greater in the former     138  group. Furthermore, adult professionals engaged in more soccer play hours than their youth professional-only counterparts.  One difference between data from my thesis and those reported elsewhere with respect to the early engagement hypothesis relates to the amount of multi-sport activity engaged in by participants. Ford and colleagues (Ford et al., 2009), reported a mean of ~1.5 sports during childhood, whereas participants in my Study 1 participated in ~5 other sports during childhood. However, multi-sport activity is not uncommon in other studies of developmental soccer activities among male elite players (for a review see, Haugaasen & Jordet, 2012). Across the male youth athletes and the women’s data, I directly compared the amount of time accumulated in childhood in soccer activities versus other sports. Across both populations, there was a clear pattern indicating that players engaged in soccer related activities the majority of time compared to other sports. As reported above, elite youth soccer players engaged in approximately six times as many hours in soccer practice during childhood than the next most popular sport. For women there was a similar trend, with elite players engaging in significantly more hours in soccer activity compared to the next five most popular sports. The high volume of sports engaged in by participants could also be viewed as evidence in support for an early diversification pathway (e.g., Côté & Erickson, 2015; Côté et al., 2009, 2012). However, there was no evidence that more hours or more diversified sport experiences were good for development. In fact, there was a tendency for the reverse to be true within Study 1 (where fewer sports were engaged by the more successful male players). Although diversity was apparent in the type of soccer activities (i.e., practice and play), the time amassed in other sport activities were considerably less than those amassed in soccer practice. Based upon these data it was concluded that athletes following an early majority engagement pathway accrue the benefits     139  related to early soccer involvement associated with early specialization and accrual of high volumes of practice and play. Although this is not at the exclusion of engagement in other sports, there was no evidence that multisport participation was an advantage for later success at adult levels.  In line with Ericsson’s monotonic beliefs assumption (Ericsson, Krampe, & Tesch-Römer, 1993), early involvement in soccer practice would maximize opportunity to engage in highly relevant deliberate practice experiences (or purposeful practice; see Ericsson, & Pool, 2016). In the thesis, evidence is presented to support the importance of high volumes of practice in Study 2, where a positive relationship was reported between accumulated soccer practice hours during development with coach ratings of tactical, technical and physical skill. Early selection into an elite academy, which differentiated between professional and academy-only male players, would have maximized the opportunity to receive expert instruction, feedback and practice designed and implemented by experienced practitioners, leading to mastery of basic movement competencies essential for sport success (see, MacNamara, Collins, & Giblin, 2015). Given the large population base of aspiring male soccer players in the UK, allied to the proportionately small number of players that obtain professional status (0.4%, FIFA, 2007), it appears increasingly likely that early soccer engagement (including both play and practice) is a pre-requisite for future success. Although female players amassed fewer hours in soccer play compared to males, National players engaged in more play than their sub-elite peers, as outlined within the early majority engagement pathway. However, the female players were not differentiated with respect to accumulated hours in practice during childhood or adolescence, but were both considered as highly skilled.       140  Overall, these data do not support an early childhood specialization pathway at the exclusion of engagement in other sports. These data suggest multi-sport participation may play an something more akin to an ancillary role in developing soccer experts, so long as participation in other sports does not detract from majority time engagement in the primary sport of interest (see also, Hendry & Hodges, 2013). There is evidence suggesting that there may be some transfer of perceptual-cognitive skill from multi-sport involvement (Broadbent, Causer, Williams, & Ford, 2014; Causer & Ford, 2014; Roca & Williams, 2017). However, any such transfer is dependent upon the extent to which each sport form contains the same, or similar, underlying perceptual-cognitive structures as those found in soccer (e.g., invasion games; see Baker, Côté, & Abernethy, 2003; Berry, Abernethy, & Côté, 2008; Causer & Ford, 2014). While, based on my data, I would advocate towards perceptual-cognitive skill being best developed from engagement within the primary domain (i.e., soccer practice and play; see Causer & Ford, 2014; Roca, Williams, & Ford, 2012), I recognize that engagement in ancillary sports (e.g., basketball, hockey, handball) could facilitate the development of decision making, pattern recognition and anticipatory capabilities required for elite level soccer performance (e.g., Causer & Ford, 2014; Roca & Williams, 2017). In addition to transfer, multisport participation is likely to reduce the physical stress placed upon key effectors, which can have a positive effect on the development of overuse injuries (e.g., Post et al., 2017). Moreover, there may be other physical benefits of engaging in other sports such as athletics to improve speed, strength, and running mechanics or gymnastic, or martial arts, to improve general flexibility and general kinesthetic awareness. However, the degree of sport science support found within most professional academies offer highly specific programs to best meet the demands of aspiring soccer experts.      141    Further research is required to determine the specific number and nature of multisport participation which might best facilitate soccer development, or at least not interfere. This would need to be considered with respect to sports that provide optimal transfer of perceptual cognitive skill (Broadbent et al., 2014; Causer & Ford, 2014; Roca & Williams, 2017) as well as physical development of the athlete or reduce the incidence of injury, such as gymnastics and swimming. According to the findings from Study 1, there appears to be an upper limit for the total number of other sports engaged during childhood which does not interfere with later success at adult levels. For example, future professional youth and adult male players engaged in significantly fewer other sports (~4) during childhood than academy players who were not selected to play professional soccer at youth or adult levels (~5). At face value the difference across groups appears quite small, however it is possible that in professional sport, whereby the margins of success are often small, participating in an additional sport may detract from accruing the hours in soccer practice and play activities associated with success. Unfortunately, there were no real specific patterns with respect to the nature of the different types of sports engaged in by participants.  6.2.2 Discriminability of skill ratings Study 2 represented one of a few research studies assessing how well various soccer related skills (i.e., technical, tactical, physical and creative skill ratings) distinguished across future success in elite youth male soccer (such a prospective design was also used by Vaeyens et al., 2006).  In this study, I also examined relationships between skill ratings and soccer practice and play estimates. Together, these methods helped us capture the dynamic nature of talent identification and development (e.g., Collins & MacNamara, 2012), as well as the multifaceted nature of skill in soccer (e.g., Ali, 2011; Unnithan, White, Georgiou, Iga, & Drust, 2012;     142  Williams & Reilly, 2000). As expected, future youth professional players received higher coach ratings for technical, tactical and physical skills (provided at T1) than players that did not progress beyond the academy level (Academy-only). However, only ratings of tactical skill at time 1 and time 2 differentiated between the future adult professional players and the youth-professional only players (technical skill approached statistical significance). This finding underlines the importance of game intelligence developing future soccer expertise. Importantly, physical skills no longer differentiated at the professional levels (youth vs. adult).  The results of Study 2 highlighted the importance of a long-term approach to talent identification and development in which the ability of players to make and execute quick and accurate decisions should be placed at the forefront of any developmental environment. The lack of discriminability of physical skill at the professional levels also has several important practical and theoretical implications. Several researchers have shown a selection bias in youth talent development programs toward players born towards the beginning of the cut-off date for selection into an age category (e.g., players born within the first 3-6 months of the calendar year; see Helsen et al., 2012; Wattie, MacDonald, & Cobley, 2015 for reviews). Termed the relative age effect, players born earlier in the selection year may be both chronologically and biologically more mature than those players born toward the end of the age category. Being relatively older than players born in the last quartile of the year provides distinct differences in physiological capacities (e.g., Meylan et al., 2010, Mujika, et al., 2009). This can lead to a predomination of physically more dominant players being selected into representative teams at national and international level, which consequently provides more physically mature players with greater practice and development opportunities (Helsen et al., 2012; Musch & Grondin, 2001; Wattie et al., 2015). Around 50% of the players selected from their respective schoolboy academies as     143  youth-professionals at age 16 yr were born in the first quartile of the year (Jan – Mar). However, only33% of successful players at the adult-professional level were born within the first quartile of the year. Allied to the discriminability of physical skill at the youth level (i.e., discriminating Academy-only from youth-professionals), but not for the transition to adult professional level, it appears that this bias towards more physically dominant youth players exists early, but has a decreased influence at the highest (older) levels. Unfortunately, this predisposition towards selecting more physically dominant players, exacerbated by the relative age effect, continues to occur despite explicit knowledge of the causes and consequences of the relative age effect (e.g., Hill & Sotiriadou, 2016).   To ward against erroneously selecting players based upon their physical capability in youth, it would appear prudent for practitioners to focus on tactical and technical skill over the long term. A long-term focus on tactical and technical skill is supported from an emerging body of work showing something of a reversal effect in the relative age effect at the elite adult level (Cobley, Baker, Wattie, & McKenna, 2009; McCarthy, Collins, & Court, 2015). Similar “reversal effects” have been shown in the upper echelons of soccer, with players that receive higher salaries (presumably some indicator of overall skill level and performance), showing a tendency to be born relatively later in the calendar year (Ashworth & Heyndels, 2007). Award-winning athletes from team sports, including soccer, tended to be born later in the year (Ford & Williams, 2011). Although the mechanisms for the reversal effect are unclear, researchers have suggested that as compensation for their lack of physicality, relatively younger athletes are believed to develop their technical and tactical skills to a greater extent than those physically more mature (Ford & Williams, 2011; McCarthy, Collins, & Court, 2016). The advantage of this technical/tactical development being that once they have reached full maturity they have     144  developed advanced technical, and tactical skills, beyond those players that previously relied more heavily on their physicality to dominate at the younger age group levels. As such, in adulthood, where there is relative physiological parity, and the temporal demands are greater, these ‘late bloomers” are better able to cope with the demands associated with higher levels of play. Physical skill ratings generally decreased from T1 to T2 suggesting that any physical advantages of more physically mature players, (as indicated by higher physical skill ratings and a greater proportion of players born in the first quartile of the selection year), may be attenuated as a function of parity in biological maturity and/or through adaptations to training. Similar declines, however, were also evident in coach ratings of technical and tactical skills. These declines do not suggest that players become less skilled with age. Ratings were made in comparison to others within their individual teams, indicating that the professional group at T2 was perhaps more homogenous than the original sample at T1. Soccer skill is multidimensional, requiring proficiencies in technical, tactical, physical, and creative skills (see Meylan et al., 2010; Williams & Reilly, 2000). While deficiencies in one skill area may be compensated for by strengths in others (termed the “compensation phenomenon,” Bartmus, Neumann, & de Marées, 1987), a minimum standard of proficiency in technical, tactical and physical skill would be required to meet the temporal and cognitive demands of professional soccer (e.g., Vaeyens et al., 2006). Aspiring players must possess a relatively high level of physical fitness to cope with the physiological demands of match-play. They must be able to select and perform the correct technical actions at the correct time under significant temporal and spatial constraints to maintain basic possession of the ball.      145   6.2.3 Development of skill  In Study 2, as a secondary question, I examined the relationships between developmental soccer activities and various soccer-related skills. Across all skill dimensions, these relationships emerged over a long-time period. There were few significant correlations between childhood activities and coach skill ratings provided at T1. In fact, only skill ratings of the youth professional players (collected at T2) showed significant relationships with practice-related variables. Within the group of youth-professionals, T2 ratings of technical, tactical and creative skills were generally positively related to hours accumulated in soccer specific practice engaged in across the full developmental period (i.e., childhood and adolescence). These results support ideas that highly specific practice in the primary domain is an important precursor for the development of expertise (Ford, et al., 2015). Contrary to expectations, childhood and career soccer play yielded few significant, or even moderately sized, correlations with coach ratings of skill. Given the centrality of play to the early engagement hypothesis (Ford et al., 2009) and results from Study 1, in which career play differentiated those players that made it to the adult professional level versus those that did not, we expected to see relationships with play. One possibility is that soccer play does little more than add to the overall volume of soccer activity. Very little is known about the specific nature of play, at least in terms of the underlying learning principles that are involved within unstructured practice. Yet there tends to be an assumption within the literature that play offers optimal practice conditions primarily through increased practice variability, opportunities for decision making and autonomy (see Côté & Erickson, 2015). However, we do not have the data to support these claims. From these current data it appears that soccer practice provides the most beneficial, or at least a more efficient learning environment in comparison to unstructured play     146  since practice was generally most associated with the development of technical and tactical skill. While we are not discounting that play benefits skill development, based upon these data, we hypothesize that the lack of external feedback, instruction, practice design and focus may lead to diminished returns compared to more structured practice conditions whereby these conditions are present (Hendry & Hodges, 2013; MacNamara, Collins, & Giblin, 2015).    One skill dimension we did expect to see positively related to soccer play was creativity. Several authors have suggested that deliberate play activities afford more opportunities to experiment with new ideas, movements and techniques, more than structured deliberate practice activities (Bowers, Green, Hemme, & Chalip, 2014; Côté et al., 2012; Memmert, 2015; Memmert, Baker, & Bertsch, 2010; Santos, Memmert, Sampaio, & Leite, 2016). Contrary to expectations, no relationships between soccer play and creative skill were present. Rather, ratings of creative skill were generally associated with soccer practice amounts. This latter finding is in keeping with Ericsson and Lehman’s (Ericsson & Lehman, 1999) conclusion that creativity is developed upon a basis of mastery, which is primarily developed via deliberate practice.  Overall, the data in this paper point towards the development of soccer skills primarily through involvement in high quality soccer practice activities, particularly for technical, tactical and creative skills. The quality of the practice experience is likely to contribute to the effect, considering that professional players were rated as being more skilled, were recruited into an academy earlier and thus had likely amassed more hours in high quality practice than less successful players. Alone this finding may lead practitioners towards “a race to the bottom” involving selection of increasingly younger players into professional academies as a means of maximizing the development of skill. However, any such considerations should be met with     147  caution, due in part to the proposed negative psycho-social outcomes associated with early specialization (see Côté & Erickson, 2015). Yet, the high level of competition, and large volumes of players competing for places within professional soccer, dictate that those players not engaging in high volumes of soccer activity are at a distinct disadvantage in comparison to those amassing more soccer activity hours. Finding an optimal balance between overall positive youth development and expertise should be the end goal for researchers and practitioners alike. However, the unfortunate reality is that for many involved in elite soccer development whose livelihood depends on producing first team players (e.g., Academy Directors), the latter often takes precedence over the former.  6.2.4 Motivation Involvement in high volumes of deliberate play during childhood is postulated to foster the development of intrinsic and self-determined motivation (Côté & Erickson, 2015; Côté & Hancock, 2014; Côté et al., 2009, 2012). I failed to show support for this postulate, at least within a developmentally elite sample of soccer players (see Study 3 and also Hendry et al., 2014). I have not shown evidence of a relationship between hours in childhood soccer play and measures of self-determined motivation (SDM: see Ryan & Deci, 2017). However, in Study 3, a shift was observed towards less self-determined forms of motivation as a function of age among the elite level players, which were on the cusp of receiving a professional contract. Furthermore, a significant negative correlation was shown between recent practice estimates and indices of SDM (in an earlier study, not part of this thesis, I also showed negative relationships between practice encountered within an elite academy setting and indices of SDM, Hendry et al., 2014).  According to Côté and colleagues (Côté & Erickson, 2015; Côté et al., 2009, 2012), deliberate play activities are thought to evoke in participants a sense of autonomy over the game     148  conditions. That is, participants have a sense of agency over the what, where, when and with whom they will engage in deliberate play. As a basic psychological need (along with competence and relatedness), this sense of autonomy over deliberate play conditions is believed to foster intrinsic motivation and more self-determined forms of motivation. Furthermore, deliberate play is believed to be highly enjoyable and participated in for inherently rewarding motives (e.g., Côté et al., 2012). This stands in contrast to deliberate practice activities, which are portrayed as more controlling, less enjoyable developmental activities (Côté & Hancock, 2014). However, this comparison appears to be overly simplistic and represents deliberate practice activities, such as those encountered with soccer practice, as a generally constraining and demotivating experience. Rather than practice activities, this comparison reflects a poor coaching environment. Moreover, the idea that deliberate practice activities are always not inherently enjoyable is false. In early conceptualizations of deliberate practice, Ericsson and colleagues (Ericsson et al., 1993) suggested that deliberate practice may not always be inherently enjoyable. However, this did not necessarily mean that a condition of deliberate practice is that is not enjoyable. In Ericsson et al.’s original work, practice activities were rated quite high for enjoyment, just not as high as other recreational-type activities. Several researchers have shown that deliberate practice activities can be enjoyable but perceptions of practice enjoyment can be highly variable, often changing on a day to day basis (Hodges, Kerr, Starkes, Weir, & Nananidou, 2004). Perceptions of enjoyment derived from deliberate practice may shift over time from enjoyment in the activity itself to enjoyment received from the benefits of practice (i.e., improved performance; Ward & Williams, 2003; for a review, see Hodges & Baker, 2013).  