Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

From level to field : factors influencing student choice of undergraduate field of study Breton-Skagen, Camille 2015

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

Item Metadata

Download

Media
24-ubc_2015_november_bretonskagen_camille.pdf [ 903.27kB ]
Metadata
JSON: 24-1.0216544.json
JSON-LD: 24-1.0216544-ld.json
RDF/XML (Pretty): 24-1.0216544-rdf.xml
RDF/JSON: 24-1.0216544-rdf.json
Turtle: 24-1.0216544-turtle.txt
N-Triples: 24-1.0216544-rdf-ntriples.txt
Original Record: 24-1.0216544-source.json
Full Text
24-1.0216544-fulltext.txt
Citation
24-1.0216544.ris

Full Text

    FROM LEVEL TO FIELD: FACTORS INFLUENCING STUDENT CHOICE OF UNDERGRADUATE FIELD OF STUDY by CAMILLE BRETON SKAGEN B.A University Laval, 2013 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Sociology) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) © Camille Breton Skagen, 2015
Abstract  As universities face unprecedented numbers of applicants, competition for access to the more prestigious fields of higher education has become increasingly important. This study focuses on “what one studies” rather than “where one studies” based on how a student’s family background (as measured both by socioeconomic status and ethnicity) and gender influence their choice of undergraduate field of study. This paper addresses two main theoretical traditions: the “liberated” theory and  Effectively  Maintained  Inequality  (EMI).  Whereas  the  latter  suggests  that  social background may actually become more important for later transitions than for earlier ones,  the “liberated” theory emphasizes that as children age, they become more independent of parents and therefore  social  background  effects  are  lower  for  later  transitions  and  less  likely  to  impact student’s  choice  of  field.  The data for this study are drawn from Statistics Canada’s National Graduate Survey (NGS) of 2007 (class of 2005) and 2013 (class of 2009-2010). I conduct logistic and OLS regressions in order to assess the influence of a variety of independent variables, such as socioeconomic status, on the dependent variable “Field of study”, measured using 3 types of dichotomous codings: cultural fields, professional fields, and hard sciences, as well as the categorical coding of field by average income. Findings are not in keeping with the EMI theory, but rather lean towards a more “liberated” theory of choice of undergraduate field of study. Indeed, instead of maintaining the intergenerational transmission of status, as EMI predicts, field of study may well weaken the disadvantageous effect of family background, thereby providing a means for upward social mobility for lower SES students.  !iiPreface This thesis is original, unpublished work by the author Camille Breton Skagen.  !iiiTABLE OF CONTENTS Abstract……………………………………………………………………………………………ii Preface……………………………………………………………………………………………iii Table of contents………………………………………………………………………………….iv List of Tables………………………………………………………………………………………v List of Figures……………………………………………………………………………………vi Acknowledgments……………………………………………………………………………….vii Dedication………………………………………………………………………………………viii Introduction………………………………………………………………………………………..1 Review of Literature………………………………………………………………………………3  Effectively and Maximally Maintained Inequality………………………………………..3  “Liberated” theory………………………………….……………………………..………5  Socio-economic Status (SES)……………………………………………………………..7  Gender and Ethnicity………………………….…………………………………………..9 Research questions and Hypotheses…………………….………………………………….……12 Methods………………………………………………………………………………………….15  Data and Variables.………………………………………………………………………15   Dependent variable: “Field of Study”……………………………………………16 Independent variables: Demographic and Socioeconomic factors………………19Models and Results……..………….…………………………………………………………….24Class of 2005 (Survey of 2007)………………………………………………………….24Class of 2010 (Survey of 2013)………………………………………………………….34Conclusion……………………………………………………………………………………….48 Bibliography……………………………………………………………..………………………55 Appendix I……………………………………………………………………………………….62 Appendix II………………………………………………………………………………………63 Appendix III……………..…………………….…………………………………………………65  !ivList of Tables Table 1 Logistic regression models for hard sciences……………………………………25 Table 2 Logistic regression models for cultural fields……………………………………27 Table 3 Logistic regression models for professional fields……………………………….29 Table 4 OLS regression models for average income per field of study…………………..31 Table 5 Logistic regression models for hard sciences……………………………………35 Table 6 Distribution of field of study by age of respondent for the 2010 cohort…..…….38 Table 7 Logistic regression models for cultural fields……………………………………39 Table 8 Logistic regression models for professional fields……………………………….43 Table 9 OLS regression models for average income per field of study…………………..45  !vList of Figures Figure 1 RESP and government student loans for the 2005 cohort……………………….20 Figure 2 RESP and government student loans for the 2010 cohort……………………….20 Figure 3 Postsecondary savings set aside for children 17 years old or younger,  by household income, 2013…………………………………………………………21  !viAcknowledgments I owe particular thanks to Dr. Neil Guppy and Dr. Elizabeth Hirsh who have inspired and supported me throughout the sometimes challenging adventure that is building a Master’s thesis. I thank Dr. Guppy for the opportunity he has given me in working at this side, and Dr. Hirsh for transmitting her passion of quantitative research.  Very special thanks are owed to my parents who have supported me throughout my many years of education, both morally and financially. They have been there through the tears and the joys, and have never failed to show me how proud they are of my accomplishments.   !viiTo those who, uncelebrated, work in the shadows to make the world a better place.  !viiiIntroduction  As higher education becomes a mass and increasingly universal enterprise, its institutions have become less exclusive, more differentiated, and more internally stratified. Scholars in the sociology of education have traditionally examined factors influencing access to higher levels, or more years, of schooling. This area of research is important since years of schooling and especially higher credentials mean a greater likelihood that an individual can command better occupational careers and higher income (Simpson, 2001).  Recently, however, scholarship has also turned to examining different avenues or tracks within levels of schooling, suggesting that in addition to the level of the credential one attains, the field in which the credential was earned also commands greater payoffs for individuals. Much like universities with the most prestigious reputations offer their graduates access to elite jobs, higher wages, contacts, and other advantages, fields of study differ greatly in their payoffs, with those that have the tightest links to the job market, particularly in the professions, offering the greatest returns (NCES, 2002, p. 446; Fitzgerald, 2000; Walters, 2002; Karen, 2002, as seen in Davies & Hammack, 2005). Field of study  As universities face unprecedented numbers of applicants, competition for access to the more prestigious fields of higher education has become increasingly contested (Davies & Hammack, 2005). Higher education can now be seen as stratified along two main dimensions: selectivity of institution and field of study (Davies & Guppy, 1997). Davies and Hammack (2005) efficiently sum up the distinction between higher education in Canada and the US: Whereas students in the U.S. compete for access to elite colleges, students in Canada compete 
1for elite majors. Where one studies is seen as more important in the U.S., while what one studies is more salient in Canada. Indeed, Canadian students face remarkable stratification within institutions, shifting the importance from ‘where one studies’ to ‘what one studies’ (Davies & Hammack, 2005). For instance, although tuition fees differ only slightly between institutions, Canadian standards and tuition fees are increasingly disparate between programs within institutions; advantageous fields of study are setting their standards and tuition fees apart from the rest of the university (see Canadian Association of University Teachers [CAUT], 2013). At the University of British Columbia (UBC) for instance, student tuition for a one-year full-time course load in the Arts program is $4,988.10; students in Engineering pay $6,151.99; those in Law pay $11,677.12 a year; and those in Medicine are expected to spend $17,065.92 . Similar 1disparities are found in universities across the country. Also in British Columbia, at the University of Victoria, students in Arts pay $5,262 in annual tuition fees, those in Engineering pay $5,822 a year, and those in Law pay $9,209 . The difference between the two universities are 2negligible, at least in comparison to the difference between the fields of study where much wider discrepancies are found. Disciplines are therefore unequal with respect to power, prestige, and economic payoffs (Clark 1983; Hagstrom 1971; Rumberger & Thomas 1993, as cited in Davies & Guppy, 1997; Guppy, Grabb & Mollica, 2013).  University of British Columbia, Undergraduate Tuition Fees, site visited July 2015. http://students.ubc.ca/1enrolment/finances/tuition/undergraduate-tuition-fees University of Victoria, Undergraduate Tuition Fees, site visited July 2015. http://web.uvic.ca/calendar2015-05/2FACS/UnIn/UTanOF/AcDe.html2Review of literature  Effectively and Maximally Maintained Inequality     Over the course of the last century, school attendance has grown such that by the early 1920s almost all children attended elementary school and by the early 1970s the same was true of high school enrolment. Attendance at each level had become universal in that everyone was enrolling in those respective levels of schooling. In recent decades, attendance at the post-secondary level has also become nearly universal (Davies and Guppy, 2014). Due to the increasingly universal access to education and in an attempt to explain the shift from a focus on level to that of field, Effectively Maintained Inequality theory (EMI) suggests that “for levels of education that are universal, competition will occur around the type of education attained” (Lucas, 2001). Therefore, social background will affect differences in kind of schooling. In other words, when a particular level of schooling is not universal, the socioeconomically advantaged secure that level for themselves. However, once that level of schooling becomes nearly universal and easily accessible to the less socioeconomically advantaged students, the upper class seek out whatever qualitative differences there are at that level and use their advantages to secure quantitatively similar but qualitatively better education (e.g. fields of study). This may challenge the long standing argued equality in education. Indeed, as undergraduate enrolments continue to rise, undergraduate field of study has become an “effective” way of differentiating graduates and maintaining inequality (Lucas, 2001). Torche (2011) suggests that, although evidence about the stratification of field of study remains limited, and while the association between social origins and a lucrative major appears to be weak 