Overall, the results from Study 3 indicated that shifts towards less self-determined forms of motivation in older players, as exhibited by increased controlled motivation, occurred as a     149  function of age and skill. Although motivation tended to become less self-determined in older elite players, SDM scores within older (U17) elite players only dropped to a level commensurate with age matched, non-elite, yet competitive players. Thus, despite engaging in vastly greater quantities of time in highly structured soccer practice activities, scores of SDM were not negatively affected beyond the control group.  While the causative factors for the decline in self-determined forms of motivation associated with age (U15 to U17) and skill (i.e., only in the elite athletes) are speculative, I suggest several possibilities. First, introducing external rewards to previously intrinsically motivating behaviors has been shown to have an undermining effect on SDM. Termed “the undermining effect” (Deci et al., 2001; Deci, Ryan, & Koestner, 1999), the potential rewards associated with achieving professional status such as increased wealth, status, and fame are vast and may contribute towards this decline. Second, an increase in controlled motivation may actually benefit those older elite players. By its very nature, professional sport is competitive and likely requires a determination driven, not only by an innate desire for self-improvement (e.g., Deci & Ryan, 2002) but also by a motivation to achieve and exert superiority over others (Cook, Crust, Littlewood, Nesti, & Allen-Collinson, 2014). The shift towards more controlled motivation is likely reflective of this innate desire and the associated competition for contracts within the older cohort of players. This type of co-existing motivational profile, appears to be consistent within high performing athletes and has been associated with increased persistence, interest, satisfaction and mental toughness (Briere et al., 1995; Cook et al., 2014; Pelletier et al., 2001). Many of these characteristics are viewed essential to cope with the often challenging realities associated with professional sport (Collins & MacNamara, 2012; Hardy et al., 2016; MacNamara & Collins, 2010).      150  6.2.5 Developmental activity challenge Using the challenge point framework (Guadagnoli & Lee, 2004) as a theoretical backdrop, women National team players engaged in more challenging play as children than sub-elite, women Varsity players. Further, challenge ratings during childhood showed further skill-group discriminability both with respect to activity type (e.g., play, practice and competition) and age-period (e.g., childhood & adolescence). The fact that these challenge ratings discriminated across groups of highly skilled athletes provides support for the validity of these methods.  Competition was viewed as the most challenging developmental activity by both groups of elite players. Although Ericsson and colleagues (1993) used competition performance as a variable for comparing success; it was not thought to positively contribute to the development of expertise. Their categorization of competition as a work activity that it is time constrained, motivated by external rewards, lacking repeatability and opportunities for experimentation is not universally shared. Several researchers have proposed that experience of competition is a vital component of the talent development process (Abernethy, Farrow, & Berry, 2003; Singer & Janelle, 1999). In youth development circles, competition is often viewed as an extension of the practice session with the experience of playing against unfamiliar opposition, psychological preparation, and dealing with travel, fluctuations in score and crowds thought to facilitate the development of future experts (Cook et al., 2014; Holt & Dunn, 2004). Further, referring to competition as a work activity is a distinction that may not translate from music to sport, since competition is generally rated as a highly enjoyable and relevant activity in sports (Ward, Hodges, Starkes, & Williams, 2007). There are also anomalies  between the soccer expertise research and the “real world” of youth development with respect to the relative importance of competition (e.g., Ford et al., 2009; Ward et al., 2007). Across skill classes, there tends to be no     151  significant difference in the total number of hours that prospective talents engage in competitive match play during development. However, the total number and length of each game, during childhood and adolescence are often externally controlled by a league organization. Consequently, it is not surprising that differences across a skill level for time spent in competition are small.  A “best v best” approach to games is continuously championed by coaches as being central to the development of young players. The premise being that players are repeatedly placed into optimally challenging environments that equally tax the player’s technical, tactical, physical and psychological capacities, forcing them to adapt and improve. Therefore, like practice and play, it is highly unlikely that all competition is equal.  Analysis of the degree of relative challenge during competition may inform us of the quality of developmental experiences that define elite players.  Given the centrality of soccer play to the development of soccer expertise generally we were hopeful that challenge would facilitate our understanding of soccer play’s importance, beyond merely contributing to more hours in soccer activity generally. In our previous work, we have suggested that play and practice should not be viewed as dichotomous constructs, as they can often share similar perceptual-motor conditions. Using the constraints approach to skill acquisition (e.g., Chow, Davids, Renshaw, & Button, 2013; Davids, Button, & Bennet, 2008) as a theoretical backdrop, the combination of individual, task and environments can be manipulated by the coach, or by peers, to produce conditions of optimal challenge, and even transfer (see Chow et al., 2013; Hendry, Ford, Williams, & Hodges, 2015). For example, the differences between participation in a 3v3 small sided game (e.g., task) during soccer practice versus street soccer are likely minimal, when the skill level of the performer is relatively matched with other     152  participants (e.g., individual), the playing area is small, and the overall psychosocial environment is competitive and supportive (e.g., environment).  In Study 3, when we questioned elite women players about childhood soccer, play was generally viewed as the least challenging developmental activity and rated as significantly less than challenging than coach-led practice. However, in comparing across groups, National players viewed soccer play as more challenging in comparison to Varsity player. In this sense, it is possible that the soccer play engaged in by the elite, National players was closer to the degree of perceived challenged evoked in practice, albeit there was no Skill group X Activity interaction. It appears that within this cohort, the specific conditions of soccer play (or interactions between the constraining variables) were sub-optimal. This may not be the case in other cultures and contexts, whereby street soccer is considered the norm and an integral component of development (e.g., Araújo et al., 2010; Ford et al., 2012; Koslowsky & Botelho, 2010). It may be that the specific intention is the key to interpretations of challenge (e.g., Ericsson et al., 1993). Unfortunately, we were unable to examine the extent to which intention influenced challenge, due to low response rate amongst all participants for the total number of hours in individual practice activities (i.e., self-led practice activities conducted with the specific intention of improving performance), which could then be compared to self-led play.  When reflecting on childhood development, practice was generally viewed as being sub-optimally challenging for both the National and Varsity women groups. This was unexpected, both in relation to the overall low rating of practice challenge, but also in relation to the lack of discriminability of practice across the two skill groups. We expected National players to have engaged in more challenging practice throughout their development. The generally low ratings may point to a general lack of quality coaching for the female players, particularly during     153  childhood, whereby practice was viewed as less challenging than adolescence. Alternatively, the lack of perceived challenge may be viewed as preliminary evidence for early signs of precocity within elite level players. Although not presented in this thesis, both National and Varsity players rated themselves as being within the top 10% of players within their respective teams throughout childhood and adolescence. This perception of competence continued to persist in adulthood, suggesting an inflated sense of one’s own ability. In sum, the use of subjective ratings of perceived challenge provided promising early findings which may advance the existing expertise literature beyond counting practice hours. Further research is required to explore the potential mechanisms responsible for challenge, but the discriminability of these data across two elite samples provides a measure of validity upon which to base future research.  6.3 Practical implications  The research findings from my thesis have numerous implications for both practitioners and academics involved in the study of expertise. First, the data from this thesis supports the early (majority) engagement pathway as being the most beneficial to those aspiring to become experts in adulthood. At an applied level, this has implications for players, parents, coaches and national governing bodies as they weigh up the consequences of their talent development decisions and infrastructure. I also present data supporting the idea that youth development practitioners allow time for the “fruits of their labors” to emerge, such that rather than identifying and selecting “talent” early (i.e, U13-U15), they focus on the long-term development of tactical and technical actions in player selection.  6.3.1 Benefits of early majority engagement Based upon the outlined research findings, we would advocate for aspiring experts, in large participation sports such as soccer, follow an early majority engagement pathway. This     154  involves high volumes of domain specific practice and play from an early age. This may or may not be accompanied by other sports, as long as the majority of time in childhood is spent in the “primary” sport. Engagement in other sports should be in addition to soccer activity, and not at its expense.  Elite players following the early majority engagement pathway are likely receiving behavioral and psychosocial benefits aligned within the deliberate practice framework, early engagement hypothesis and early diversification pathway based on play and practice. By engaging in relatively high volumes of soccer specific practice, young players selected into an elite academy from an early age are gaining access to the best available coaching, facilities, competition, equipment, sport science and medical support. In this regard, they are receiving all of the benefits of high quality instruction, feedback and practice conditions which are associated with enhanced levels of skill (see Study 2, also MacNamara, Collins, & Giblin, 2015) which likely meet the conditions of deliberate practice. In considering these distinct learning advantages associated within involvement in an academy, it becomes difficult to reconcile how such favorable conditions could not be considered advantageous over developmental activities without such clear bases of support, at least in sports where the participation base is high, and peak attainment is not required until adulthood. However, some semblance of balance is needed to avoid the pitfalls associated with specialization such as burnout and injury (e.g., Côté et al., 2009; Post et al., 2017).  The overall contribution of soccer play was highlighted across both sexes. The degree of involvement in soccer play differentiated elite from sub-elite players in the male youth samples (i.e., adult-professional from youth professional, Study 1 and Nationalfrom Varsity women’s team during childhood, Study 4). Unfortunately, based on correlational data (Study 2 and 3),     155  there were no relationships between soccer play and either coach ratings of skill or indices of self-determined motivation. Thus, we were unable to establish either skill related, or motivational benefits associated with involvement in high volumes of soccer play during childhood. As such, based on these data I am unable to comment upon the potential mechanisms underpinning the importance of soccer play, beyond contributing to more hours in soccer activity generally. Even if soccer play does little beyond extending soccer contact time, it plays an important role in the development of soccer experts. As such, avenues for soccer play, whether in formal or informal settings should be maximized. This would be particularly important for female players, where play estimates were small in comparison to male players. Because soccer play has been traditional associated with masculinity (e.g., Clark & Paechter, 2007; Williams, 2007), soccer policy makers may wish to provide positive messaging and opportunity for female only play during recess or after school initiatives. Across both sexes, youth soccer clubs may wish to rearrange the time/location of their practice sessions to allow youth players’ greater opportunity to engage in soccer play outside their regular soccer practice. During personal communication with players from Study 1, they reported spending ~2 hours on a mini-bus to get to and from practice sessions. Perhaps, this time would be better served playing street soccer, at least part of the time and trading off some of the “quality” coaching or finding less “elite” coaching at a more convenient location. The finding that the elite male and female soccer players participated in several other sports during childhood could be associated with other developmental benefits. One of the advantages of multisport engagement could be potential transfer of perceptual-cognitive processes underlying decision making and anticipation from engaging in sports with similar     156  underlying structures to soccer (e.g., Causer & Ford, 2014; Roca & Williams, 2017). Notably, no female players and less than 5 % of the male sample from Study 1 specialized solely in soccer. Thus, the idea of specialization, as originally conceptualized within the DMSP, may not be beneficial to developing soccer expertise. The lack of specialization within participants may explain why there were not any negative relationships between soccer practice estimates and self-determined motivation (Study 3). At a practical level, the early majority engagement pathway has potential implications for sport development and national governing bodies. Soccer clubs/associations would be advised to develop partnerships with several other sports organization to provide multisport opportunities that could complement their existing soccer programing. Not only could this positively influence soccer expertise, but it may also provide sport transfer opportunities for those players that fail to make the grade in professional soccer (e.g., Bullock et al., 2009).  6.3.2 Evaluation of the DMSP The DMSP and its associated postulates have provided a framework of expertise that has instigated a large volume of research (for a citation analysis see, Bruner, Erickson, Wilson, & Côté, 2010). In its current dual pathway form, this model is not directly applicable to soccer. As arguably the most popular sport in the world, this questions the overall practical and theoretical utility of the model. Based upon multiple research sources, it appears that early specialization as defined with the DMSP, involving high volumes of deliberate practice from a young age in a single sport, is not applicable to elite soccer contexts (Study 1, Study 4; Ford et al., 2012, 2009; Haugaasen & Jordet, 2012). Other than early maturation sports, whereby the peak performance years take place during early adolescence, such as gymnastics or ice skating (Law, Côté, & Ericsson, 2008), early specialization is not necessary nor advisable. The early (majority)     157  engagement pathway offers something of a hybrid of both the early specialization and diversification pathways outlined within the DMSP. In this regard, aspiring soccer experts receive the benefits of both specialization, high levels of practice and play (i.e., diversity within a sport) and potential benefits from participation in other sports. From a theoretical standpoint, perhaps the DMSP should be modified to look at specialization and diversification along a continuum, each with related outcomes rather than the dichotomous model currently presented. In recent developments of the DMSP, diversity has been considered both as a between sport measure as well as a within sport concept, such that an early “diversification” pathway could actually be a specialization pathway, but with time spent in different variations of the same sport (i.e., play, co-rec soccer, practice, competition, futsal; Côté & Erickson, 2015). Although this is an attempt to include the early engagement hypothesis, the terminology and lack of clear messaging to “users” of sports is potentially concerning. It is beyond the scope of this thesis to provide accurate distinct recommendations as to specific ratios of primary to ancillary sport involvement. However, in future research, there may be value in probing this question. Despite variations in play between elite youth male players in the UK and Canadian female players, the general pattern of results pointing towards an early majority engagement pathway was consistent across cultures and context, at least within a soccer framework. Further testing of the early engagement hypothesis is needed to establish its external validity for other large participation sports, such as basketball in the United States of America or rugby in New Zealand or South Africa.  6.3.3 Talent identification  Overall, the results from Study 2 highlight the importance of a long-term approach to talent identification and development in which selection criteria towards adult professional status     158  is focused primarily on tactical and technical skill. The fact that these skill related factors discriminated within a nested sample of highly skilled male youth athletes, already pre-selected on the basis of skill from recreational athletes, mean that any results carry significant discriminatory weight (and may also explain why effect sizes were expected to be low). A long term approach to talent identification requires youth coaches to look beyond players’ current physical capabilities, even though we had evidence that physically stronger/fitter and “older” in terms of birth month players were predominantly being selected at the U16 year, elite age groups (see also, Meylan et al., 2010; Mujika et al., 2009). At the youth professional transition, physical skill differentiated successful versus unsuccessful players, but it did not at the adult professional level. Similarly, indicative of a relative age effect 50% of youth professional players selected were born in the first quartile of the year. However, this percentage dropped to only 33% of adult professionals. Knowledge of the advantages of the relative age effect has permeated from academia into the applied world of youth soccer, yet despite this explicit coach knowledge it appears that the general pattern of selection persists (Helsen et al., 2012; Hill & Sotiriadou, 2016). In the absence of data from women’s soccer is it unclear if there would be any discrepancy between male and female soccer players. However, it would appear unlikely that the importance of technique and game intelligence would be any less prominent within the women’s game.   6.4 Limitations and future recommendations 6.4.1 Retrospective recall techniques Arguably the largest criticism of the sport expertise literature in general surrounds the use of retrospective recall methods to capture developmental activity hours. Despite being prone to bias and error (Hodges, Huys, & Starkes, 2007) this method remains the best available method     159  for assessing engagement in developmental activities across the lifespan (Hopwood, 2015). By using a prospective design with our study of elite male players we were able to attain estimates that were more closely related to the assessed time-period than those used with adult athletes. With the talent identification and development process starting at increasingly younger age groups (Côté et al., 2011) future research may wish focus upon a cohort players as they are first identified by their respective academies and follow-up using a longitudinal design to measure which variables are associated with future success in soccer. In the absence of this longitudinal assessment, various precautions were taken to ensure the reliability and validity of the research. An adapted version of the practice history questionnaire (PHQ; Ford, Low, McRobert, & Williams, 2010) was used which has been used in several research publications to attain estimates of hours spent in developmental activities during childhood and adolescence (e.g., Ford et al., 2012; Ford & Williams, 2012; Roca et al., 2012). This method was based on the original methods used by Ericsson et al. (1993) in music and adapted versions to aid recall in sport-related studied (e.g., Hodges & Starkes, 1996; Helsen et al., 1998; Hodges et al., 2012). To aid validity of recall and to prevent inflation bias, we measured the most recent years’ activity first before working backwards towards initial engagement in sports for the adult athletes. For the first studies with boys, measures of reliability were attained by collecting estimates of time in soccer practice and play from a sample of parents and coaches. Due to a lack of access to the parents of the adult women players, these data were not available for Study 4. However, across both cohorts, player-player reliability estimates were collected from two different components of the questionnaire. Thus, we were able to ascertain the extent to which estimates were reliable at the within participant level across all studies. While the use of the PHQ (Ford et al., 2010) and adoption of recommendations made by Hopwood (2015) have enhanced the reliability of these     160  methods of collecting developmental activity hours, there are validity issues that persist. In the future, it will be important to determine better methods for enhancing the validity of these procedures. One method might be to make comparisons of means (and variability) associated with age group data collected in youth athletes in real-time versus those provided retrospectively by older athletes. Comparisons of cross-sectional and retrospective data for various age groups would help to determine how much variance should be deemed acceptable. As with some of the earlier work in this area, developmental activity diaries from a subset of the athletes would also help corroborate the validity of these estimates.  