3(Davies and Guppy 1997), an indirect influence is likely to exist—upper-class students are more likely to major in sciences and professional fields as opposed to the arts for instance, which in turn increases their chances of pursuing an advanced degree resulting in higher earnings (Goyette and Mullen, 2006 as seen in Torche, 2011). This current research project seeks to determine the extent to which socioeconomic backgrounds as measured by parental level of education and parental income influence a student’s choice of undergraduate field of study.  While we know that socioeconomic background, gender and ethnicity play a significant role in determining which postsecondary institution a student will attend (Scott, Maldonado & Zarifa, 2014) as well as the influence these factors have on the level of education one pursues (Zarifa, 2012; Mullen et al. 2003; Shavit & Blossfeld, 1993; Ethington and Smart 1986), there is a lack of research addressing the influence of family background, gender and ethnicity on the choice of field of study at the undergraduate level. In examining the extent to which the family socioeconomic  status,  student’s  gender,  race  and  age  impact  their  likelihood  of  obtaining  a graduate degree in Canada, Zarifa (2012) reviews a debate relevant to this research, between Maximally Maintained Inequality (MMI) theory and what is referred to in the literature as the “liberated”  theory.  MMI  theory  postulates  that  inequality  persists  as  advantaged  individuals migrate  towards higher  levels  of  credentials  and that  educational  choices  and transitions are based  on  rational  choice  by  the  family  (cost  and  benefit).  MMI  suggests  that  families  and children opt for constancy in the student’s educational path unless forced to change by increasing enrolment, in which case they migrate towards higher credentials (Raftery & Hout, 1993). In other  words,  whereas MMI (Maximal Maintained Inequality) theory contends that inequality persists as advantaged groups, sensing competition for better-paying jobs, migrate to higher 
4levels of credentials once less advantaged groups have better access to lower ranking institutions and programs (due to the widening access to university baccalaureate programs among lower status groups, for example), EMI (Effectively Maintained Inequality) is a process in which widened access to a credential tier encourages privileged groups to migrate toward its most advantageous, selective, and prestigious sectors or fields: “If MMI describes how inequality is maintained through upward movements between credential tiers, EMI describes how inequality is maintained through lateral movements within a tier” (Davies, Maldonado and Zarifa, 2014). “Liberated” theory The liberated theory,  on the other  hand,  views students  as  being liberated from their social origins by virtue of their undergraduate experiences, suggesting that social origins may not yield any direct effect on graduate school decisions. In essence, as students age, mature, move out of the family home, and become more independent, the influence of the family background is thought  to  wain.  Although Zarifa  (2012)  successfully  shows that  parents’ level  of  education influences  graduate  school attendance, in accordance with the MMI theory, he also finds that field of study and region of undergraduate institution play a role in shaping educational outcomes (i.e. graduate attendance), suggesting students may be partially “liberated” from their family backgrounds due to their undergraduate experiences. Despite these interesting results, Zarifa’s population of interest remains that of graduate students.   In keeping with the “liberated” theory, findings  in  stratification  research  show that  the direct influence of parental resources on the socioeconomic position of adult children is much weaker among college graduates than among those with less schooling (Hout 1984, 1988). These 
5findings  mean  that  for  those  who  attain  a  college  degree,  their  socioeconomic  standing  is independent of their socioeconomic background. One way to understand this is as a series of transitions.  In order to attend college or university, students must first enrol in elementary and secondary school, and successfully complete all the transitions entailed in earning a high school diploma. Parental background influences the lower transitions substantially, with children from higher socioeconomic backgrounds being more likely to succeed. This has a cumulative effect such that socioeconomic status (SES) influences attainment of ever higher levels of schooling. However, the influence of parental background on the higher level transitions might be weaker than  lower  level  transitions.  It  may  be  that  SES  doesn’t  have  as  much  effect  on  winning scholarships or effecting field of study choices, at least in comparison to its effect on completing high school, for example. The effect of parental SES may also wane over time due to children trying to gain independence and parents simply having less control over offspring as the latter leave  home  (in  some  cases  to  institutions  in  different  provinces  or  countries),  form  new important relationships, and eventually start new families. These views reflect the debate over the waning influence of social class on educational choices and outcomes. Whereas some authors argue that social class remains salient in terms of students’ broad social outcomes, including educational attainment (Ball, S. J. et al.1996; Reay, D. et al., 2001; Ball, S. J. et al., 2002; Ball, S. J.,  2010)  others  deem  class  to  be  irrelevant  in  understanding  such  contemporary  social phenomenon (Clark, T. N., & Lipset, S. M., 1991; Beck, U., & Beck-Gernsheim, E., 2002). In fact, previous research shows that socioeconomic status plays a negligible role in the selection of fields  that  lead  to  high  incomes  after  graduation  (Davies  &  Guppy,  1997)  or  high  status occupations (Wilson, 1978). This study will  allow for a better understanding of the changing influence of class, among other factors, on students’ choice of undergraduate field of study. 
6 In light of the above, the current study will focus on two main theoretical traditions: the “liberated” theory and Effectively Maintained Inequality (EMI). Whereas the latter suggests that social background may actually become more important for later transitions than for earlier ones, the “liberated” theory emphasizes that as children age they become more independent of parents and therefore social background effects are lower for later transitions and less likely to impact student’s choice of field. Socio-economic Status (SES)In her study of the influence of student’s SES on choice of college major, Yingyi (2009) found  that,  although  a  college education is generally regarded as an important opportunity for young people’s future, for students from lower SES families, the opportunity to attend college would be a much more cherished one compared to their higher SES counterparts, for whom college may be the natural stage of their schooling career. Lower SES children and their families would be more motivated to make sure that their college education is the surest route to success. Majors that provide good job opportunities and high financial returns are promising choices for lower SES students to maximize the returns on their education and minimize the risks of their investments. This finding supports the Relative Risk Aversion theory, in which risk is defined in this context as uncertain job prospects, including few job opportunities and low financial returns generally associated with certain college major fields.  Lucas (2001) and Van de Werfhorst (2001) address the question of field of study at the undergraduate level. Lucas’s results with students transitioning from grade 12 to college are consistent with MMI theory and furthermore prove EMI to be accurate. Results suggest that the 
7effects of social background (1) determine who completes a level of education if completion of that level is not nearly universal, and (2) determine the kind of education persons will receive within levels of education that are nearly universal. Either way, social background advantages seem to work to effectively and continuously secure for children their own privileged status.  Van de Werfhorst (2001) studies the impact of the family background on choices for fields of study as well as the value of specific investments in a field of study on the labour market and observed outcomes, seeking to examine the extent to which family background explains choice of field of study. He distinguishes four types of resources that are differentially acquired in the various fields of study: cultural, economic, communicative and technical resources. He examines four hypotheses that subsequently prove to be accurate: 1) children are likely to choose the same field of study as their father; 2) children of the economic elite (i.e., parents with higher levels of material possessions) are likely to choose an economically oriented field of study; 3) children of the cultural elite (i.e., a high level of cultural participation of parents) are likely to choose a cultural field of study; and 4) children from low SES backgrounds (low social class and low educational level of the father) are likely to choose a technical field of study. Results from the present research will shed light on the relationship between SES background and the type of academic field of study. Although Van de Werfhorst examines the topic of the present research, he does so by using the The Netherlands Family Surveys of 1992 and 1998. The particular contribution of the proposed study is found in its Canadian focus and its assessment of two much more recent cohorts (class of 2005 and class of 2010). 8Gender and Ethnicity  Although EMI and the “liberated” theory speak to socioeconomic status,  they do not directly address the issue of gender variations in choice of field of study. The majority of college and university graduates in the United States and most other industrial countries are now women (Schofer and Meyer 2005; Buchmann and DiPrete 2006; Shavit, Arum, and Gamoran 2007), and although this may give the impression of a progressive elimination of gender differentiation among higher education, some argue that it is not the case. Gendered patterns in choice of field of study have been extensively documented (Jacobs, 1986, 1995). Men have traditionally chosen fields such as business, engineering, and hard sciences while women have studied education, humanities, nursing and psychology (Goyette and Mullen, 2006). In 1999-2000, among recipients of bachelor’s degrees in the United States, 13 percent of women majored in education compared to 4 percent of men, and only 2 percent of women majored in engineering compared to 12 percent of men (Zafar, 2013). Andres and Adamuti-Trache (2007) show that over 25 years (1979-2004), gender segregation among undergraduate enrolment has decreased by only 5 percent. Indeed, increases in enrolment and completion by women have not been translated into gender integration within most fields of study, suggesting that different structures of opportunities within higher education continue to perpetuate gender inequities in the labour market (see also Davies and Guppy, 2013: 140-1). Shauman (2006) suggests that women’s segregation in college fields of study directly contributes to their segregation in later occupations, noting that sex differences in college major explain 11 to 17 percent of the sex gap in the likelihood of employment in relatively high-paying occupations. 