6.4.2 Subjective testing Subjectively assessed ratings of skill were attained in Study 2, as opposed to more objective methods of variables related to fitness, tactical awareness or technique (e.g., Ajmol Ali et al., 2007; Huijgen, Elferink-Gemser, Ali, & Visscher, 2013; Memmert et al., 2010; Meylan et al., 2010; Roca, Williams, & Ford, 2012b). The lack of relationships between soccer activities with coach ratings of skill at time 1 may be attributed to reliability issues associated with subjective skill ratings. However, the coach ratings across the two-time points and the player-coach ratings did show satisfactory levels of reliability and consistency. The subjectivity of these data may reflect a more ecologically valid method for assessment of skills and making conclusions about factors related to continued participation in an elite youth development setting. Coaches and directors of football are generally regarded as the gatekeepers to professional status. Decisions as to which players’ progress versus those that do not are generally based upon the coaches’ “eye for talent”, which can be used to assess the multifaceted nature of expertise in elite soccer (Meylan et al., 2010; Unnithan et al., 2012; Williams & Reilly, 2000). Just because a test is more objective, it does not make it a more valid test (e.g., reaction time, cognitive recall test),     161  nor does it make it more discriminatory. Moreover, more refined measures of skill are often lacking in their ability to fully capture the multidimensional nature of skill in soccer (Ali, 2011; Unnithan et al., 2012). For instance, technical skills’ tests may not incorporate the key perceptual-cognitive cognitive components that are required to make accurate and relevant passes. A mixed methods approach might be advisable in future research, which would provide a basis of comparison across both objective and subjective measures of skill. As recommended in the discussion of Study 2, it would be beneficial to use multiple item, versus single-item scales to measure different components of the various skills, similar to methods used in the Tactical Skills Inventory for Sports (TACSIS; Elferink-Gemser et al., 2004). In addition to potentially providing a more valid measure of tactical skill, it may enhance the possibilities of detecting relationships between different facets of skill and developmental soccer activities. Similar arguments could be made for assessing multiple components of physical skill (e.g., Meylan et al., 2010) or tactical creativity (e.g., Memmert et al., 2015).  6.4.3 Sample size One of the strengths of this series of studies was the overall “eliteness” of the participant sample. However, elite participants, by their very definition are a statistically rare group. As such the sample size may not necessarily meet criteria for “adequate” power (Cohen, 1988). This had implications on the types of analyses, since we lacked participants to conduct multi-level modeling based on regression techniques (for Study 1-3). The prospective, follow-up method used in Study 1 provided opportunity to obtain data from players that had been released at T1 (academy-only) and T2 (youth professional-only). Resultantly, data was collected from participants that are typically excluded from expertise research. Further, across Studies 1-3, the overall number of elite youth participants involved across each study was relatively large (n =     162  102) adding to the overall generalizability of the elite, male data. The focus on expertise meant that we primarily tested the hypotheses forwarded by Côté and colleagues (Côté & Erickson, 2015; Côté & Hancock, 2014; Côté et al., 2009, 2012) within an elite sport context, albeit we did include a sample of recreational Canadian male soccer players. It is possible that some of the postulated relationships between play and positive outcomes may be more prevalent within a recreational sample. However, since the DMSP encompasses the development of expertise, as well as long-term participation in sport, any postulate based upon the DMSP should stand up to scrutiny within an elite context. Where we did recruit age matched recreational players (i.e., Study 3), the general pattern of results was similar in that play amounts were not significantly not significantly correlated with self-determined motivations. However, there were negative correlations between practice and autonomous forms of motivation.   6.4.4 Cause or consequence of expertise  Early recruitment into a professional soccer Academy appears to be an important factor in the long term development of future male soccer experts in the UK (Le Gall, Carling, Williams, & Reilly, 2010; Meylan, Cronin, Oliver, & Hughes, 2010; Zibung & Conzelmann, 2013). Indeed, those male players that progressed to elite professional youth status at age 16 yr entered the academy system and joined their current academy at an earlier that those players that did not successfully transition (Study 1). While there were no differences among professional players with respect to these two soccer milestones, this early entry serves to raise the question as whether future success was a cause or a consequence of early involvement in an elite soccer academy. That is, did these players exhibit a level of soccer precocity that would have resulted in future sporting success or was the talent development system integral in causing the development of “talent”? Of course, the interaction between the two is also likely to be a reason for success,     163  whereby early-entry reflects initial precocity, but quality developmental experiences are needed to develop talent and skills. Future longitudinal research is needed to explore the initial skill level of participants as they enter the academy system to identify the overall efficacy of the academy system and to track learning/improvement over time. This type of research would afford opportunities to assess the antecedents of adaptability to an elite soccer environment over time (e.g., basic physical and perceptual competencies).  6.5 Conclusion Through this thesis, I have provided support for an early majority engagement pathway, consisting of primary engagement in soccer related activities in comparison to either a specialized or diversified pathway, as outlined within the DMSP. This early engagement pathway was consistent across male and female soccer experts across two continents. Together, these data question the overall utility of the DMSP in its current dual stream framework. Studies, 2 and 3 also failed to provide support for postulates made by Côté and colleagues outlining the skill development and motivational benefits from engaging in high volumes of deliberate play during childhood. In addition, the results from Study 3 highlighted the importance for practitioners to “decide slowly” when selecting players that progress through the soccer development pathway, potentially allowing physiological benefits associated with the relative age effect to level out.  In selecting future talents, emphasis should be placed on the technical and tactical capabilities to execute the correct decision at the correct time, so long as a basic physical capability exists. The attempt to better assess activity quality through the challenge point framework shows early signs of promise (Study 4). A finer grained of analysis of the specific nature of challenge in general, as well as of “optimally challenging conditions” which     164  might elicit specific skill related outcomes is important and offers some impetus for future research.        165  References  Abernethy, B., Farrow, D., & Berry, J. (2003). Constraints and issues in the development of a general theory: A critique of the deliberate practice framework. of expert perceptual-motor performance. In J.L. Starkes & K.A. Ericsson (Eds.), Expert performance in sports: Advances in research on sport expertise. (1st ed., pp. 349–369). Champaign: IL: Human Kinetics. Adie, J. W., Duda, J. L., & Ntoumanis, N. (2012). Perceived coach-autonomy support, basic need satisfaction and the well- and ill-being of elite youth soccer players: A longitudinal investigation. Psychology of Sport and Exercise, 13(1), 51–59. http://doi.org/10.1016/j.psychsport.2011.07.008 Ali, A. (2011). Measuring soccer skill performance: A review. Scandinavian Journal of Medicine and Science in Sports, 21(2), 170–183. http://doi.org/10.1111/j.1600-0838.2010.01256.x Ali, A., Williams, C., Hulse, M., Strudwick, A., Reddin, J., Howarth, L., … McGregor, S. (2007). Reliability and validity of two tests of soccer skill. Journal of Sports Sciences, 25(13), 1461–1470. http://doi.org/10.1080/02640410601150470 Araújo, D., Fonseca, C., Davids, K., Garganta, J., Volossovitch, A., Brandão, R., & Krebs, R. (2010). The Role of Ecological Constraints on Expertise Development. Talent Development and Excellence, 2(2), 165–179. Araújo, D., Travassos, B., & Vilar, L. (2010). Tactical skills are not verbal skills: A comment on Kennekens and colleagues. Perceptual and Motor Skills, 110(3), 1086–1088. Retrieved     166  from http://www.amsciepub.com/doi/abs/10.2466/pms.110.C.1086-1088 Ashworth, J., & Heyndels, B. (2007). Selection Bias and Peer Effects in Team Sports: The Effect of Age Grouping on Earnings of German Soccer Players. Journal of Sports Economics, 8(4), 355–377. http://doi.org/10.1177/1527002506287695 Atkinson, G., & Nevill, A. (1998a). Statistical Methods for Assssing Measurement Error (Reliability) in Variables Relevant to Sports Medicine. Sports Medicine, 26(4), 217–238. Atkinson, G., & Nevill, A. M. (1998b). Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Medicine (Auckland, N.Z.), 26(4), 217–38. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9820922 Baker, J., & Côté, J. (2003). Resources and commitment as critical factors in the development of “Gifted” athletes. High Ability Studies, 14(2), 139–140. http://doi.org/10.1080/1359813032000163816 Baker, J., Côté, J., & Abernethy, B. (2003). Sport-specific practice and the development of expert decision-making in team ball sports. Journal of Applied Sport Psychology, 15(1), 12–25. http://doi.org/10.1080/10413200305400 Baker, J., & Hodges, N. J. (2011). Expertise: The goal of performance development. In D. J. Collins, A. Abbott, & H. Richards (Eds.), Performance psychology: a practitioner’s guide (pp. 31–46). Elsevier Health Sciences. Baker, J., & Horton, S. (2004). A review of primary and secondary influences on sport expertise. High Ability Studies, 15(2), 211–228. http://doi.org/10.1080/1359813042000314781 Baker, J., Wattie, N., & Schorer, J. (2015). Defining expetise: A taxonomy for researchers in     167  skill acquisition and expertise. In J. Baker & D. Farrow (Eds.), Routledge handbook of sports expertise (pp. 145–155). New York, NY. Baker, J., & Young, B. (2014). 20 years later: Deliberate practice and the development of dxpertise in sport. International Review of Sport and Exercise Psychology, 7(1), 135–137. http://doi.org/10.1080/1750984X.2014.896024 Barkoukis, V., Taylor, I., Chanal, J., & Ntoumanis, N. (2014). The relation between student motivation and student grades in physical education: A 3-year investigation. Scandinavian Journal of Medicine & Science in Sports, 24(5), e406–e414. http://doi.org/10.1111/sms.12174 Barreiros, A., Côté, J., & Fonseca, A. M. (2014). From early to adult sport success: analysing athletes’ progression in national squads. European Journal of Sport Science, 14 Suppl 1, S178-82. http://doi.org/10.1080/17461391.2012.671368 Bartholomew, K. J., Ntoumanis, N., & Thøgersen-Ntoumani, C. (2010). The controlling interpersonal style in a coaching context: development and initial validation of a psychometric scale. Journal of Sport & Exercise Psychology, 32(2), 193–216. Bartmus, U., Neumann, E., & de Marées, H. (1987). The talent problem in sports. International Journal of Sports Medicine, 8, 415–416. Berry, J., Abernethy, B., & Côté, J. (2008). The contribution of structured activity and deliberate play to the development of expert perceptual and decision-making skill. Journal of Sport & Exercise Psychology, 30(8), 685–708. Retrieved from http://hub.hku.hk/handle/10722/60471     168  Boccia, G., Rainoldi, A., & Brustio, P. R. (2017). Relative age effect in males, but not females, undergraduate students of sport science. Sport Sciences for Health, 13(2), 349–353. http://doi.org/10.1007/s11332-017-0364-7 Bowers, M. T., Green, B. C., Hemme, F., & Chalip, L. (2014). Assessing the Relationship Between Youth Sport Participation Settings and Creativity in Adulthood. Creativity Research Journal, 26(3), 314–327. http://doi.org/10.1080/10400419.2014.929420 Briere, N. M., Vallerand, R. J., Blais, M. R., & Pelletier, L. G. (1995). Development and validation of the EMS for the measurement of intrinsic motivation, extrinsic motivation and non-motivation in sports - Echelle de Motivation dans les Sports. International Journal of Sport Psychology, 26(4), 465–489. Broadbent, D. P., Causer, J., Williams, A. M., & Ford, P. R. (2014). Perceptual-cognitive skill training and its transfer to expert performance in the field: Future research directions. European Journal of Sport Science, 1–10. http://doi.org/10.1080/17461391.2014.957727 Bruner, M. W., Erickson, K., Wilson, B., & Côté, J. (2010). An appraisal of athlete development models through citation network analysis. Psychology of Sport and Exercise, 11(2), 133–139. http://doi.org/10.1016/j.psychsport.2009.05.008 Bullock, N., Gulbin, J. P., Martin, D. T., Ross, A., Holland, T., & Marino, F. (2009). Talent identification and deliberate programming in skeleton: Ice novice to Winter Olympian in 14 months. Journal of Sports Sciences, 27(4), 397–404. http://doi.org/10.1080/02640410802549751 Bulmer, M. G. (1979). Principles of Statistics. New York: Dover.     169  Causer, J., & Ford, P. R. (2014). “Decisions, decisions, decisions”: transfer and specificity of decision-making skill between sports. Cognitive Processing, (2004), 1–5. http://doi.org/10.1007/s10339-014-0598-0 Chow, J. Y., Davids, K. W., Renshaw, I., & Button, C. (2013). The Acquisition of Movement Skill in Children through Nonlinear Pedagogy. In J. Côté & R. Lidor (Eds.), Conditions of children’s talent development in sport (pp. 41–60). Morgantown, WV: Fitness Information Technology. Clark, S., & Paechter, C. (2007). “Why can”t girls play football?’ Gender dynamics and the playground. Sport, Education and Society, 12(3), 261–276. http://doi.org/10.1080/13573320701464085 Cobley, S., Baker, J., Wattie, N., & McKenna, J. (2009). Annual age-grouping and athlete development: a meta-analytical review of relative age effects in sport. Sports Medicine (Auckland, N.Z.), 39(3), 235–56. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/19290678 Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Statistical Power Analysis for the Behavioral Sciences. http://doi.org/10.1234/12345678 Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155–159. http://doi.org/10.1037/0033-2909.112.1.155 Collins, D., & MacNamara, Á. (2012). The Rocky Road to the Top. Sports Medicine, 42(11), 907–914. http://doi.org/10.1007/BF03262302 Cook, C., Crust, L., Littlewood, M., Nesti, M., & Allen-Collinson, J. (2014). “What it takes”:     170  perceptions of mental toughness and its development in an English Premier League soccer academy. Qualitative Research in Sport, Exercise and Health, 6(September 2014), 329–347. http://doi.org/10.1080/2159676X.2013.857708 Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78(1), 98–104. http://doi.org/10.1037/0021-9010.78.1.98 Côté, J. (1999). The influence of the family in the development of talent in sport. Sport Psychologist, 13(4), 395–417. Retrieved from https://www.researchgate.net/publication/232566000_The_influence_of_the_family_in_the_development_of_talent_in_sport Côté, J., Baker, J., & Abernethy, B. (2007). Play and practice in the development of sports expertise. In G. Eklund & R. Tenenbaum (Eds.), Handbook of sport psychology (3rd ed., pp. 184–202). New York: Wiley. Côté, J., Coakley, C., & Bruner, M. W. (2011). Childrens’ talent development in sport: Effectiveness or efficiency? In S. Dagkas & K. Armour (Eds.), Inclusion and exclusion through youth sport (pp. 172–185). London: Routledge. Côté, J., & Erickson, K. (2015). Diversification and deliberate play during the sampling years. In J. Baker & D. Farrow (Eds.), Routledge handbook of sports expertise (pp. 305–316). London: Routledge. Côté, J., Ericsson, K. A., & Law, M. P. (2005). Tracing the development of athletes using retrospective interview methods: A proposed interview and validation procedure for reported information. Journal of Applied Sport Psychology, 17(1), 1–19. http://doi.org/10.1080/10413200590907531     171  Côté, J., & Hancock, D. J. (2014). Evidence-based policies for youth sport programmes. International Journal of Sport Policy and Politics, 8(1), 51–65. http://doi.org/10.1080/19406940.2014.919338 Côté, J., Lidor, R., & Hackfort, D. (2009). ISSP position stand: To sample or to specialize? Seven postulates about youth sport activities that lead to continued participation and elite performance. International Journal of Sport and Exercise Psychology, 7(1), 7–17. http://doi.org/10.1080/1612197X.2009.9671889 Côté, J., Murphy-Mills, J., & Abernethy, B. (2012). The development of skill in sport. In A. M. Williams & N. J. Hodges (Eds.), Skill acquisition in sport: Research, theory and practice (2nd ed., pp. 269–86). London: Routledge. Coughlan, E. K., Williams, A. M., McRobert, A. P., & Ford, P. R. (2014). How experts practice: a novel test of deliberate practice theory. Journal of Experimental Psychology. Learning, Memory, and Cognition, 40(2), 449–58. http://doi.org/10.1037/a0034302 Coutinho, P., Mesquita, I., Davids, K., Fonseca, A. M., & Côté, J. (2016). How structured and unstructured sport activities aid the development of expertise in volleyball players. Psychology of Sport and Exercise, 25, 51–59. http://doi.org/10.1016/j.psychsport.2016.04.004 Cushion, C., Ford, P. R., & Williams, A. M. (2012). Coach behaviours and practice structures in youth soccer: Implications for talent development. Journal of Sports Sciences, (January 2013), 1–11. http://doi.org/10.1080/02640414.2012.721930 Cushion, C. J., & Jones, R. L. (2012). A Bourdieusian analysis of cultural reproduction: socialisation and the “hidden curriculum” in professional football. Sport, Education and     172  Society, 19(3), 276–298. http://doi.org/10.1080/13573322.2012.666966 Davids, K., Button, C., & Bennet, S. (2008). Dynamics of skill acquisition: A constraints led approach. Chalmpaign, IL: Human Kinetics. Deci, E. L., Koestner, R., & Ryan, R. M. (2001). Extrinsic rewards and intrinsic motivation in education: Reconsidered once again. Review of Educational Research Spring, 71(1), 1–27. http://doi.org/10.3102/00346543071001001 Deci, E. L., & Ryan, R. M. (2002). Handbook of Self-determination Research. New York: University Rochester Press. Retrieved from https://books.google.com/books?hl=en&lr=&id=DcAe2b7L-RgC&pgis=1 Deci, E. L., & Ryan, R. M. (2011). Multiple identities within a single self: A self-determination theory perspective on internalization within contexts and cultures. In M. R. Leary & J. P. Tagney (Eds.), Handbook of self & identity (pp. 253–274). New York: The Guilford Press. Deci, E. L., Ryan, R. M., & Koestner, R. (1999). A Meta-Analytic Review of Eperiments Examining the Effects of Extrinsic Rewards on Intrinsic Motivation. Psychological Bulletin, 125(6), 627–668. http://doi.org/10.1037/0033-2909.125.6.627 Deci, E., & Ryan, R. (2012). Motivation, personality, and development within embedded social contexts: An overview of self-determination theory. (R. M. Ryan, Ed.)The Oxford handbook of human motivation. Oxford, UK: Oxford University Press. Retrieved from https://books.google.ca/books?hl=en&lr=&id=YU8IpnTnQasC&oi=fnd&pg=PT104&dq=Motivation,+personality,+and+development+within+embedded+social+contexts:+An+overview+of+self-determination+theory&ots=5OFSAb3skF&sig=7xPI_zjX_9VsqQnchnG_73moImA     173  Deci, E., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior. Springer Science & Business Media. Retrieved from https://books.google.com/books?hl=en&lr=&id=p96Wmn-ER4QC&pgis=1 Dent, C. W., Sussman, S. Y., & Stacy, A. W. (1997). The impact of a written parental consent policy on estimates from a school-based drug use survey. Evaluation Review, 21(6), 698-712. Duda, J. L. (2013). The conceptual and empirical foundations of Empowering CoachingTM: Setting the stage for the PAPA project. International Journal of Sport and Exercise Psychology, 11(4), 311–318. http://doi.org/10.1080/1612197X.2013.839414 Elferink-Gemser, M. T., Visscher, C., Richart, H., & Lemmink, K. A. P. M. (2004). Development of the Tactical Skills Inventory for Sports. Perceptual and Motor Skills, 99(3 Pt 1), 883–95. http://doi.org/10.2466/pms.99.3.883-895 Ericsson, K. A. (2006). The Influence of Experience and Deliberate Practice on the Development of Superior Expert Performance. In The Cambridge Handbook of Expertise and Expert Performance (pp. 685–705). http://doi.org/10.1017/CBO9780511816796.038 Ericsson, K. A. (2014). Why expert performance is special and cannot be extrapolated from studies of performance in the general population: A response to criticisms. Intelligence, 45(1), 81–103. http://doi.org/10.1016/j.intell.2013.12.001 Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review. Ericsson, K. A., & Lehman, C. (1999). Expertise. In M. A. Runco & S. R. Pritzker (Eds.),     174  Encyclopedia of Creativity (pp. 695–707). San Diego: Academic Press. Ericsson, K. A., & Pool, R. (2016). Peak: Secrets from the new science of expertise. New York: Houghton Mifflin Harcourt. Ericsson, K. A., & Towne, T. J. (2010). Expertise. Wiley Interdisciplinary Reviews. Cognitive Science, 1(3), 404–16. http://doi.org/10.1002/wcs.47 Fogarty, G. J., & Else, D. (2005). Performance calibration in sport: Implications for self‐confidence and metacognitive biases. International Journal of Sport and Exercise Psychology, 3(1), 41–57. http://doi.org/10.1080/1612197X.2005.9671757 Ford, P. R., Carling, C., Garces, M., Marques, M., Miguel, C., Farrant, A., … Williams, M. (2012). The developmental activities of elite soccer players aged under-16 years from Brazil, England, France, Ghana, Mexico, Portugal and Sweden. Journal of Sports Sciences, 30(15), 1–11. http://doi.org/10.1080/02640414.2012.701762 Ford, P. R., Coughlan, E. K., Hodges, N. J., & Williams, A. M. (2015). Deliberate practice in sport. In J. Baker & D. Farrow (Eds.), Routledge handbook of sports expertise (pp. 347–363). London: Routledge. Ford, P. R., Low, J., McRobert, A. P., & Williams, A. M. (2010). Developmental activities that contribute to high or low performance by elite cricket batters when recognizing type of delivery from bowlers’ advanced postural cues. Journal of Sport & Exercise Psychology, 32(5), 638–54. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/20980708 Ford, P. R., Ward, P., Hodges, N. J., & Williams, A. M. (2009). The role of deliberate practice and play in career progression in sport: the early engagement hypothesis. High Ability     175  Studies, 20(1), 65–75. http://doi.org/10.1080/13598130902860721 Ford, P. R., & Williams, A. M. (2011). No relative age effect in the birth dates of award-winning athletes in male professional team sports. Research Quarterly for Exercise and Sport, 82(3), 570–573. http://doi.org/10.1080/02701367.2011.10599790 Ford, P. R., & Williams, A. M. (2012). The developmental activities engaged in by elite youth soccer players who progressed to professional status compared to those who did not. Psychology of Sport and Exercise, 13(3), 349–352. http://doi.org/10.1016/j.psychsport.2011.09.004 G. Hoare, D., & Warr, C. R. (2000). Talent identification and women’s soccer: An Australian experience. Journal of Sports Sciences, 18(9), 751–758. http://doi.org/10.1080/02640410050120122 Gill, D. L. (2001). Feminist sport psychology: A guide for our journey. The Sport Psychologist, 15(4), 363–372. Retrieved from http://search.proquest.com/docview/619648044?accountid=11440%5Cnhttp://pr7mz9rq5v.search.serialssolutions.com/?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&rfr_id=info:sid/ProQ%3Apsycinfo&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.jtit Gillet, N., Berjot, S., & Gobancé, L. (2009). A motivational model of performance in the sport domain. European Journal of Sport Science, 9(3), 151–158. http://doi.org/10.1080/17461390902736793 Gillet, N., Berjot, S., Vallerand, R. J., Amoura, C., & Rosnet, E. (2012). Examining the motivation-performance relationship in competitive sport: A cluster-analytic approach.     176  International Journal of Sport Psychology, 42(2), 79–102. Retrieved from https://www.researchgate.net/publication/237088605_Examining_the_Motivation-Performance_Relationship_in_Competitive_Sport_A_Cluster-Analytic_Approach Glass, G. V, Peckham, P. D., & Sanders, J. R. (1972). Consequences of failure to meet assumptions underlying the fixed effects analyses of variance and covariance. Review of Educational Research, 42(3), 237–288. http://doi.org/10.3102/00346543042003237 Gledhill, A., & Harwood, C. (2014). Developmental experiences of elite female youth soccer players. International Journal of Sport and Exercise Psychology, 12(2), 150–165. http://doi.org/10.1080/1612197X.2014.880259 Gledhill, A., & Harwood, C. (2015). A holistic perspective on career development in UK female soccer players: A negative case analysis. Psychology of Sport and Exercise, 21, 65–77. http://doi.org/10.1016/j.psychsport.2015.04.003 Gobet, F., & Campitelli, G. (2007). The role of domain-specific practice, handedness, and starting age in chess. Developmental Psychology, 43(1), 159–172. Guadagnoli, M. a, & Lee, T. D. (2004). Challenge point: a framework for conceptualizing the effects of various practice conditions in motor learning. Journal of Motor Behavior, 36(2), 212–224. http://doi.org/10.3200/JMBR.36.2.212-224 Güllich, A. (2014). Many roads lead to Rome - Developmental paths to Olympic gold in men’s field hockey. European Journal of Sport Science, Ahead of P(0), 1–9. http://doi.org/10.1080/17461391.2014.905983 Güllich, A. (2016). International medallists’ and non-medallists’ developmental sport activities –     177  a matched-pairs analysis. Journal of Sports Sciences, 1–8. http://doi.org/10.1080/02640414.2016.1265662 Hambrick, D. Z., Altmann, E. M., Oswald, F. L., Meinz, E. J., Gobet, F., & Campitelli, G. (2014). Accounting for expert performance: The devil is in the details. Intelligence, 45(1), 112–114. http://doi.org/10.1016/j.intell.2014.01.007 Hambrick, D. Z., Oswald, F. L., Altmann, E. M., Meinz, E. J., Gobet, F., & Campitelli, G. (2014). Deliberate practice: Is that all it takes to become an expert? Intelligence, 45(1), 34–45. http://doi.org/10.1016/j.intell.2013.04.001 Handford, C., Davids, K., Bennett, S., & Button, C. (1997). Skill acquisition in sport: Some applications of an evolving practice ecology. Journal of Sports Sciences. http://doi.org/10.1080/026404197367056 Hardy, L., Barlow, M., Evans, L., Rees, T., Woodman, T., & Warr, C. (2016). Great British medallists: Psychosocial biographies of Super-Elite and Elite athletes from Olympic sports. Progress in Brain Research, 1–119. http://doi.org/10.1016/bs.pbr.2017.03.004 Haugaasen, M., & Jordet, G. (2012). Developing football expertise: a football-specific research review. International Review of Sport and Exercise Psychology, 5(2), 177–201. http://doi.org/10.1080/1750984X.2012.677951 Helsen, W. F., Baker, J., Michiels, S., Schorer, J., Van winckel, J., & Williams,  a. M. (2012). The relative age effect in European professional soccer: Did ten years of research make any difference? Journal of Sports Sciences, 30(15), 1–7. http://doi.org/10.1080/02640414.2012.721929     178  Helsen, W. F., Hodges, N. J., Kel, J., & Starkes, J. L. (2000). The roles of talent, physical precocity and practice in the development of soccer expertise. Journal of Sports Sciences, 18(9), 727–736. http://doi.org/10.1080/02640410050120104 Hendry, D. T., Crocker, P. R. E., & Hodges, N. J. (2014). Practice and play as determinants of self-determined motivation in youth soccer players. Journal of Sports Sciences, 32(11), 1091–9. http://doi.org/10.1080/02640414.2014.880792 Hendry, D. T., Ford, P. R., Williams, A. M., & Hodges, N. J. (2015). Five evidenced based principles of effective practice and instruction. In J. Baker & D. Farrow (Eds.), Routledge handbook of sports expertise (pp. 414–429). New York, NY: Routledge. Hendry, D. T., & Hodges, N. J. (2013). Getting on the right track: athlete centred practice for expert performance in sport. In T. McGarry, P. O’Donoghue, & J. Sampaio (Eds.), Routledge handbook of sports performance analysis (pp. 5–20). London: Routledge. Hendry, D.T., & Hodges, N.J., (2018). Early majority engagement pathway best defines transitions from youth to adult elite men's soccer in the UK: A three time-point retrospective and prospective study Psychology of Sport & Exercise. 36, 81-89 Hendry, D.T., Williams, A.M., & Hodges, N. J. (2018). Coach ratings of skills and their relations to practice, play and successful transitions from youth-elite to adult-professional status in soccer. Journal of Sport Sciences,  Hill, B., & Sotiriadou, P. (2016). Coach decision-making and the relative age effect on talent selection in football. European Sport Management Quarterly, 16(3), 292–315. http://doi.org/10.1080/16184742.2015.1131730     179  Hodge, K., & Lonsdale, C. (2011). Prosocial and antisocial behavior in sport: the role of coaching style, autonomous vs. controlled motivation, and moral disengagement. Journal of Sport & Exercise Psychology, 33(4), 527–547. Hodges, NJ, Starkes, J. (1996). Wrestling with the nature expertise: a sport specific test of Ericsson, Krampe and Tesch-Römer’s (1993) theory of “deliberate practice.” International Journal of Sport Psychology. Hodges, N. J., Ford, P. R., Hendry, D. T., & Williams, A. M. (2017). Getting gritty about practice and success: Motivational characteristics of great performers. Progress in Brain Research. http://doi.org/10.1016/bs.pbr.2017.02.003 Hodges, N. J., Huys, R., & Starkes, J. L. (2007). Methodological review and evaluation of research in expert performance in sport. In G. Tenenbaum & R. C. Eklund (Eds.), Handbook of Sport Psycology (Vol. 53, pp. 161–183). New Jersey: John Wiley & Sons. http://doi.org/10.1017/CBO9781107415324.004 Hodges, N. J., Kerr, T., Starkes, J. L., Weir, P. L., & Nananidou, A. (2004). Predicting performance times from deliberate practice hours for triathletes and swimmers: What, when, and where is practice important? Journal of Experimental Psychology: Applied, 10(4), 219–237. Holt, N. L., & Dunn, J. G. H. (2004). Toward a Grounded Theory of the Psychosocial Competencies and Environmental Conditions Associated with Soccer Success. Journal of Applied Sport Psychology, 16(3), 199–219. http://doi.org/10.1080/10413200490437949 Höner, O., Votteler, A., Schmid, M., Schultz, F., & Roth, K. (2014). Psychometric properties of the motor diagnostics in the German football talent identification and development     180  programme. Journal of Sports Sciences, 33(2), 1–15. http://doi.org/10.1080/02640414.2014.928416 Hopwood, M. J. (2015). Issues in the collection of athlete training histories. In J. Baker & D. Farrow (Eds.), Routledge handbook of sports expertise (pp. 156–165). New York: Routledge. Hornig, M., Aust, F., & Güllich, A. (2014). Practice and play in the development of German top-level professional football players. European Journal of Sport Science, (December 2014), 37–41. http://doi.org/10.1080/17461391.2014.982204 Hornig, M., Aust, F., & Güllich, A. (2016). Practice and play in the development of German top-level professional football players. European Journal of Sport Science, 16(1), 96–105. http://doi.org/10.1080/17461391.2014.982204 Huberty, C. J., & Morris, J. D. (1989). Multivariate analysis versus multiple univariate analyses. Psychological Bulletin, 105(2), 302–308. http://doi.org/10.1037/0033-2909.105.2.302 Huijgen, B. C. H., Elferink-Gemser, M. T., Ali, A., & Visscher, C. (2013). Soccer skill development in talented players. International Journal of Sports Medicine, 34(8), 720–6. http://doi.org/10.1055/s-0032-1323781 Huijgen, B. C. H., Elferink-Gemser, M. T., Post, W. J., & Visscher, C. (2009). Soccer skill development in professionals. International Journal of Sports Medicine, 30(8), 585–591. http://doi.org/10.1055/s-0029-1202354 Jordet, G. (2015). Psychological characteristics of experts. In J. Baker & D. Farrow (Eds.), Routledge handbook of sports expertise (pp. 106–120). London: Routledge.     181  Kannekens, R., Elferink-Gemser, M. T., Post, W. J., & Visscher, C. (2009). Self-assessed tactical skills in elite youth soccer players: a longitudinal study. Perceptual and Motor Skills, 109(2), 459–72. http://doi.org/10.2466/PMS.109.2.459-472 Kannekens, R., Elferink-Gemser, M. T., & Visscher, C. (2009). Tactical skills of world-class youth soccer teams. Journal of Sports Sciences, 27(8), 807–12. http://doi.org/10.1080/02640410902894339 Koslowsky, M., & Botelho, M. D. C. (2010). Domains in the practice of the football learning: comparative study among football athletes of junior category in Portugal and Brazil. Journal of Human Sport and Exercise, 5(3), 400–410. http://doi.org/10.4100/jhse.2010.53.10 Kruger, J., & Dunning, D. (1999). Unskilled and unaware of it: How difficulties in recognizing one’s own incompetence lead to inflated self-assessments. Journal of Personality and Social Psychology, 77(6), 1121–1134. http://doi.org/doi.org/10.1037/0022-3514.77.6.1121 Law, M. P., Côté, J., & Ericsson, K. A. (2008). Characteristics of expert development in rhythmic gymnastics: A retrospective study. International Journal of Sport and Exercise Psychology, 5(1), 82–103. http://doi.org/10.1080/1612197X.2008.9671814 le Gall, F., Carling, C., Williams, M., & Reilly, T. (2010). Anthropometric and fitness characteristics of international, professional and amateur male graduate soccer players from an elite youth academy. Journal of Science and Medicine in Sport / Sports Medicine Australia, 13(1), 90–5. http://doi.org/10.1016/j.jsams.2008.07.004 Leite, N., Baker, J., & Sampaio, J. (2009). Paths to expertise in Portuguese national team athletes. Journal of Sports Science and Medicine, 8(4), 560–566.     182  Lonsdale, C., Hodge, K., & Rose, E. a. (2008). The behavioral regulation in sport questionnaire (BRSQ): instrument development and initial validity evidence. Journal of Sport & Exercise Psychology, 30(3), 323–355. Low, J., Williams, A. M., McRobert, A. P., & Ford, P. R. (2013). The microstructure of practice activities engaged in by elite and recreational youth cricket players. Journal of Sports Sciences, 31(11), 1242–50. http://doi.org/10.1080/02640414.2013.778419 MacNamara, Á., & Collins, D. (2010). The role of psychological characteristics in managing the transition to university. Psychology of Sport and Exercise, 11(5), 353–362. http://doi.org/10.1016/j.psychsport.2010.04.003 MacNamara, A., Collins, D., & Giblin, S. (2015). Just let them play? Deliberate preparation as the most appropriate foundation for lifelong physical activity. Frontiers in Psychology. http://doi.org/10.3389/fpsyg.2015.01548 MacNamara, Á., Collins, D., & Giblin, S. (2015). Just let them play? Deliberate preparation as the most appropriate foundation for lifelong physical activity. Frontiers in Psychology, 6, 1548. http://doi.org/10.3389/fpsyg.2015.01548 Macnamara, B. N., Hambrick, D. Z., & Oswald, F. L. (2014). Deliberate practice and performance in music, games, sports, education, and professions: A meta-analysis. Psychological Science, 25(8), 1608–1618. http://doi.org/10.1177/0956797614535810 Macnamara, B. N., Moreau, D., & Hambrick, D. Z. (2016). The relationship between deliberate practice and performance in sports: A meta-analysis. Perspectives on Psychological Science, 11(3), 333–350. http://doi.org/10.1177/1745691616635591     183  Mann, D. L., & van Ginneken, P. J. M. A. (2017). Age-ordered shirt numbering reduces the selection bias associated with the relative age effect. Journal of Sports Sciences, 35(8), 784–790. http://doi.org/10.1080/02640414.2016.1189588 Masters, R. S. W., Poolton, J. M., Maxwell, J. P., & Raab, M. (2008). Implicit motor learning and complex decision making in time-constrained environments. Journal of Motor Behavior, 40(1), 71–9. http://doi.org/10.3200/JMBR.40.1.71-80 McCarthy, N., & Collins, D. (2014). Initial identification & selection bias versus the eventual confirmation of talent: evidence for the benefits of a rocky road? Journal of Sports Sciences, 32(17), 1604–1610. http://doi.org/10.1080/02640414.2014.908322 McCarthy, N., Collins, D., & Court, D. (2015). Start hard, finish better: further evidence for the reversal of the RAE advantage. Journal of Sports Sciences, 1–5. http://doi.org/10.1080/02640414.2015.1119297 McCarthy, N., Collins, D., & Court, D. (2016). Start hard, finish better: Further evidence for the reversal of the RAE advantage. Journal of Sports Sciences, 34(15), 1461–1465. http://doi.org/10.1080/02640414.2015.1119297 McGaghie, W. C., Issenberg, S. B., Cohen, E. R., Barsuk, J. H., & Wayne, D. B. (2011). Does simulation-based medical education with deliberate practice yield better results than traditional clinical education? A meta-analytic comparative review of the evidence. Academic Medicine : Journal of the Association of American Medical Colleges, 86(6), 706–11. http://doi.org/10.1097/ACM.0b013e318217e119 McMorris, T. (2004). Acquisition and performance of sports skills. Chichester: Wiley & Sons.     184  Memmert, D. (2015). Development of tactical creativity in sport. In J. Baker & D. Farrow (Eds.), Routledge handbook of sports expertise (pp. 363–372). London: Routledge. Memmert, D., Baker, J., & Bertsch, C. (2010). Play and practice in the development of sport‐specific creativity in team ball sports. High Ability Studies, 21(1), 3–18. http://doi.org/10.1080/13598139.2010.488083 Memmert, D., & Roth, K. (2007). The effects of non-specific and specific concepts on tactical creativity in team ball sports. Journal of Sports Sciences, 25(12), 1423–32. http://doi.org/10.1080/02640410601129755 Meylan, C., Cronin, J., Oliver, J., & Hughes, M. (2010). Reviews: Talent identification in soccer: The role of maturity status on physical, physiological and technical characteristics. International Journal of Sports Science and Coaching, 5(4), 571–592. Retrieved from http://multi-science.atypon.com/doi/abs/10.1260/1747-9541.5.4.571?journalCode=ijssc Mujika, I., Vaeyens, R., Matthys, S. P. J., Santisteban, J., Goiriena, J., & Philippaerts, R. (2009). The relative age effect in a professional football club setting. Journal of Sports Sciences, 27(11), 1153–1158. http://doi.org/10.1080/02640410903220328 Musch, J., & Grondin, S. (2001). Unequal competition as an impediment to personal development: A review of the relative age effect in sport. Developmental Review, 21(2), 147–167. http://doi.org/10.1006/drev.2000.0516 Neinstein, L. S., & Kaufman, F. R. (2002). Normal Physical Growth and Development. In Adolescent Health Care: A Practical Guide, 4th edition. Ng, J. Y. Y., Lonsdale, C., & Hodge, K. (2011). The Basic Needs Satisfaction in Sport Scale     185  (BNSSS): Instrument development and initial validity evidence. Psychology of Sport and Exercise, 12(3), 257–264. http://doi.org/10.1016/j.psychsport.2010.10.006 Otis, N., Grouzet, F. M. E., & Pelletier, L. G. (2005). Latent motivational change in an academic setting: A 3-year longitudinal study. Journal of Educational Psychology, 97(2), 170–183. http://doi.org/10.1037/0022-0663.97.2.170 Pachman, M., Sweller, J., & Kalyuga, S. (2013). Levels of knowledge and deliberate practice. Journal of Experimental Psychology. Applied, 19(2), 108–19. http://doi.org/10.1037/a0032149 Pallant, J. (2007). SPSS survival manual: a step by step guide to data analysis using SPSS. Step by step guide to data analysis using the SPSS program (3rd ed.). Sydney: McGraw-Hill. Retrieved from http://solo.bodleian.ox.ac.uk/primo_library/libweb/action/search.do?dscnt=1&vl(drStartDay6)=00&vl(60606418UI3)=all_items&scp.scps=scope:(OULS),scope:(NET)&tab=local&srt=rank&mode=Advanced&vl(1UIStartWith1)=contains&tb=t&indx=1&vl(drEndYear6)=Year&vl(freeT Partington, M., Cushion, C., & Harvey, S. (2014). An investigation of the effect of athletes’ age on the coaching behaviours of professional top-level youth soccer coaches. Journal of Sports Sciences, 32(5), 403–414. http://doi.org/10.1080/02640414.2013.835063 Pelletier, L. G., Fortier, M. S., Vallerand, R. J., & Brière, N. M. (2001). Associations Among Perceived Autonomy Support, Forms of Self-Regulation, and Persistence: A Prospective Study. Motivation and Emotion, 25(4), 279–306. http://doi.org/10.1023/A:1014805132406 Pfister, G. (2015). Assessing the sociology of sport: On women and football. International     186  Review for the Sociology of Sport, 50(4–5), 563–569. http://doi.org/10.1177/1012690214566646 Post, E. G., Trigsted, S. M., Riekena, J. W., Hetzel, S., McGuine, T. A., Brooks, M. A., & Bell, D. R. (2017). The Association of Sport Specialization and Training Volume with Injury History in Youth Athletes. American Journal of Sports Medicine, 45(6), 1405–1412. http://doi.org/10.1177/0363546517690848 Ratelle, C. F., Guay, F., Vallerand, R. J., Larose, S., & Senécal, C. (2007). Autonomous, controlled, and amotivated types of academic motivation: A person-oriented analysis. Journal of Educational Psychology, 99(4), 734–746. http://doi.org/10.1037/0022-0663.99.4.734 Rees, T., Hardy, L., Güllich, A., Abernethy, B., Côté, J., Woodman, T., … Warr, C. (2016). The Great British Medalists Project: A Review of Current Knowledge on the Development of the World’s Best Sporting Talent. Sports Medicine, 46(8), 1041–1058. http://doi.org/10.1007/s40279-016-0476-2 Roca, A., Ford, P. R., McRobert, A. P., & Williams, A. M. (2013). Perceptual-cognitive skills and their interaction as a function of task constraints in soccer. Journal of Sport & Exercise Psychology, 35(2), 144–55. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23535973 Roca, A., & Williams, A. M. (2017). Does decision making transfer across similar and dissimilar sports? Psychology of Sport and Exercise, 31, 40–43. http://doi.org/10.1016/j.psychsport.2017.04.004 Roca, A., Williams, A. M., & Ford, P. R. (2012a). Developmental activities and the acquisition of superior anticipation and decision making in soccer players. Journal of Sports Sciences,     187  30(15), 1643–52. http://doi.org/10.1080/02640414.2012.701761 Roca, A., Williams,  a. M., & Ford, P. R. (2012b). Developmental activities and the acquisition of superior anticipation and decision making in soccer players. Journal of Sports Sciences, 30(15), 1–10. http://doi.org/10.1080/02640414.2012.701761 Ryan, R. M., & Deci, E. L. (2017). Self-Determination Theory: Basic Psychological Needs in Motivation, Development, and Wellness (2nd ed.). New York: Guilford Press. Retrieved from https://www.guilford.com/books/Self-Determination-Theory/Ryan-Deci/9781462528769 Santos, S. D. L., Memmert, D., Sampaio, J., & Leite, N. (2016). The spawns of creative behavior in team sports: A creativity developmental framework. Frontiers in Psychology. http://doi.org/10.3389/fpsyg.2016.01282 Singer, R. N., & Janelle, C. M. (1999). Determining sport expertise: From genes to supremes. International Journal of Sport Psychology, 30(2), 117–150. Soberlak, P., & Côté, J. (2003). The developmental activities of elite ice hockey players. Journal of Applied Sport Psychology, 15(1), 41–49. http://doi.org/10.1080/10413200305401 Swann, C., Moran, A., & Piggott, D. (2015). Defining elite athletes: Issues in the study of expert performance in sport psychology. Psychology of Sport and Exercise, 16, 3–14. http://doi.org/10.1016/j.psychsport.2014.07.004 Unger, J. M., Keith, N., Hilling, C., Gielnik, M. M., & Frese, M. (2009). Deliberate practice among South African small business owners: Relationships with education, cognitive ability, knowledge, and success. Journal of Occupational and Organizational Psychology,     188  82(1), 21–44. http://doi.org/10.1348/096317908X304361 Unnithan, V., White, J., Georgiou, A., Iga, J., & Drust, B. (2012). Talent identification in youth soccer. Journal of Sports Sciences, 30(15), 1719–1726. http://doi.org/10.1080/02640414.2012.731515 Vaeyens, R., Malina, R. M., Janssens, M., Van Renterghem, B., Bourgois, J., Vrijens, J., & Philippaerts, R. M. (2006). A multidisciplinary selection model for youth soccer: the Ghent Youth Soccer Project. British Journal of Sports Medicine, 40(11), 928–934. http://doi.org/10.1136/bjsm.2006.029652 Vallerand, R. J. (2010). On passion for life activities: The dualistic model of passion. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (1st ed., Vol. 42, pp. 97–193). New York: Academic Press. http://doi.org/10.1016/S0065-2601(10)42003-1 Vansteenkiste, M., Lens, W., & Deci, E. L. (2006). Intrinsic versus extrinsic goal contents in Self-determination theory: Another look at the quality of academic motivation. Educational Psychologist, 41(1), 19–31. http://doi.org/10.1207/s15326985ep4101_4 Vink, K., Raudsepp, L., & Kais, K. (2015). Intrinsic motivation and individual deliberate practice are reciprocally related: Evidence from a longitudinal study of adolescent team sport athletes. Psychology of Sport and Exercise, 16, 1–6. http://doi.org/10.1016/j.psychsport.2014.08.012 Ward, P., Hodges, N. J., Starkes, J. L., & Williams, A. M. (2007). The road to excellence: deliberate practice and the development of expertise. High Ability Studies, 18(2), 119–153. http://doi.org/10.1080/13598130701709715     189  Ward, P., & Williams, A. M. (2003). Perceptual and cognitive skill development in soccer : The multidimensional nature of expert performance. Journal of Sport & Exercise Psychology, 25, 93–111. Wattie, N., MacDonald, D. J., & Cobley, S. (2015). Birthdate and birthplace effects on expertise attainmnet. In J. Baker & D. Farrow (Eds.), Routledge handbook of sports expertise (pp. 373–383). London: Routledge. Williams, A. M., & Hodges, N. J. (2005). Practice, instruction and skill acquisition in soccer: challenging tradition. Journal of Sports Sciences, 23(6), 637–650. http://doi.org/10.1080/02640410400021328 Williams, A. M., & Reilly, T. (2000). Talent identification and development in soccer. Journal of Sports Sciences, 18(9), 657–67. http://doi.org/10.1080/02640410050120041 Williams, A. M., Ward, P., Bell-Walker, J., & Ford, P. R. (2012). Perceptual-cognitive expertise, practice history profiles and recall performance in soccer. British Journal of Psychology, 103(3), 393–411. http://doi.org/10.1111/j.2044-8295.2011.02081.x Williams, A. M., Ward, P., Ward, J. D., & Smeeton, N. J. (2008). Domain specificity, task specificity, and expert performance. Research Quarterly for Exercise and Sport, 79(3), 428–433. Williams, J. (2007). A Beautiful Game: International Perspectives on Women’s Football. Berg. Retrieved from https://www.dora.dmu.ac.uk/handle/2086/695 Zibung, M., & Conzelmann, A. (2013). The role of specialisation in the promotion of young football talents : A person-oriented study. European Journal of Sport Science, 13(5), 452–    190  60. http://doi.org/10.1080/17461391.2012.749947 Zuber, C., Zibung, M., & Conzelmann, A. (2014). Motivational patterns as an instrument for predicting success in promising young football players. Journal of Sports Sciences, 33(2), 160–168. http://doi.org/10.1080/02640414.2014.928827       191  Appendices Appendix A: T1 questionnaire package  Study 1, 2 and 3 A.1: Club contact letter   Developmental Football Activities, Skill and Motivation Questionnaires A Prospective Study  Principal Investigator:  Co-Investigator: Nicola Hodges, Ph.D.   David T Hendry School of Kinesiology  School of Kinesiology University of British Columbia University of British Columbia nicola.hodges@ubc.ca  hendrydt@interchange.ubc.ca  Dear Coach,  By way of introduction, my name is David Hendry, a Master’s degree student currently studying ‘Skill Acquisition’ at the University of British Columbia (UBC) in Vancouver, Canada. Whilst living in Scotland I was heavily involved in Football Development as an employee of the Scottish F.A., Rangers F.C., Falkirk F.C. and Stirling Albion F.C. for over 10 years, and I hope to continue my contribution to the game through research. In particular, I am interested in career progressions of youth soccer players and the factors that make highly skilled, creative and passionate players and successful teams. These aspects of skill development have been identified as important to success in sport and hence I wish to study the relationships between these variables and the practice environments of individuals involved in football.  I am working with Dr Nicola Hodges, who runs the ‘Motor Skills Lab’ (http://msl.kin.educ.ubc.ca/) at UBC and who has been studying expert performance and motor learning for the past 15 years. Dr Hodges has conducted research related to soccer skills and she has both worked with professional soccer players (Tottenham Hotspur FC) and conducted research with professional clubs in the UK (Liverpool, Everton, Tottenham Hotspur and Sheffield United), with a primary focus on youth development. For this reason, I am writing to enquire about the possibility of administering questionnaires to your U13, U15 & U17 coaches, players and a number of parents. These questionnaires have been designed to take approximately 30 minutes to complete and they can be completed either before or after practice or competition. A brief rationale for the proposed study can be found below. By consenting to allow us to administer questionnaires does not mean that you consent for all parties to participate. Individual consent will be sought from the players, parents and coaches as evidenced by their willingness to complete a questionnaire.  Study proposal:  We aim to assess how early sport activity experiences affect the development of football related skills (technical, tactical, creative) and psychological skills related to passion, desire and motivation. Through specifically designed, validated questionnaires, we will investigate the     192  relationships between early sport experiences in general, engagement in football-specific activities, (i.e. street soccer and organized practice) and current levels of passion and motivation. These analyses will form part of a wider understanding of optimal youth sport development and the benefits and costs of early sport-specific specialization. Because we aim to be collecting data from players and coaches, in the UK and in Canada, we will be able to study cross-cultural differences as well as differences in perceptions of ability across players and coaches. Although not part of this study, we hope to validate our questionnaire methods with specific skill tests (tactical and technical) and relate these to psychological indices of desire and passion. We expect that the results of our study will contribute to the existing body of literature on the development of expertise in soccer. In a practical sense, we hope that the findings will help foster successful and positive youth sport development in soccer, with research-based evidence serving to guide the design of effective practice environments. All proposed measurement scales have been verified and deemed appropriate by the research ethics’ board at the University of British Columbia. Moreover any information collected will also be held in the strictest of confidence and no specific details will be included in subsequent publication or presentation. General results will be made available to interested persons, but no information will be provided that will serve to identify individual coaches, parents or players. If your club is willing to participate in the study please confirm by emailing me at hendrydt@interchange.ubc.ca .  Data collection will take place from October 1st -14th and specific times/dates can be negotiated in due course.  Please do not hesitate to contact me via email if you require any further information.  Kind Regards David Hendry Addendum It is widely recognized that practice is the most important variable in becoming an elite youth soccer player. Recent research has also shown that time spent in playful activities (i.e. street soccer) during early development is also an important factor in the development of expert soccer players. This has led top European clubs including F.C. Porto and Bayern Munich F.C. to recreate soccer play activities as part of their sessions. Moreover Long Term Athlete Development models advocate involvement in a variety of sports during the early years (5-12 years) as being beneficial to skill acquisition. Finally, key psychological variables such as passion and motivation have been shown to be correlated with greater persistence and adherence in sport as well as investing more time in relevant practice activities.        193  A.2:  Parent/Guardian Information Letter  Developmental Football Activities, Skill and Motivation Questionnaires   Principal Investigator:  Co-Investigator: Nicola Hodges, Ph.D.   David T Hendry School of Kinesiology  School of Kinesiology University of British Columbia University of British Columbia nicola.hodges@ubc.ca  hendrydt@interchange.ubc.ca     Dear Parent/Guardian,  We are a group of researchers at the University of British Columbia in Vancouver, Canada, interested in career progressions of youth soccer players and the factors that make highly skilled, creative and passionate players and successful teams. These aspects of skill development have been identified as important to success in sport and hence we wish to study the relationships between these variables and the practice environments of individuals involved in football. In 3 weeks’ time we will be going in to your child’s club and will be inviting players to complete a number of short questionnaires. The questionnaires will ask players a number of questions about the types of football activities that they have engaged in, how they rate their skill levels and also information about their passion and desire for football.  A copy of the questionnaires can be forwarded to you via email by contacting David Hendry using the address above.  It will take participants approximately 30 minutes to complete the questionnaires. None of the questions that we ask are of a delicate or intrusive nature and there are no known risks associated with a child’s involvement in this study.  Player participation is entirely voluntary, and even if players initially choose to take part in this study they may subsequently withdraw at any time without having to give any reason and without experiencing any negative consequences.   All answers that your child provides will be combined with those of other players who are taking part in this research and any information players provide will remain completely confidential. All completed questionnaires will be kept in a locked cabinet at the University of British Columbia and shall not be made available to anyone other than the researchers involved in this study.   If you DO NOT wish for your child to take part in this research, all we ask you to do is complete this form and return it to your child’s coach. Alternatively, you can email David Hendry using the contact details identified above and we will ensure that your son does not take part in this study. Also, even if you have consented for your child to take part in this     194  study, we also require his own consent as well before he can be invited to take part. If you have any questions or want further information about the study please contact the researchers. Alternatively, if you have any concerns about your rights or treatment as a research subject please contact the UBC Office of Research Services via email to RSIL@ors.ubc.ca.   IF YOU DO NOT WANT YOUR CHILD TO TAKE PART PLEASE SIGN THIS FORM AND RETURN THIS TO YOUR CHILD’S COACH:   I……………………………………………………………………………………    (Parent/Guardian Name)   DO NOT wish for my child …………………………………….……………..       (Child’s Name) to take part in this research.    Signed…………………………………………… Date………………………………………..   (Parent/Guardian Name)     Yours sincerely,           Nicola Hodges, PhD  David T Hendry (Principal Investigator (Co-Investigator)        195  A.3: Coach Information Letter  Developmental Football Activities, Skill and Motivation Questionnaires   Principal Investigator:  Co-Investigator: Nicola Hodges, Ph.D.   David T Hendry nicola.hodges@ubc.ca  hendrydt@interchange.ubc.ca  Dear Coach,  We are a group of researchers at the University of British Columbia in Vancouver, Canada, interested in how the types of football activities, like street football and organized practice sessions, may make highly skilled, creative and passionate players and successful teams. These aspects of skill development have been identified as important to success in sport and we wish to learn more about these relationships.  In 3 weeks’ time we will visit your club and would like you to complete a number of short questionnaires. The purpose of the questionnaires is to provide some information on your football coaching background as well as information relating to typical practice sessions and passion for coaching.  It will take you approximately 20-25 minutes to complete the questionnaires. None of the questions that we ask are of a delicate or intrusive nature and there are no known risks associated with your involvement in this study.  Participation is entirely voluntary, and even if you initially choose to take part in this study you may leave at any time without having to give any reason and without experiencing any negative consequences.   All answers that you provide will be combined with those of other coaches who are taking part in this research and any information you provide will remain completely confidential. All completed questionnaires will be kept in a locked cabinet at the University of British Columbia and shall not be made available to anyone other than the researchers involved in this study.   Your consent will be provided by completing the questionnaire. Please note that you are not obligated to participate in the study and are free, at any time, to discontinue completing the questionnaires.  If you have any questions or want further information about the study please contact David Hendry.  Yours sincerely,          David T Hendry       196    A.4: Practice History Questionnaire (Time 1)                                     “The Role of Developmental Activities on Motivation,   Passion and Skill in Youth Soccer Players”    The purpose of this questionnaire is to find out information on sporting participation, motivation, and passion from elite youth football players and coaches ranging from 12 to 17 years of age.  Questions are asked about the amount of match-play, organized practice and play-related activities engaged in by youth players and the number of sports other than soccer that you have participated in.   There are 3 sections in total (A-C), you can answer them in any order. You do not have to complete the questionnaire if you do not want to and may take a questionnaire and return it blank. If you decide to take part then please complete the questionnaire to the best of your ability. It will take around 30 minutes to complete. By completing the questionnaire you are agreeing to take part in the study. If you need help answering or understanding the questions please ask one of the assistants. Try to answer as best as you can remember or as best as you think (not someone else). Please note that all information will be treated in strictest confidence. Only those directly involved in the study (that is, the researchers and not the coaches) will have access to the information that you give in this questionnaire. The coaches will only have access to general information that does not identify you or any other players. If you have any question about this questionnaire please contact David, the researcher running the study.     197   Section A: GENERAL INFORMATION  Please fill in the details below:1. 1. Name:_____________________________________________Today’s date______________  2. Your age:  3.  Date of birth (day, month, year):  4.  Name of current Football Club/Academy:  5.  What age group do you currently play in?:  6. Have you ever played for the first team/seniors (if yes, what age)?     7. What age group were you in when you joined this Club/Academy?     8. Have you played for any previous Clubs/Academy (Yes / No, please circle)?  9. If YES, please give the name and the age group when you joined.       198  SECTION B: Participation in All Sports  Start by listing all the sports /physical activities you have ever played/taken part in (any order) and as many as you like, include football.  From this list, pick out your top 5 sports and number them 1 -5, with the sport you’ve played the most as number 1 (football), second most as number 2 etc.    •    •    •    •    •    •    •    •    •    •    •    •              199   Practice History in all Sports Please answer all of the following questions as accurately as possible. Start by entering the names of the top 5 sports you have played in the boxes which read ‘Sport X = ’. We have included Football for you under ‘Sport 1 = ’. Enter the sport that you play the most, or used to play a lot, next to ‘Sport 2 =’.  Write your answers in the boxes directly below the name of each sport. See the EXAMPLE in grey for help with your answers.      START HERE    Questions  EXAMPLE Sport 1 = FOOTBALL Sport 2 = Sport 3 = Sport 4 = Sport 5 = 1 How old were you when you first started this sport?  8      2 During your first year, typically how many hours per week did you play/practice?  2½      3 How old were you when you first received organised practice (formal coaching)   10      4 Apart from in P.E., do you still take part in this sport? If you do not play anymore, enter the age you stopped. Yes / No  Age Stopped =  . Yes / No  Age Stopped =   Yes/ No  Age Stopped Yes/ No  Age Stopped =   Yes/ No  Age Stopped =   Yes/ No  Age Stopped =   5        6 On average, how many  days per week did you practice at:   5-8 years old =   9-12 years old =  12 years to now = Number of days per week =  2  3  0 Number of days per week = Number of days per week = Number of days per week = Number of days per week = Number of days per week = Typically, how long did you practice each day at:              5-8 years old =             9-12 years old =              12 years to now = Number of hoursper day = 1½ 2½ 0 Number of hours per day =    Number of hours per day = Number of hours per day = Number of hours per day = Number of hours per day =     200  SECTION C: Practice History in Football  As accurately as possible, try to recall and write down an average of how often (sessions per week) and how much time (hours per week) you spent in organised football practice, play (i.e. street football) and playing matches. Write down each number underneath each of the age categories. An example, in grey, can be found in the table below.  If you have taken a significant break from football at some point in your career (e.g., due to injury, long-term illness, etc.) then please enter the number of weeks that you took off in that year.  Organised Practice includes:  Practice activities that are conducted with a coach/teacher/adult and that are primarily designed to improve skills (formal coaching). This is typically team-led practice and could include things such as football drills, technical skills, tactical skills, strategic skills, coached small- sided games, conditioned games, set-play practices, football-related fitness work etc.  Play (non-organised practice) includes: Unstructured activities that are not conducted with a coach or teacher.This includes fun games, general kick around, pick-up games, individual play/practice, keep-ups etc.  Match-play includes: Playing competitive matches against another team or playing uninterrupted matches against other players in your team/club (bounce game)                     201    EXAMPLE: Sport Name Football  Age (years) =  5-6  7-8  9-10  11-12  13-14  15-16  17 Organised Practice Number of sessions per week =  1  2  3  4  4  6  10   Average length of session (hours) =  1  1 ½  2  1 ¾  2  2  2  Play Number of hours per week =  6  10  10  6  6  4  3  Match play Number of hours per week =  1  1  1  1 ½  1 ½  2  3  Significant breaks from football (through injury, or long term illness, exclude holiday & off season) Number of weeks off =  0  0  6  0  12  2  0              202          START HERE    Sport Name Football  Age (years) =  5-6  7-8  9-10  11-12  13-14  15-16  17 Organised Practice Number of sessions per week =        Average length of sessions (hours) =        Play Number of hours per week =         Match play Number of hours per week =         Significant breaks from football (through injury, or long term illness, exclude holiday & off season) Number of weeks off =                    203   A.5: Football Enjoyment Questionnaire SECTION D: Football Enjoyment Questionnaire (part a)  Please complete the following questionnaire by circling the appropriate number using a rating scale from 1-7, where 1 = Not at all true, 4 = somewhat true and 7 = very true. For each of the statements circle only ONE of the numbers in each row. Please make sure you answer this on your own.  1= Not at all true, 4 = Somewhat true, 7 = Very True  1             2           3             4             5             6            7  I participate in football because I enjoy it. 1 2 3 4 5 6 7 I participate in football because it’s part of who I am. 1 2 3 4 5 6 7 I participate in football because the benefits of football are important to me 1 2 3 4 5 6 7 I participate in football because I would feel ashamed if I quit 1 2 3 4 5 6 7 I participate in football because if I don’t other people will not be pleased with me 1 2 3 4 5 6 7 I participate in football but I wonder what’s the point 1 2 3 4 5 6 7 I participate in football because I like it 1 2 3 4 5 6 7 I participate in football because it is an opportunity to just be who I am 1 2 3 4 5 6 7 I participate in football because it teaches me self-discipline 1 2 3 4 5 6 7 I participate in football because I would feel like a failure if I quit 1 2 3 4 5 6 7 I participate in football because I feel pressure from other people to play 1 2 3 4 5 6 7     204       I participate in football but I question why I continue 1 2 3 4 5 6 7 I participate in football because it’s fun 1 2 3 4 5 6 7 I participate in football because what I do in football is an expression of who I am 1 2 3 4 5 6 7 I participate in football because I value the benefits of football 1 2 3 4 5 6 7 I participate in football because I feel obligated to continue 1 2 3 4 5 6 7 I participate in football because people push me to play 1 2 3 4 5 6 7 I participate in football but the reasons why are not clear to me anymore 1 2 3 4 5 6 7 I participate in football because I find it enjoyable. 