9 Although most research addresses aspects of education in which women trail men, such as gender segregation in majors (Charles and Bradley 2002; Jacobs 1995; Turner and Bowen 1999), women’s underrepresentation at top-tier institutions (Jacobs 1999), and their underrepresentation in science and engineering (Fox 2001; Long 2001; Xie and Shauman, 2003) some suggest that an array of other factors including differences in preferences, labor market expectations, and gender-specific effects of the college experience account for the main part of today’s gender gaps in choice of academic major (Turner and Bowen, 1999). Among others are an individual's preferences for various courses of study, which may be encouraged by parental and societal expectations; and the labor market prospects associated with a given set of skills, which may provide more encouragement for one sex than the other to pursue certain fields of study. Zafar (2013) for instance, suggests that males and females primarily differ in their preferences for the workplace outcomes, with females valuing less money-related aspects of the job and more reconciling work and family, and enjoying work relatively more than males. This may explain, to a certain extent, differences in choice of field of study.   Charles and Bradley (2009) argue that contrary to what would be expected as for the “degendering” of curricular fields within modern societies, sex segregation of academic fields is greater in more economically developed societies.  Authors of “Indulging Our Gendered Selves? Sex Segregation by Field of Study in 44 Countries”, Charles  and  Bradley  find that sex segregation by field of study has not waned as would be expected, suggesting that enduring cultural forces of gender-essentialist ideology (i.e., cultural beliefs in fundamental and innate gender differences) are still strongly at work. These cultural ideologies are said to be deeply influential in “shaping life experiences, expectations, and aspirations, even in the most liberal 
10egalitarian societies”. The analysis of gendered choices of field of study in this research will allow me to assess the accuracy of such “degendering” of curricular fields in Canadian undergraduate programs and institutions.   Finally, if class and gender influence choice of field, it would be expected that ethnicity also impact students choice of academic program. Simpson (2001) found that in the United States, European Americans do not significantly differ in their choice of baccalaureate academic degree from African Americans, Hispanic Americans, and Native Americans. Instead, the significant differences in choice of degree program occur between Asian Americans and non-Asians. Considerable attention has been given to the factors predicting the selection of science majors, and studies have found that women and non-Asian minority students are under-represented in these fields compared to their white male counterparts (Maple and Stage, 1991; Mullen, 2001). Xie and Goyette (2003) found that ethnicity does in fact influence choice of a more profitable field of study. Specifically among Asian-Americans, many of whom are recent immigrants, and may choose fields with higher possible earnings as a way to ensure upward mobility.   In addition, Asian Americans are said to have higher educational expectations than whites. Goyette and Xie (1999) suggest that this may be due to favorable socioeconomic characteristics, academic ability, and/or parents’ high expectations. Seeking to go beyond the role of family SES in the intergenerational reproduction of educational attainment and the role of middle-class cultural capital in reproducing advantage, Lee and Zhou (2014) introduce a model of cultural frames to explain how the children of immigrants in Los Angeles whose families 
11exhibit low SES and lack middle-class cultural capital attain exceptional educational outcomes. In fact, the low-SES origins of Vietnamese and Chinese parents do not hamper their children’s educational outcomes in the same way that it obstructs the educational outcomes of other racial/ethnic minority groups. Going beyond theories concerned with Asian culture such as parents’ rigid and authoritarian child-rearing practices (Chua, 2011) and values related to hard work (Brooks, 2012; Murray, 2012) for example, Lee and Zhou explain that these families share a frame for academic success that entails “getting straight A’s in high school, graduating from an elite university, and pursuing an advanced degree”. Academic achievement becomes the goal for Chinese and Vietnamese immigrant parents who perceive education as the only sure path to mobility—a perception that they have passed on to their children (Lee and Zhou 2013; Steinberg 1996; Sue and Okazaki 1990). Indeed, according to Lee and Zhou, given the potential discrimination that they fear their children may experience if they pursue cultural fields for example, Asian immigrant parents push their children into fields and professions in which they believe their children will experience the least possibility of bias and discrimination. Professions such as medicine, law, and engineering require advanced degrees and are understood as being more objective in their assessment of educational success which, immigrant parents believe, may shield their children from potential bias from employers, fellow employees, peers, customers, and clients. Research questions and Hypotheses This paper will investigate two particular research questions: How does a student’s family background, as measured both by socioeconomic status and ethnicity, influence their choice of 