1 2 3 4 5 6 7 I participate in football because it allows me to live in a way that is true to my values 1 2 3 4 5 6 7 I participate in football because it is a good way to learn things which could be useful to me in my life 1 2 3 4 5 6 7 I participate in football because I would feel guilty if I quit 1 2 3 4 5 6 7 I participate in football to satisfy people who want me to play 1 2 3 4 5 6 7 I participate in football but I question why I am putting myself through this 1 2 3 4 5 6 7                                                                                                                                 1=Not at all true, 4 = Somewhat true, 7 = Very True     1          2          3           4           5           6           7     205  A.6: Player skill ratings  My Skill ratings  1. In comparison to other players in your team, rate your current tactical, technical, creative and physical skill by circling the appropriate number (1 = Poor, 2 = Below average, 3 = Average, 4 = Above average, 5 = Excellent).  1 = Poor, 2 = Below average, 3 = Average, 4 = Above average, 5 = Excellent.  Tactical Skill (i.e. decision making/pass selection) 1 2 3 4 5 Technical Skill (i.e. passing, shooting, dribbling,) 1 2 3 4 5 Physical Skill (i.e. endurance, physical condition) 1 2 3 4 5 Creative Skill (i.e. unexpected, original and useful) 1 2 3 4 5  2. In comparison to other players of your age (e.g. school mates, relatives) who play football, rate your current  tactical, technical, physical and creative skill by circling the appropriate number (1 = Poor, 2 = Below average,  3 = Average, 4 = Above average, 5 = Excellent).   1 = Poor, 2 = Below average, 3 = Average, 4 = Above average, 5 = Excellent  Tactical Skill (i.e. decision making/pass selection) 1 2 3 4 5 Technical Skill (i.e. passing, shooting, dribbling,) 1 2 3 4 5 Physical Skill (i.e. endurance and physical condition) 1 2 3 4 5 Creative Skill (i.e. unexpected, original and useful) 1 2 3 4 5 Thank you very much for completing this questionnaire!  206   A.7: Parent questionnaire FOR PARENTS: Career Practice and Motivation Questionnaire   The purpose of this parental questionnaire is to find out information on sporting participation, and developmental activities of elite youth football players, ranging from 12 to 17 years of age. We would like you to provide this information on your own (or with your spouse/partner), but not in consultation with your son. Questions are asked about match-play, organised practice and play-related activities, and the amount of football specific practice in comparison to other sports.  Organised Practice includes: Practice activities that are conducted with a coach/teacher/adult that are used mainly to improve skills (formal practice). This is typically team-led practice and could include things such as football drills, technical skills, conditioned games, tactical skills, strategic skills, set- play practices, and football-related fitness work.  You do not have to complete the questionnaire and may stop filling out the questionnaire at any point. The questionnaire will take approximately 20 minutes to complete and by completing the questionnaire you are providing your consent. Your answers will provide information about the roles of key practice behaviours useful to the development of skilled youth football players as well as the reliability of this information. Please note that all information will be treated in strictest confidence. Only those directly involved in the study (that is, the researchers and not the coaches) will have access to the information that you give in this questionnaire. The coaches will only have access to general information that does not identify you, your son or any other player. If you have any question about this questionnaire please contact David Hendry, the researcher running the study.     PLAYER NAME:  AGE GROUP:    CLUB: 207  SECTION A: Practice History in Sport Start by entering the names of all sports your son has played in the boxes which read ‘Sport 1-5 = ’ (up to a max of 5, if there are more just list these over the page). We have included ‘Football’ for you in ‘Sport 1’. Enter the sport that your son plays the most, or used to play a lot, next to ‘Sport 2 =’ etc. Write your answers in the boxes directly below each sport. See the EXAMPLE in grey for help.             START HERE   Questions  EXAMPLE Sport 1 = FOOTBALL Sport 2 = Sport 3 = Sport 4 = Sport 5 = 1 How old were you when you first started this sport?  8      2 During your first year, typically how many hours per week did you play/practice?  2½      3 How old were you when you first received organised practice (formal coaching)   10      4 Apart from in P.E., do you still take part in this sport? If you do not play anymore, enter the age you stopped. Yes / No  Age Stopped =  . Yes / No  Age Stopped =   Yes/ No  Age Stopped Yes/ No  Age Stopped =   Yes/ No  Age Stopped =   Yes/ No  Age Stopped =   5        6 On average, how many  days per week did you practice at:               5-8 years old =             9-12 years old =              12 years to now = Number of days per week =   2 3 0 Number of days per week = Number of days per week = Number of days per week = Number of days per week = Number of days per week = Typically, how long did you practice each day at:              5-8 years old =             9-12 years old =              12 years to now = Number of hours/ day = 1½ 2½ 0 Number of hours per day =    Number of hours per day = Number of hours per day = Number of hours per day = Number of hours per day =  208     SECTION B: Practice History in Football  As accurately as possible, try to recall and write down an average of how often (sessions per week) and how much time (hours per week) your son spent in organised football practice, play (i.e. street football) and playing matches. Write down each number underneath each of the age categories. An example, in grey, can be found in the table below. If your son has taken a significant break from football at some point in his career (e.g., due to injury, long-term illness, etc.) then please enter the number of weeks that he took off in that year.  Organised Practice includes: Practice activities that are conducted with a coach/teacher/adult that are used mainly to improve skills (formal coaching). This is typically team-led practice and could include things such as football drills, technical skills, conditioned games, tactical skills, strategic skills, set- play practices, and football-related fitness work    Play (non-organised practice)                 Unstructured activities that are not conducted with a coach or teacher. includes: This       includes fun games, general kick around, football games with friends, individual play/practice, keep-ups etc.  Match-play includes: Playing competitive matches against another team or playing uninterrupted matches against other players in the team/club (bounce game)              209   EXAMPLE:  Sport Name Football  Age (years) =  5-6  7-8  9-10  11-12  13-14  15-16  17 Organised Practice Number of sessions per week =  1  2  3  4  4  6  10   Average length of session (hours) =  1  1 ½  2  1 ¾  2  2  2  Play Number of hours per week =  6  10  10  6  6  4  3  Match play Number of hours per week =  1  1  1  1 ½  1 ½  2  3  Significant breaks from football (through injury, or long term illness, exclude holiday & off season) Number of weeks off =  0  0  6  0  12  2  0          210        START HERE   Sport Name Football  Age (years) =  5-6  7-8  9-10  11-12  13-14  15-16  17 Organised Practice Number of sessions per week =        Average length of sessions (hours) =         Play Number of hours per week =         Match play Number of hours per week =         Significant breaks from football (through injury, or long term illness, exclude holiday & off season) Number of weeks off =          If you helped your son complete his practice history questionnaire, please tick here: □  211  A.8: Coach Questionnaire and Skill Ratings   “The Role of Developmental Activities on Self-Determined Motivation, Passion and Skill in Youth Soccer Players”   The purpose of this questionnaire is to acquire information on sporting participation, skill, motivation, and passion from elite youth football players and coaches. Information will be collected from coaches and a number of different age groups players, ranging from 12 to 17 years of age. Many questions are asked about the type of activities that players are involved in during typical training sessions. Other aspects of the questionnaire relate to passion, motivation and a rating of each players technical, tactical, creative and physical skill level.   Please complete the questionnaire to the best of your ability. Your answers will provide information as to key psychological aspects and practice habits in the development of skilled performance. You do not have to complete the questionnaire and may stop filling out the questionnaire at any point. The questionnaire will take approximately 15 minutes to complete and by completing the questionnaire you are providing your consent. Please note that all information will be treated in strictest confidence. Only those directly involved in the study will have access to information given in this questionnaire and at no time will the information you provide be made available to any person in your Club. We will never disclose personnel or identifying information about individuals, only group-based, summary data will be made available to all interested persons. If you have any queries regarding this questionnaire please contact David Hendry from the Motor Skills Lab at the University of British Columbia via email at hendrydt@interchange.ubc.ca or the Principal Investigator, Dr Nicola Hodges (nicola.hodges@ubc.ca: http://msl.kin.educ.ubc.ca/). Many thanks for your participation.     212  SECTION A: Coaches Demographic Information   Name:_______________________________   Email address:___________________________________________________________   Coaching Qualifications held:    ____________________________________________________               ____________________________________________________        ____________________________________________________        ____________________________________________________   Other relevant qualifications     ____________________________________________________        ____________________________________________________        ____________________________________________________       Number of years involved in coaching youth football      _______________years.  Number of years at your current club        _______________years.    Number of years coaching current age group squad      _______________years      Did you play professional soccer? (circle as appropriate)  yes / no  As a player, what was the highest level of soccer play  that you reached (i.e. amateur, international)        ____________________     As a player what position did you normally play?             ____________________     213  SECTION B: Typical Practice Sessions in Football  Please indicate as accurately as possible the amount of time spent in ONE WEEK of typical practice in each of the activities listed below (report your answers to the nearest ¼ hour/15 minutes).  How much time do you spend in each of the following activities?  Activity  Time Training Form Definition Hours minutes Fitness activity Improving fitness aspects of the game without a ball (e.g. warm-up, cool-down, conditioning, rest.   Technical  Isolated technical skills unopposed alone or in a group   Skills    Re-enacting isolated simulated game incidents with or without focus on particular technical skills   Playing Form    Small-sided games Match-play with reduced number of players and two goals.    Free play games  Small-sided games with no coaching or interruption    Conditioned games Small-sided games with altered rules, goals & or areas of play (e.g. possession, limited touches, zone games)   11v11 training games Coached 11v11 games on a full size pitch during training sessions.    Phase of play  Uni-directional match-play towards one goal   Other (please specify)       214  SECTION C: Technical, Tactical, Physical Skill Ratings  We would like you to rate the players who you are currently coaching in terms of their current tactical, technical, physical and creativity skill level.  To do this, please first list all the players names in the Table below under the heading “tactical skill” with their initials in brackets in the first table (you can list them in any order). In subsequent tables, all you need to do is enter the players in the same order, but this time, just give their initials.  Please use a 5 point scale to rate each player with respect to each of these 4 “skills”. When rating a player, please rate them in comparison to other players in the team, where 1 = Poor, 2= Below average, 3 = Average, 4 = Above average and 5 = excellent.  215  Tactical Skill Tactical skills are defined by the player’s ability to make fast and accurate decisions with respect to picking out open players, reading the game well, smart playing, good pass selection/decisions.   When rating a player TACTICAL skill ask yourself “in comparison to his team mates, how well is this player able to read the game and make good decisions?”    1 = Poor, 2 = Below average, 3 = Average, 4 = Above average, 5 = Excellent  Player Name and Initials 1 2 3 4 5 1.      2.      3.      4.      5.      6.      7.      8.      9.      10.      11.      12.      13.      14.      15.      16      17.      18.      19.      20.        216  Technical Skill Technical skills are defined by skills such as the ability of the player to pass accurately, dribble well with the ball, shoot, perform accurate and effective throw-ins/free-kicks etc.   When rating a player in terms of their TECHNICAL skill ask yourself “in comparison to his team mates, how technically skilled is this player?”   1 = Poor, 2 = Below average, 3 = = Average, 4 = Above average, 5 = Excellent  Player Name and Initials 1 2 3 4 5 1.      2.      3.      4.      5.      6.      7.      8.      9.      10.      11.      12.      13.      14.      15.      16      17.      18.      19.      20.       217  Physical Skill Physical skills are defined by a person’s overall physical condition.  When rating a player in terms of their PHYSICAL skill ask yourself “in comparison to his team mates, how physical fit and/or fast is this player?”    1 = Poor, 2 = Below average, 3 = = Average, 4 = Above average, 5 = Excellent   Player Name and Initials 1 2 3 4 5 1.      2.      3.      4.      5.      6.      7.      8.      9.      10.      11.      12.      13.      14.      15.      16      17.      18.      19.      20.        218  Creativity Creativity is defined by a person’s overall flair and originality in making decisions and displaying unusual skills and effective creative plays  When rating a player in terms of their CREATIVE skill ask yourself “in comparison to his team mates, how creative is this player on the ball and in making original decisions?”   1 = Poor, 2 = Below average, 3 = = Average, 4 = Above average, 5  = Excellent   Player Name and Initials 1 2 3 4 5 1.      2.      3.      4.      5.      6.      7.      8.      9.      10.      11.      12.      13.      14.      15.      16      17.      18.      19.      20.      Thank you for your help!     219  Appendix B: T2 questionnaire package Study 1, 2 and 3:  B.1: Professional club email.  Dear Craig I hope that this find you well.  In October 2011 we visited a number of Scottish youth football academies to study the effects of early practice experiences on skill level and motivation. We are hoping to conduct a final follow-up study to a chart the success of the players from the initial cohort and identify any trends in relation to the predictive capabilities of these data.  I am therefore contacting you to request access to the U13, U15 and U17 players at Glasgow Rangers F.C.  In addition, coaches would be invited to rate players’ technical, tactical, creative and skill level using a 1-5 scale.  Data collection will take place at the end of January/beginning of February and will take 25 minutes maximum to complete. This can be administered at the beginning or end of a training session. I have already confirmed visits to 2 other Scottish clubs but need one more club to give me the statistical power for valid and reliable results. Could you please let me know if Rangers F.C. is interested in participating in this follow-up study?   Many thanks, David       220  B.2: Parent/Guardian Information Letter  Developmental Football Activities, Skill and Motivation Questionnaires   Principal Investigator:  Co-Investigator: Nicola Hodges, Ph.D.   David T Hendry School of Kinesiology  School of Kinesiology University of British Columbia University of British Columbia nicola.hodges@ubc.ca  david_t_hendry@yahoo.co.uk     Dear Parent/Guardian,  We are a group of researchers at the University of British Columbia in Vancouver, Canada, who are conducting research into career progressions of youth soccer players and the factors that make highly skilled, creative and passionate players and successful teams. These aspects of skill development have been identified as important to success in sport and hence we wish to study the relationships between these variables and the practice environments of individuals involved in football. We have previously collected data from a sample of elite Scottish youth players (which might have included your child) and wish to conduct follow up research to see how these relationships change over time and whether we can predict success and positive well-being from practice histories. As part of a cross cultural study we are also collecting data from a sample of Canadian youth soccer players.  The Academy Director has agreed to disseminate information about this study and has also agreed to provide us with club records of anthropometrical (i.e. height, weight) and fitness test results (i.e. yo-yo intermittent, 10m sprint) in order to develop our understanding of these variables in elite youth soccer. In February and March 2016 we will be visiting your child’s club and inviting players, and coaches to complete a number of short questionnaires. The questionnaires will ask players a series of questions about the types of football activities that they have engaged in, and information about their passion and motivation in football. A copy of the questionnaires can be forwarded to you in advance via email by contacting David Hendry using the address above. We are also happy to provide general feedback on the study’s results should you so wish.  It will take players new to the study, (including all U13 players), approximately 40 minutes to complete. For returning players (i.e., those who completed questionnaires in the January/February of 2014), it should only take 20- 30 minutes. All players will be asked to give their assent to participate (see copy attached) on the day of the study. None of the questions that we ask are of a delicate or intrusive nature and there are no known risks associated with a child’s involvement in this study. Participation is entirely voluntary, and even if players or parents initially choose to take part in this study they may subsequently withdraw at any time without having to give any reason and without experiencing any negative consequences.    221  All of the answers provided will be combined with those of others taking part in this research and any information will remain completely confidential (i.e. club coaches and the Academy Director will not have access to these data). The results of this study will be reported in a graduate thesis and may also be published in journal articles and books. All completed questionnaires will be kept in a locked cabinet at the University of British Columbia and shall not be made available to anyone other than the researchers involved in this study.   If you DO NOT wish for your child to take part in this research, all we ask you to do is complete this form and return it to your child’s coach. Alternatively, you can email David Hendry using the contact details identified above and we will ensure that your son does not take part in this study. Acknowledgement of this request will then be sent to you detailing alternative arrangements for your son during this time (i.e., access to soccer related reading).  If you have any concerns or complaints about your rights as a research participant and/or your experiences while participating in this study, contact the Research Participant Complaint Line in the UBC Office of Research Ethics at 604-822-8598 or if long distance e-mail RSIL@ors.ubc.ca or call toll free 1-877-822-8598."  IF YOU DO NOT WANT YOUR CHILD TO TAKE PART PLEASE SIGN THIS FORM AND RETURN THIS TO YOUR CHILD’S COACH:   I……………………………………………………………………………………    (Parent/Guardian Name)   DO NOT wish for my child …………………………………….……………..       (Child’s Name) to take part in this research.    Signed…………………………………………… Date………………………………………..   (Parent/Guardian Name)     Yours sincerely,      Nicola Hodges, PhD  David T Hendry (Principal Investigator (Co-Investigator)    222  B.3: Player Consent Form  Developmental Football Activities, Skill and Motivation Questionnaires   Principal Investigator:  Co-Investigator: Nicola Hodges, Ph.D.   David T Hendry School of Kinesiology  School of Kinesiology University of British Columbia University of British Columbia nicola.hodges@ubc.ca  hendrydt@interchange.ubc.ca   Dear Player,  We are a group of researchers at the University of British Columbia in Vancouver, Canada, interested in how the types of football activities, like street football and organized practice sessions, may make highly skilled, creative and passionate players. These aspects of skill development have been identified as important to success in sport and we wish to learn more about these relationships.  As part of a follow up study of Scottish and Canadian youth football players, we would like you to complete a number of short questionnaires to provide information on the amount and types of football practice that you have been involved in so far. The questionnaires will also ask you to provide information about your passion and motivation in football. The Academy Director has approved this study and has also agreed to provide us with club records of your height, weight and fitness test results (i.e. yo-yo intermittent, 10m sprint) in order to develop our understanding of the relationships between practice, motivation and performance.  If you were part of the original study then you will only be asked to fill out a short questionnaire that will take around 20-30 minutes to complete. If you were not part of the original study or are an U13 player then the questionnaire will be slightly longer, taking about 40 minutes to complete. None of the questions that we ask are of a delicate or intrusive nature and there are no known risks associated with your involvement in this study. Participation is entirely voluntary, and even if you initially choose to take part in this study you may leave at any time without having to give any reason and without experiencing any negative consequences.   