12undergraduate field of study? And how does a student’s gender influence their choice of undergraduate field of study? In other words, are higher SES or lower SES students more likely to choose lucrative fields? Does family SES have similar or differential effects on men and women and on whites and other racial/ethnic groups? This study systematically examines these questions.   On a more theoretical level, does the long arm of family background reach across all levels of education (supporting EMI theory) or do students rather liberate themselves from its influence on their academic decisions once they reach university, as suggested by the “liberated” theory? An initial hypothesis in line with EMI theory would be that, due to the universalization of education which has become more easily accessible to the less socioeconomically advantaged students, the upper class seeks out whatever qualitative differences there are at that level and use their advantages to secure quantitatively similar but qualitatively better education (e.g. fields of study). In other words, supporting this hypothesis would be to show that upper class students are found in fields promising higher payoffs (prestige, income, etc), such as the hard sciences and professional programs (business, engineering, etc) rather than the soft sciences and cultural fields. On the other hand, a hypothesis in line with the “liberated” theory suggests a waning influence of  social  class  on educational  choices  and outcomes.  We would therefore  see  that choice of field of study is not significantly influenced by level of SES. As for gender, considering previous research, a relevant hypothesis would be that men are more likely to go into hard sciences and professional fields than women, despite the seemingly increasing “degendering” of fields of study. Women would be over represented in soft sciences 
13and cultural fields. Finally, as for ethnicity, in light of the above, a relevant hypothesis is that Asian student immigrants will be overrepresented in fields with higher payoffs such as the hard sciences and professional programs. In addition, Lyons et. al. (2010) found that visible minorities in Canada have higher salary expectations than domestic students. This could be a reflection of a culture of loyalty and obedience, especially among East Asians (Lee and Jablin, 1992), and also strong motivation for upward socioeconomic mobility, given that many visible minorities are first-generation immigrants (Somerville and Walsworth, 2009). Although current data do not allow for detailed information on student’s specific origins but rather assess if they are Canadian or landed immigrants, we can hypothesize that non-whites will choose fields with higher possible income and prestige as opposed to whites born in Canada. From here the thesis is divided into three separate sections.  First, the methods section presents an overview of the data (National Graduate Survey for both the 2005 and 2010 cohorts) and variables used in the analysis. The dependent variable (field of study) is explored through 4 different  codings:  1)  hard science/soft  science,  2)  cultural  fields/other,  3)  professional  fields/liberal arts and sciences, and 4) average income by field. The independent variables are measures of  socioeconomic  status  and  demographic  characteristics.  The  second  section  presents  the models for all 8 analyses (all four measures of the dependent variable for both cohorts) as well as the  results  detailing  the  salient  findings  and  their  implications  for  this  research.  Third,  the discussion concludes this paper by reiterating the objectives of this study and summarizing the relevant findings, as well as the implications for future research. 14METHODSData & Variables The data for this study are drawn from Statistics Canada’s National Graduate Survey (NGS) of 2007 (class of 2005) and 2013 (class of 2009-2010). Concerning variables of interest for this study, the questionnaire used for both years was the same. The NGS was conducted via computer-assisted telephone interviews, two or three years after respondents had graduated from a postsecondary institution. The NGS surveys provide extensive information on the educational experiences, program history, and early labor market experiences of recent graduates. The survey population is composed of all graduates of Canadian postsecondary educational institutions who had completed the requirements for degrees, diplomas, or certificates during the appropriate graduation year. The sampling frame used by Statistics Canada consisted of a listing of all the names and addresses of students who graduated from a specific institution in either of the survey years.  The institutions also provided the field of study, in which the student received a credential. The initial sample is of approximately 15,000 observations in each data set. The analyses for this paper will be restricted to only university graduates who completed their first bachelor’s degree (e.g., B.A., B.Sc., B.Ed., B.A.Sc., B.Eng.), allowing me to expand the existing body of literature on level to that of field. The final sample size is of 6,939 observations for the class of 2005, and 5,774 observations for the class of 2009-2010. The overall response rate for the survey of 2005 is 68% and 49.1% for the 2010 survey. In addition, in order for estimates produced from survey data to be representative of the target population, and not just of the sample itself, the NGS requests that users incorporate the survey weights into their calculations. 