All answers that you provide will be combined with those of other players who are taking part in this research and any information you provide will remain completely confidential. The results of this study will be reported in a graduate thesis and may also be published in journal articles and books. All completed questionnaires will be kept in a locked cabinet at the University of British Columbia and shall not be made available to anyone other than the researchers involved in this study. No-one from your club, including the Academy Director or coaching staff will have access to your information. We would of course be happy to provide you with general feedback about the results of this study or some summary data of your own should you desire (please see email and website details for contact information).   223  Details of the study have already been provided to your parents by letter. If your parent/guardian did not return the attached form, then your parent/guardian has provided consent to take part in the study.  By signing this form you are demonstrating that you wish to take part in the study. If you have any questions or want further information about the study please ask the researcher.   If you have any concerns or complaints about your rights as a research participant and/or your experiences while participating in this study, contact the Research Participant Complaint Line in the UBC Office of Research Ethics at 604-822-8598 or if long distance e-mail RSIL@ors.ubc.ca or call toll free 1-877-822-8598."   SO, IF YOU WANT TO TAKE PART PLEASE SIGN THIS FORM AND RETURN THIS TO YOUR COACH OR THE RESEARCHER:   I……………………………………………………………………………………    (Name in Capitals)   wish to take part in this research.    Signed…………………………………………… Date………………………………………..        Yours sincerely,     Nicola Hodges, PhD  David T Hendry  (Principal Investigator (Co-Investigator)          224  B.4: Practice history questionnaire (new players)                                    “The Role of Developmental Activities on Motivation, Passion and Skill in Youth Soccer Players”    The purpose of this questionnaire is to find out information on sporting participation, motivation, and passion from elite youth football players and coaches ranging from 12 to 17 years of age.  Questions are asked about the amount of match-play, organized practice and play-related activities engaged in by youth players and the number of sports other than soccer that you have participated in.   There are 3 sections in total (A-C), you can answer them in any order. You do not have to complete the questionnaire if you do not want to and may take a questionnaire and return it blank. If you decide to take part then please complete the questionnaire to the best of your ability. It will take around 30 minutes to complete. By completing the questionnaire you are agreeing to take part in the study. If you need help answering or understanding the questions please ask one of the assistants. Try to answer as best as you can remember or as best as you think (not someone else). Please note that all information will be treated in strictest confidence. Only those directly involved in the study (that is, the researchers and not the coaches) will have access to the information that you give in this questionnaire. The coaches will only have access to general information that does not identify you or any other players. If you have any question about this questionnaire please contact David, the researcher running the study.  225   SECTION A: General Information  Please fill in the details below:  Name -  Today’s date -  Current age -  Date of Birth -  Current club - Age group -   1. How old were you when you first started playing football?  _______ 2. During your first year, how many hours per week did you play/practice? _______ 3. How old were you when you first received organized football practice (formal coaching with a coach)?  _______ 4. Apart from P.E. lessons, how many sports did you participate in between age 5 and 12 that was led by a coach?_______ 5. Apart from P.E. lessons, how many sports do participate in now that are led by a coach?______   226  SECTION B: Practice History in Football  As accurately as possible, try to recall and write down an average of how often (sessions per week) and how much time (hours per week) you spent in organized football practice, play (i.e. street football) and playing matches. Write down each number underneath each of the age categories. An example, in grey, can be found in the table below.  If you have taken a significant break from football at some point in your career (e.g., due to injury, long-term illness, etc.) then please enter the number of weeks that you took off in that year.  Organized Practice  Practice activities that are conducted with a coach/teacher/adult and are primarily designed to improve skills (formal coaching). This is typically team-led practice and could include things such as football drills, technical skills, tactical skills, strategic skills, coached small- sided games, conditioned games, set-play practices, football-related fitness work etc.  Play (non-organized practice)  Unstructured activities conducted without a coach or teacher. This includes fun games, general kick around, pick-up games, individual play/practice, keep-ups etc.  Match-play  Playing competitive matches against another team or playing uninterrupted matches against other players in your team/club (bounce game)                     227    EXAMPLE: Sport Name Football  Age (years) =  5-6  7-8  9-10  11-12  13-14  15-16  17 Organised Practice Number of sessions per week =  1  2  3  4  4  6  10   Average length of session (hours) =  1  1 ½  2  1 ¾  2  2  2  Play Number of hours per week =  6  10  10  6  6  4  3  Match play Number of hours per week =  1  1  1  1 ½  1 ½  2  3  Significant breaks from football (through injury, or long term illness, exclude holiday & off season) Number of weeks off =  0  0  6  0  12  2  0          228                START HERE     Sport Name Football  Age (years) =  5-6  7-8  9-10  11-12  13-14  15-16  17 Organised Practice Number of sessions per week =        Average length of sessions (hours) =        Play Number of hours per week =         Match play Number of hours per week =         Significant breaks from football (through injury, or long term illness, excludeholiday & off season) Number of weeks off =                229  SECTION C: Football Enjoyment Questionnaire (part a)  Please complete the following questionnaire by circling the appropriate number using a rating scale from 1-7, where 1 = Not at all true, 4 = somewhat true and 7 = very true. For each of the statements circle only ONE of the numbers in each row. Please make sure you answer this on your own.  1= Not at all true, 4 = Somewhat true, 7 = Very True  1             2           3             4             5             6            7  I participate in football because I enjoy it. 1 2 3 4 5 6 7 I participate in football because it’s part of who I am. 1 2 3 4 5 6 7 I participate in football because the benefits of football are important to me 1 2 3 4 5 6 7 I participate in football because I would feel ashamed if I Quit 1 2 3 4 5 6 7 I participate in football because if I don’t other people will not be pleased with me 1 2 3 4 5 6 7 I participate in football but I wonder what’s the point 1 2 3 4 5 6 7 I participate in football because I like it 1 2 3 4 5 6 7 I participate in football because it is an opportunity to just be who I am 1 2 3 4 5 6 7 I participate in football because it teaches me self-discipline 1 2 3 4 5 6 7 I participate in football because I would feel like a failure if I quit 1 2 3 4 5 6 7 I participate in football because I feel pressure from other people to play 1 2 3 4 5 6 7   230      I participate in football but I question why I continue 1 2 3 4 5 6 7 I participate in football because it’s fun 1 2 3 4 5 6 7 I participate in football because what I do in football is an expression of who I am 1 2 3 4 5 6 7 I participate in football because I value the benefits of football 1 2 3 4 5 6 7 I participate in football because I feel obligated to continue 1 2 3 4 5 6 7 I participate in football because people push me to play 1 2 3 4 5 6 7 I participate in football but the reasons why are not clear to me anymore 1 2 3 4 5 6 7 I participate in football because I find it enjoyable. 1 2 3 4 5 6 7 I participate in football because it allows me to live in a way that is true to my values 1 2 3 4 5 6 7 I participate in football because it is a good way to learn things which could be useful to me in my life 1 2 3 4 5 6 7 I participate in football because I would feel guilty if I quit 1 2 3 4 5 6 7 I participate in football to satisfy people who want me to play 1 2 3 4 5 6 7 I participate in football but I question why I am putting myself through this 1 2 3 4 5 6 7    1=Not at all true4 = Somewhat true, 7 = Very True      231  B.5: Player skill ratings  My Skill ratings  1. In comparison to other players in your team, rate your current tactical, technical, creative and physical skill by circling the appropriate number (1 = Poor, 2 = Below average, 3 = Average, 4 = Above average, 5 = Excellent).  1 = Poor, 2 = Below average, 3 = Average, 4 = Above average, 5 = Excellent.  Tactical Skill (i.e. decision making/pass selection) 1 2 3 4 5 Technical Skill (i.e. passing, shooting, dribbling,) 1 2 3 4 5 Physical Skill (i.e. endurance, physical condition) 1 2 3 4 5 Creative Skill (i.e. unexpected, original and useful) 1 2 3 4 5  2. In comparison to other players of your age (e.g. school mates, relatives) who play football, rate your current  tactical, technical, physical and creative skill by circling the appropriate number (1 = Poor, 2 = Below average,  3 = Average, 4 = Above average, 5 = Excellent).   1 = Poor, 2 = Below average, 3 = Average, 4 = Above average, 5 = Excellent  Tactical Skill (i.e. decision making/pass selection) 1 2 3 4 5 Technical Skill (i.e. passing, shooting, dribbling,) 1 2 3 4 5 Physical Skill (i.e. endurance and physical condition) 1 2 3 4 5 Creative Skill (i.e. unexpected, original and useful) 1 2 3 4 5 Thank you very much for completing this questionnaire!   232  B.6: Follow-up player questionnaire                 “The Role of Developmental Activities on Motivation, Passion and Skill in Youth Soccer Players”    The purpose of this questionnaire is to find out information on sporting participation, motivation, and passion from elite youth football players and coaches ranging from 12 to 17 years of age. Other parts of the questionnaire relate to passion and desire as well as technical, tactical, creative and physical skill level. Questions are asked about match-play, organised practice and play-related activities, and the amount of football specific practice in comparison to other sports.  Organised Practice includes: Practice activities that are conducted with a coach/teacher/adult that are used mainly to improve skills (formal practice). This is typically team-led practice and could include things such as football drills, technical skills, conditioned games, tactical skills, strategic skills, set- play practices, and football-related fitness work.  There are 4 sections in total (A-D), you can answer them in any order. You do not have to complete the questionnaire if you do not want to and may take a questionnaire and return it blank. If you decide to take part then please complete the questionnaire to the best of your ability. It will take around 30 minutes to complete. By completing the questionnaire you are agreeing to take part in the study. If you need help answering or understanding the questions please ask one of the assistants. Try to answer as best as you can remember or as best as you think (not someone else). Please note that all information will be treated in strictest confidence. Only those directly involved in the study (that is, the researchers and not the coaches) will have access to the information that you give in this questionnaire. The coaches will only have access to general information that does not identify you or any other players. If you have any question about this questionnaire please contact David, the researcher running the study.    233    SECTION A: General Information  Please fill in the details below: Name -  Today’s date -  Current age -  Date of Birth -  Current club - Age group -   1. How old were you when you first started playing football?  _______ 2. During your first year, how many hours per week did you play/practice? _______ 3. How old were you when you first received organized football practice (formal coaching with a coach)?  _______ 4. What age group were you when you joined the Academy system?  _______ 5. What age group were you when you joined this current Academy?  _______ 6. If you have left the Academy system for any period of time please give the age range when this occurred? _________ 7. Apart from P.E. lessons, how many sports did you participate in between age 5 and 12 that was led by a coach?_______    8. Apart from P.E. lessons, how many sports do participate in now that are led by a coach?______    234  SECTION B: Football Activity Questionnaire   As accurately as possible, try to recall and write down an average of how often (sessions per week) and how much time (hours per week) you spent in organised football practice, play (i.e. street football) and playing matches. Write down each number underneath each of the age categories. An example, in grey, can be found in the table below.  If you have taken a significant break from football at some point in your career (e.g., due to injury, long-term illness, etc.) then please enter the number of weeks that you took off in that year.  Organised Practice includes:  Practice activities that are conducted with a coach/teacher/adult and that are primarily designed to improve skills (formal coaching). This is typically team-led practice and could include things such as football drills, technical skills, tactical skills, strategic skills, coached small- sided games, conditioned games, set-play practices, football-related fitness work etc.  Play (non-organised practice) includes: Unstructured activities that are not conducted with a coach or teacher. This includes fun games, general kick around, pick-up games, individual play/practice, keep-ups etc.  Match-play includes: Playing competitive matches against another team or playing uninterrupted matches against other players in your team/club (bounce game)    235  EXAMPLE: Time period Age (years) = This year Age = 13 Last year Age = 12 Year before last Age = 11 Organised Practice Number of sessions per week =  4  3  3   Average length of session (hours) =  1 1/2  1 ½  2  1 1/2   1 1/4  Play Number of hours per week =  6  8  10  Match play Number of hours per week =  1 1/2  1  1  Significant breaks from football (through injury, or long term illness, exclude holiday & off season) Number of weeks off =  5  0  0     236         START HERE     Time period Age (years) = This year Age =  Last year Age =  Year before last Age =  Organised Practice Number of sessions per week =      Average length of session (hours) =     Play Number of hours per week =     Match play Number of hours per week =     Significant breaks from football (through injury, or long term illness, exclude holiday & off season) Number of weeks off =        237  SECTION C: Football Enjoyment Questionnaire (part a)  Please complete the following questionnaire by circling the appropriate number using a rating scale from 1-7, where 1 = Not at all true, 4 = somewhat true and 7 = very true. For each of the statements circle only ONE of the numbers in each row. Please make sure you answer this on your own. 1= Not at all true, 4 = Somewhat true, 7 = Very True  1             2           3             4             5             6            7  I participate in football because I enjoy it. 1 2 3 4 5 6 7 I participate in football because it’s part of who I am. 1 2 3 4 5 6 7 I participate in football because the benefits of football are important to me 1 2 3 4 5 6 7 I participate in football because I would feel ashamed if I Quit 1 2 3 4 5 6 7 I participate in football because if I don’t other people will not be pleased with me 1 2 3 4 5 6 7 I participate in football but I wonder what’s the point 1 2 3 4 5 6 7 I participate in football because I like it 1 2 3 4 5 6 7 I participate in football because it is an opportunity to just be who I am 1 2 3 4 5 6 7 I participate in football because it teaches me self-discipline 1 2 3 4 5 6 7 I participate in football because I would feel like a failure if I quit 1 2 3 4 5 6 7 I participate in football because I feel pressure from other people to play 1 2 3 4 5 6 7                                                                                                                  Not at all true, 4 = Somewhat true, 7 = Very True              238   1          2          3           4           5           6           7   I participate in football but I question why I continue 1 2 3 4 5 6 7 I participate in football because it’s fun 1 2 3 4 5 6 7 I participate in football because what I do in football is an expression of who I am 1 2 3 4 5 6 7 I participate in football because I value the benefits of football 1 2 3 4 5 6 7 I participate in football because I feel obligated to continue 1 2 3 4 5 6 7 I participate in football because people push me to play 1 2 3 4 5 6 7 I participate in football but the reasons why are not clear to me anymore 1 2 3 4 5 6 7 I participate in football because I find it enjoyable. 1 2 3 4 5 6 7 I participate in football because it allows me to live in a way that is true to my values 1 2 3 4 5 6 7 I participate in football because it is a good way to learn things which could be useful to me in my life 1 2 3 4 5 6 7 I participate in football because I would feel guilty if I quit 1 2 3 4 5 6 7 I participate in football to satisfy people who want me to play 1 2 3 4 5 6 7 I participate in football but I question why I am putting myself through this 1 2 3 4 5 6 7   239  B.7: Coach questionnaire and skill ratings  “The Role of Developmental Activities on Self-Determined Motivation, Passion and Skill in Youth Soccer Players”   The purpose of this questionnaire is to acquire information on sporting participation, skill, motivation, and passion from elite youth football players and coaches. Information will be collected from coaches and a number of different age groups players, ranging from 12 to 17 years of age. Many questions are asked about the type of activities that players are involved in during typical training sessions. Other aspects of the questionnaire relate to passion, motivation and a rating of each players technical, tactical, creative and physical skill level.   Please complete the questionnaire to the best of your ability. Your answers will provide information as to key psychological aspects and practice habits in the development of skilled performance. You do not have to complete the questionnaire and may stop filling out the questionnaire at any point. The questionnaire will take approximately 15 minutes to complete and by completing the questionnaire you are providing your consent. Please note that all information will be treated in strictest confidence. Only those directly involved in the study will have access to information given in this questionnaire and at no time will the information you provide be made available to any person in your Club. We will never disclose personnel or identifying information about individuals, only group-based, summary data will be made available to all interested persons. If you have any queries regarding this questionnaire please contact David Hendry from the Motor Skills Lab at the University of British Columbia via email at hendrydt@interchange.ubc.ca or the Principal Investigator, Dr Nicola Hodges (nicola.hodges@ubc.ca: http://msl.kin.educ.ubc.ca/). Many thanks for your participation.     240  SECTION A: Coaches Demographic Information   Name:_______________________________   Email address:___________________________________________________________   Coaching Qualifications held:   ____________________________________________________              ____________________________________________________        ____________________________________________________        ____________________________________________________   Other relevant qualifications    ____________________________________________________        ____________________________________________________        ____________________________________________________       Number of years involved in coaching youth football      _______________years.  Number of years at your current club        _______________years.    Number of years coaching current age group squad      _______________years      Did you play professional soccer? (circle as appropriate)  yes / no  As a player, what was the highest level of soccer play  that you reached (i.e. amateur, international)        ____________________     As a player what position did you normally play?             ____________________     241  SECTION B: Technical, Tactical, Physical Skill Ratings  We would like you to rate the players who you are currently coaching in terms of their current tactical, technical, physical and creativity skill level.  To do this, please first list all the players names in the Table below under the heading “tactical skill” with their initials in brackets in the first table (you can list them in any order). In subsequent tables, all you need to do is enter the players in the same order, but this time, just give their initials.  Please use a 5 point scale to rate each player with respect to each of these 4 “skills”. When rating a player, please rate them in comparison to other players in the team, where 1 = Poor, 2= Below average, 3 = Average, 4 = Above average and 5 = excellent.  242  Tactical Skill Tactical skills are defined by the player’s ability to make fast and accurate decisions with respect to picking out open players, reading the game well, smart playing, good pass selection/decisions.   When rating a player TACTICAL skill ask yourself “in comparison to his team mates, how well is this player able to read the game and make good decisions?”    1 = Poor, 2 = Below average, 3 = Average, 4 = Above average, 5 = Excellent  Player Name and Initials 1 2 3 4 5 1.      2.      3.      4.      5.      6.      7.      8.      9.      10.      