15For the purpose of this study, probability weights have been included into the analyses in order to insure representativity. Dependent variable: “Field of Study”For  both  2005  and  2010  cohorts,  the  categorical  variable  “Aggregated  classification  of institutional  program (CIP) at  graduation” is  divided into the following 10 programs,  plus a residual  category:  1)  Education;  2)  Visual  and  performing  arts,  and  communications technologies;  3)  Humanities;  4)  Social  and  behavioural  sciences,  and  law;  5)  Business, management  and  public  administration;  6)  Physical  and  life  sciences,  and  technologies;  7) Mathematics,  computer  and  information  sciences;  8)  Architecture,  engineering  and  related technologies;  9)  Health,  parks,  recreation  and fitness;  10)  Agriculture,  natural  resources  and conservation; personal, protective and transportation services; and 11) other. Given that there are a variety of ways to understand and operationalize fields of study, this research examines four different coding methods in order to fully grasp the complexity of this variable and assess the most efficient way to measure its  variation.  Firstly,  the dependent variable is  recoded into a dichotomous measure, 1 for the academic fields encompassed in the “hard sciences” (programs 6 through 10 as listed above) and 0 for fields within the soft sciences (programs 1 through 5). This dichotomous coding implies that variation in family background and socio-economic status may influence a student’s choice of pursuing hard sciences, often regarded as more prestigious and rewarding, rather than other, possibly less financially gratifying disciplines. An initial hypothesis may be that students from higher SES backgrounds are more likely to pursue hard sciences, in keeping with EMI theory in that fortunate students seek to secure more advantageous fields or fields that have stronger prestige reputations among the public. As for gender and ethnicity, a relevant hypothesis in light of previous research would be that women are more likely to pursue 
16soft sciences and minority groups are more likely to study hard sciences given the latter’s greater prestige and promise for upward mobility. Secondly, the fields are recoded into a dichotomous variable opposing the cultural fields (coded  1)  and  all  other  fields  (coded  0).  In  this  case,  “Visual  and  performing  arts,  and communications technologies” is understood as the only cultural field, the 9 other programs are coded as 0 . This coding explores the possible impact of family background, socio-economic 3status,  gender  and  ethnicity  on  a  student’s  choice  to  go  into  the  cultural  fields.  A relevant hypothesis  consistent  with  the  EMI  perspective  would  be  that  students  from  a  lower  SES background are more likely to go into these fields,  lower class students thus pursuing fields promising less payoff and prestige,  effectively maintaining inequalities.  It  could also be that students from lower SES backgrounds, especially if they are the first in the family to attend college or university, don’t explicitly make a choice but end up taking courses in high school, and perhaps the early years of their post-secondary education, that by default lead them into certain fields of study. Here women are expected to be overrepresented in the cultural fields, and minority groups in the more economically and professionally oriented fields. Thirdly,  the  fields  are  recoded  into  a  dichotomous  variable  comparing  professional faculties  with  liberal  arts  and  sciences.  In  this  instance,  “Business,  management  and  public administration”,  “Architecture,  engineering  and  related  technologies”,  and  “Health,  parks, recreation and fitness” are included in the professional schools. On the other hand, “Education”, “Visual and performing arts, and communications technologies”, “Humanities”, “Physical and life sciences, and technologies”, “Mathematics, computer and information sciences”, and “Social and behavioural sciences, and law” are included in the liberal arts and sciences. Although law is 
 The humanities were also tested as part of the “cultural fields”, but did not present significantly different results 3than when only accounting for the “visual and performing arts”. 17understood as a professional school, the available data combines social and behavioural sciences with law into a single field and cannot be analyzed otherwise. Therefore, for the purpose of this particular coding, law is interpreted as a liberal arts and science field rather than a professional faculty. This coding of the dependent variable implies that SES factors, gender and ethnicity may influence whether a student chooses to pursue the professional fields, often seen as promising higher  social  and  financial  benefits.  EMI  theory  suggests  that  students  with  high  SES backgrounds would be over represented in these faculties, as would be men and certain minority groups such as Asian Americans, as suggested by existing literature.Lastly, the variable is recoded into ten average income amounts for each field, based on employee income in each field and related occupations, at the year of graduation. These results are based on Statistic Canada’s estimated gross annual earnings of 2009-2010 graduates with a bachelor degree working full-time in 2013, by selected fields of study (Stat. Can., 2013) (See Appendix I). The ten fields were coded with the following respective incomes: 1) $53,000; 2) $38,000; 3) $48,000; 4) $50,000; 5) $50,000; 6) $44,000; 7) $56,000; 8) $61,500; 9) $65,000; 10) $52,000. With the exception of the last field, all were represented in Statistic Canada’s graph and incomes were measured directly on the graph. The last field is coded with the average total income of all  fields.  Although the study from Statistics Canada is  representative of the total income for 2013, the same amounts are used in the following analysis for the 2007 graduating cohort. Income by field may have slightly varied from 2007 to 2013, but we assume that these variations are not large enough as to significantly influence results. This method of coding the dependent variable is explained by the fact that some fields are more predictable in terms of future jobs and earnings. Some programs, such as humanities and the social sciences, have a liberal arts orientation and students in these programs are expected to value learning for its own 