11.      12.      13.      14.      15.      16      17.      18.      19.      20.        243  Technical Skill Technical skills are defined by skills such as the ability of the player to pass accurately, dribble well with the ball, shoot, perform accurate and effective throw-ins/free-kicks etc.   When rating a player in terms of their TECHNICAL skill ask yourself “in comparison to his team mates, how technically skilled is this player?”   1 = Poor, 2 = Below average, 3 = = Average, 4 = Above average, 5 = Excellent  Player Name and Initials 1 2 3 4 5 1.      2.      3.      4.      5.      6.      7.      8.      9.      10.      11.      12.      13.      14.      15.      16      17.      18.      19.      20.       244  Physical Skill Physical skills are defined by a person’s overall physical condition.  When rating a player in terms of their PHYSICAL skill ask yourself “in comparison to his team mates, how physical fit and/or fast is this player?”    1 = Poor, 2 = Below average, 3 = = Average, 4 = Above average, 5 = Excellent   Player Name and Initials 1 2 3 4 5 1.      2.      3.      4.      5.      6.      7.      8.      9.      10.      11.      12.      13.      14.      15.      16      17.      18.      19.      20.        245  Creativity Creativity is defined by a person’s overall flair and originality in making decisions and displaying unusual skills and effective creative plays  When rating a player in terms of their CREATIVE skill ask yourself “in comparison to his team mates, how creative is this player on the ball and in making original decisions?”   1 = Poor, 2 = Below average, 3 = = Average, 4 = Above average, 5  = Excellent   Player Name and Initials 1 2 3 4 5 1.      2.      3.      4.      5.      6.      7.      8.      9.      10.      11.      12.      13.      14.      15.      16      17.      18.      19.      20.      Thank you for your help!      246  Appendix C: Professional status contact letters  C.1: Adult professional status email  Dear Craig,  Thank you again for your continued participation in our study investigating the factors associated with developing elite soccer players. As the final part of our study we are now attempting to obtain information as to those players that successfully transitioned from receiving a youth professional contract at approximately age 16, and play in your respective first team as of September 1st, 2016. Therefore, could you please pass on the names and DOB’s of any players that successfully made this transition. As I understand from our last conversation, there was a small number of players that were released from Glasgow Rangers F.C. that subsequently received youth contracts at other clubs. Could you please pass their information also?  Many thanks,  David     247  Appendix D: Study 4 question package  D.1: Recruitment letter  Developmental Football Activities: Skill, Creativity and Passion  Principal Investigator:  Co-Investigator: Nicola Hodges, Ph.D.   David T Hendry School of Kinesiology  School of Kinesiology University of British Columbia University of British Columbia nicola.hodges@ubc.ca  hendrydt@interchange.ubc.ca  Dear Marissa,  As you are aware I am conducting some research on the development of elite women soccer players as part of my PhD Thesis. Along with Dr Nicola Hodges, from the Motor Skills Laboratory at UBC, we have collected data to explore the developmental soccer activities engaged in and psychological characteristics of these elite female soccer athletes. As part of this research we wish to gain access to your players at UBC women’s soccer to administer a series of questionnaires. These questionnaires have been designed to take approximately 60 minutes to complete and they can be completed either before or after practice or competition. A brief rationale for the proposed study can be found below. By consenting to allow us to administer questionnaires does not mean that you consent for all parties to participate. Individual consent will be sought from the players as evidenced by their willingness to complete the questionnaires.  Study proposal We aim to assess the early sport activity experiences of women soccer players and its relation to skill, success and psychological indices of passion, desire and motivation. Through specifically designed, validated questionnaires, we will investigate the relationships between early sport experiences in general, engagement in football-specific activities, (i.e. street soccer and organized practice) and current levels of passion, motivation and self-regulation. These analyses will form part of a wider understanding of optimal youth sport development and the benefits and costs of early sport-specific specialization. Because ourselves and colleagues in the UK have collected data from international level athletes we will be able to assess differences across skill levels. We expect that the results of our study will contribute to the existing body of literature on the development of expertise in soccer. In a practical sense, these findings will help foster successful and positive youth sport development in soccer, particularly in women’s soccer, with research-based evidence serving to guide the design of effective practice environments. All proposed measurement scales have been verified and deemed appropriate by the research ethics’  248  board at the University of British Columbia. Moreover any information collected will also be held in the strictest of confidence and no specific details allowing identification of individuals will be included in subsequent publication or presentation. General results will be made available to interested persons, but no information will be provided that will serve to identify individual coaches, parents or players. If you /your club are willing to participate in the study please confirm by emailing me at david_t_hendry@yahoo.co.uk (phone: 778-318-5200). Data collection is planned to take place anytime from September 11th – December 2015, although specific times/dates can be negotiated. In view of these time lines, please email me as soon as possible so I can make plans with people involved.  Nothing is required from you at this stage and we will discuss requirements after initial contact. I will of course try to make this as easy on all involved as possible. Please do not hesitate to contact me via email if you require any further information.   Kind Regards David Hendry Nicola Hodges Selected references from our Lab. • Ford P., NJ Hodges and AM Williams (2013). “Creating champions: The development of expertise in sports”. Beyond Talent or Practice: The Complexity of Greatness (pp391-413). Ed. S.Kaufman. Oxford University Press. • Hendry D, P Crocker, Hodges NJ (2014). Practice and play as determinants of self-determined motivation in youth soccer players. Journal of Sports Science, 32(11):1091-1099.  DOI:10.1080/02640414.2014.880792 • Hodges, NJ. and J. Baker (2011). "Expertise: The goal of performance development". Performance Psychology: A practitioner’s guide. Ed. D Collins, A Button and H Richards, Oxford, Elsevier Publishers, (pp 31-46). • Ward, P, NJ Hodges, AM Williams and JL Starkes (2007). "The road to excellence in soccer: A developmental look at deliberate practice". High Ability Studies, 18 (2): 119–153. • Williams, AM and NJ Hodges (Eds., 2004). Skill Acquisition in Sport: Research, Theory and Practice. London, UK: Routledge (Taylor & Francis Group) (new release due in 2012). • Williams, A.M. and N.J. Hodges (2005). "Practice, instruction and skill acquisition in soccer: Challenging tradition."  Journal of Sports Sciences, 23: 637 - 650.   249  D.2: Women’s soccer questionnaire Developmental Activities in Soccer  The primary purpose of this questionnaire is to find out information on sporting participation, practice and motivation from elite youth and adult, male and female soccer players. Questions are asked about match-play, organised practice and play-related activities, and the amount of soccer specific practice in comparison to other sports. We in the Motor Skills Lab. in the School of Kinesiology at UBC are conducting this study as part of 2 larger projects designed to (i) ascertain information on the developmental histories of National-level women athletes in soccer, across a range of countries and (ii) determine how different types of practice activities relate to success in soccer and motivation, among youth, adult, male and female athletes. The first part of the study is being conducted in conjunction with researchers from Liverpool John Moores University in the UK.  Organised Practice includes: Practice activities that are conducted with a coach/teacher/adult that are used mainly to improve skills (formal practice). This is typically team-led practice and could include things such as soccer drills, technical skills, conditioned games, tactical skills, strategic skills, set- play practices, and soccer-related fitness work.  There are 4 sections in total, you can answer them in any order (although the information in the first 2-3 sections is the most critical). You do not have to complete the questionnaire if you do not want to and you may take a questionnaire and return it blank. If you decide to take part then please complete the questionnaire to the best of your ability. It will take about 60 minutes to complete (depending on your age/experience). By completing the questionnaire you are agreeing to take part in the study. If you need help answering or understanding the questions please ask one of the assistants. Try to answer as best as you can remember or as best as you think (not someone else). Please note that all information will be treated in strictest confidence. Only those directly involved in the study (that is, the researchers and not the coaches) will have access to the specific information that you give in this questionnaire. The coaches will only have access to general information that does not identify you or any other players. If you have any question about this questionnaire please contact Dr Nicola Hodges (nicola.hodges@ubc.ca), the researcher running the study.  250  1. ‘Milestones’  What is your name (your name will be kept completely confidential and you will be assigned an alphanumeric code to identify your country, and skill level)? ___________________________________  What is your date of birth and town/city and country of birth? __________________________________  Primary playing position (please circle):    Goalkeeper    /     Defender    /    Midfield    /    Forward   Soccer-specific milestones  ___ years old when you first started playing soccer (of any type) ____ have never done it  ___ years old when you first took part in supervised practice/training by an adult in soccer ____ have never done it  ___ years old when first played in a girl’s organized soccer league ____ have never done it  ___ years old when first played in a boy’s or co-rec organized soccer league ____ have never done it   ___ years old when you stopped playing in a boy’s/ co-rec. organized soccer league ____ have never done it        years old when first took part at youth level of a club with a semi pro/pro team         ____ have never done it    ___ years old when first began regular fitness/ non-soccer training (eg. running, strength, etc)   ____ have never done it  How many hours/wk did you spend on fitness training when you first began to do this regularl _____ hrs/week (typical)  How many hours/wk do you currently spend on fitness training (outside of formal practice)     _____ hrs/week (typical)  ___ years old when first began regular Performance Analysis (PA) /self-video feedback        ____ have never done it    How many hours/wk did you spend on PA when you first began to do this regularly       _____ hrs/week (typical)  How many hours/wk do you currently spend on PA (outside of formal practice)?        ______ hrs/week (typical)        years old when first took part at youth international level               ____ not applicable       ____ have never done it        years old when first took part at Olympic development level         ____ not applicable       ____ have never done it        years old when first took part at senior level for semi pro/pro team            ____ have never done it        years old when first took part at international level                ____ have never done it        years old when first took part at Olympic games                  ____ have never done it        international appearances made (approximate if not available)                       ____ have never done it  How many hours/week do you currently take part in organized soccer practice?     _______ hrs/week (typical)     251  Please list your senior honours (best achievements) and year here (or over) (e.g., Olympic silver medal - 2012, Premier League Champion - 2010)      252  2.  Engagement in soccer-related activities (a = Adult; b = Youth)  The following section focuses on the soccer-related activities you have participated in from when you began playing to the present day, the number of hours spent in these activities per week, and the number of months per year you spent in each of the activities. This will be done for each year you have participated.   Please group the activities you have participated in into the categories listed below:  1. Match-play/games:  Organized competition in a group engaged in with the intention of winning and supervised by adult(s), e.g. league games.  2. Coach-led group practice:  Organized group practice engaged in with the intention of performance improvement and supervised by coach(es) or adult(s),  e.g. team practice.  3. Individual practice:  Practice alone engaged in with the intention of performance improvement, e.g. practicing dribbling skills alone. This is not fitness work.  4. Peer-led play:  Play-type games with rules supervised by yourself/peers and engaged in with the intention of fun and enjoyment, e.g. game of soccer with friends.     253  Overleaf there are 2 ‘participation history’ tables, which list these four categories and groups them into years. There is a separate table for a) Adult (22 yr & older) and b)Youth years (21 yr & younger). Please fill these in as accurately as possible, starting from this year (i.e., 2013/2014) and working downwards until you have completed the first year you played soccer (just alternate years). <Age groups refer to your actual age in that season, not the team you played for if different>  For each year, please complete: 1a. Number of hours spent taking part in activities related to each category. 1b. Number of months of the year that you spent taking part in activities related to each category <A soccer season usually equals 9 months, although this may vary between countries> 2. How Challenging you found each activity, where challenge would be seen as an activity that continually tests your abilities, that is demanding and/or stimulating. Please rate from 0-4: 0 = Easy/Not challenging 1 = Some/Low challenge 2 = Moderate challenge 3 = High challenge 4 = Too much/Extremely challenging  3. The number of weeks from the relevant year that you were injured and unable to take part in the soccer activity. Leave blank if no injury and only fill this in once for each year. Table a) overleaf pertains to Adult Activity. To save time, we only ask you to fill in each second year. First, add the year that you were U22 in the bottom row year column. Second, work downwards from the current year (2013/14) adding your age every second year in the year column until you reach U22 <you likely will not use all the rows and do not repeat information>. Please then write the name of your primary coach and team you played for in each season in the space provided (to aid recall). Third, fill in all other columns.    254  Table a) ADULT YEARS Year & Age (yr)  Team and coach  Activities # of hrs/wk Months /yr Challenge level  (0-4)  Injury wks/yr     1. Match-play /game 2 9 3 3 2013/14 John Smith 2. Coach-led practice 5 9 2  29-30yr Birmingham Rovers  <example> 3. Individual practice - self 2 12 3    4. Peer-led play 5 12 1    1. Match-play /game     2013/14  2. Coach-led practice        3. Individual practice     ------ yr  4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%   1. Match-play/game     2011/12  2. Coach-led practice        3. Individual practice     ------ yr  4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%   1. Match-play/game      2009/10  2. Coach-led practice        3. Individual practice     ------ yr  4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%   1. Match-play/game      2007/08  2. Coach-led practice        3. Individual practice     ------ yr  4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%   1. Match-play/game      2005/06  2. Coach-led practice        3. Individual practice     ------ yr  4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50% 2003/04 ------ yr   1. Match-play/game      2. Coach-led practice      3. Individual practice     4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50% _______   21-22 yr    This is the year you were U22, please add year. 1. Match-play/game      2. Coach-led practice      3. Individual practice     4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%   255  Adult Activity: Please list here any other season(s) when you did not play soccer because of injury or other circumstances (not listed above) and whether it was the whole season or ~half a season:  Whole season missed: ___________________Half a season missed: ___________________      The next table is designed to record your Youth Activity. Please complete this in the same way as for the Adult activity table above and stop when you reach the age you started to play soccer. If you started at U8, U10, U12 (or any age not listed), please cross out the box below this age group (e.g., U7 if you started at U8) and add this information for your start year. We appreciate it will get harder to recall these activities as you think further back, but as best as you can, please try and fill in all columns.   Youth Activity: After completing the table overleaf, please list here any other season(s) when you did not play soccer because of injury or other circumstances (not listed in the next table) and whether it was the whole or ~half a season:  Whole season missed: ___________________Half a season missed: ___________________     256  Table b) YOUTH YEARS Year & Age (yr)  Team and coach  Activities # of hrs/wk Months /yr Challenge level  (0-4) Injury wks/yr   1. Match-play /game     U21  2. Coach-led practice        3. Individual practice     20-21yr  4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%   1. Match-play/game     U19  2. Coach-led practice        3. Individual practice     18-19 yr  4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%   1. Match-play/game      U17  2. Coach-led practice        3. Individual practice     16-17 yr  4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%   1. Match-play/game      U15  2. Coach-led practice        3. Individual practice     14-15 yr  4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%   1. Match-play/game      U13  2. Coach-led practice        3. Individual practice     12-13 yr  4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%  U11  10-11 yr   1. Match-play/game      2. Coach-led practice      3. Individual practice     4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%  U9  8-9 yr    1. Match-play/game      2. Coach-led practice      3. Individual practice     4. Peer-led play     Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%  U7  6-7 yr    1. Match-play/game      2. Coach-led practice      3. Individual practice     4. Peer-led play      Please rate yourself (circle) in relation to others in your primary team: Top:         10%,    25%,     50%  <based on overall skill/ability/contribution>                                           Bottom:   10%,    25%,     50%  257   U6  5-6 yr    1. Match-play/game      2. Coach-led practice      3. Individual practice     4. Peer-led play       3.  Engagement in other sport activities  The following Table overleaf focuses on the other sporting activities you have engaged in, the period of your life in which you took part in this activity, the number of hours per week, and months per year spent in these activities, and the highest standard of this activity. For each activity, please complete:  1. Please place a tick next to the other sports that you have participated in during your life, outside of timetabled school physical education classes.  <Please only record sports that you engaged in for a total of three months of activity or more>.  2a. The age you started taking part in each activity.   2b. The age you finished taking part in each activity (if you are still participating in an activity, then please indicate this by placing a dash (–) in that box).  3. The total number of hours per week spent taking part in each activity.  4. The number of months of the year in which you took part in each activity.  5. The highest standard of the activity that you took part in for that sport (e.g., school, club, varsity, provincial, national, international).  258   Other sport activities.  Please tick if  yes  Please cross if  no Start age Finish age Total # of hrs/wk Months /yr Highest standard participated at e.g.    Tennis /  12 -  2 8 School Athletics        Badminton        Baseball        Basketball        Boxing/Kick boxing        Canoeing        Cricket        Cycling        Cross country        Gridiron/Football        Gymnastics        Golf        Handball        Hockey - Field        Hockey – Ice        Judo/Karate        Netball        Rugby/Gaelic        Running or jogging        Snooker/Pool        Swimming        Skiing/Snowboarding        Stretching/Yoga/Pilates        Surfing        Table tennis        Tennis        Volleyball        Weights        Other:        Other:        Other:        Other:        Other:        Other:        Other:        Other:        Other:         259   If we need to follow up with you about any of the information contained in this questionnaire or potentially to ask your assistance in collecting data for any further research studies are you willing for us to contact you? □ Yes, I am willing for you to contact me to participate in future studies or to get clarification on any of the information in this questionnaire: My email address is: ____________________________________________________________ □ No, I do not wish for you to contact me again. Motor Skills Lab., School of Kinesiology, UBC http://msl.kin.educ.ubc.ca/   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0365946/manifest

Comment

Related Items