18sake and tend not to tie their study closely with materialistic concerns (Bonvillian and Murphy 1996; Brint 1998). Other programs, such as business and most engineering and health fields, orient  their  training  toward  cultivating  specific  skills  and  developing  marketable  credentials (Kerr 1991). In other words, college majors are associated with differential risks of job prospects and it is relevant to explore if students make rational choices in their decision-making process when choosing a field of study based on future salary prospects. Independent variables: Demographic and Socioeconomic factorsSESSocioeconomic status (SES) is measured by three variables: 1) whether or not students have received government  loans;  2)  whether  they received money from a  family-based registered education  savings  plan  (RESP) ;  and  3)  Parent’s  level  of  education.  Whereas  the  two  first 4variables  are  economic  measures  of  student’s  SES  background,  the  third  rather  measures  a student’s social and cultural capital. These three measures provide different indicators of SES , whether it be financial or cultural, and allow me to examine the full range of SES effects on choice of field of study. Looking first as government loans and RESP, both are dichotomous variables and allow for an estimation of student’s family SES. The logic behind this being that government student loans are means-tested and reserved for those students who have low family income and will therefore not be given to those from a higher SES background. Furthermore, a RESP is more accessible to those families with higher income who can afford to start a savings plan for their 
 RESP and government loans had initially been combined into a single categorical variable “SES”. Students who 4received government loans but had no RESP were coded “low SES” and those who did not receive government loans but did in fact have RESP were coded “high SES”. All others (those who did not have either RESP or loans and those who had both) were coded “med SES”. However, this combined variable proved non significant when included in the logistic regression.19children’s future. Lower SES families are not expected to have an RESP. As seen in Figures 1 and 2, students who receive a RESP are less likely to also have a loan, and students with a government issued loan are less likely to have a RESP. Figure 1: RESP and government student loans for the 2005 cohort Pearson chi2(1) =  16.9641   Pr = 0.000Figure 2: RESP and government student loans for the 2010 cohortPearson chi2(1) =  36.6510   Pr = 0.000RESPGovernment Loans        Yes            |            NoTotalYes 37144.6510.1546055.3513.296,284100.0088.32No 3,28352.2489.853,00147.7686.71831100.0011.68Total 3,46148.64100.003,65451.36100.007,115100.00100.00RESPGovernment Loans        Yes            |            NoTotalYes 48241.9916.7066658.0123.064,626100.0080.12No 2,40451.9783.302,22248.0376.941,148100.0019.88Total 2,88850.02100.002,88649.98100.005,774100.00100.0020The 2013 Survey of Approaches to Educational Planning (SAEP) states that almost 7 in 10  Canadian  children  (68%)  17  years  old  or  younger  had  savings  set  aside  for  their postsecondary education. Among children whose parents had a high school diploma or less, 52% had savings set  aside for  their  education.  This  proportion increased to  66% among children whose parents had a trade certificate or college diploma and 78% among children whose parents had a university degree. Among children whose parents hoped they would go into the trades or to college, 53% already had savings set aside at the time of the survey. This compares with 71% among children whose parents hoped they would attend university. Looking at Figure 3 which presents the proportions of postsecondary savings set aside for children by household income in 2013, RESP proves to be a relevant measure of family socioeconomic background. Figure 3: Postsecondary savings set aside for children 17 years old or younger,  by household income, 2013Source : Statistics Canada, Survey of Approaches to Educational Planning, 2013 Mother’s and father’s level of education are also included as measures of SES as a series of  dummy variables  (1=yes;  0=no)  for  each level  of  schooling.  The coding differs  for  both cohorts. For the year 2005 the categories are “no post-secondary”, “college”, and “university”. The referent for this year being “no post-secondary”. For 2010, the categories encompass a larger range of options: “less than high school”; “high school diploma or equivalent”; “trade certificate 
21or  diploma”;  “college  or  other  non-university  certificate  or  diploma”;  “university  up  to  and including Bachelors”; “university above the Bachelors” . The referent for this model is “less than 5high school”. As early as 1967, Blau and Duncan noted that the strongest predictor of a son’s occupation is his father’s occupation, proving that advantages are transmitted intergenerationally. Including parents level of education into the current study allows to assess the accuracy of the status attainment model in the study of intergenerational mobility, which emphasizes the role of family socioeconomic status in determining children’s educational and occupational outcomes. As seen in Tables 11 through 14 (see Appendix II), in the 2005 cohort for instance, for students who don’t receive a RESP, nearly 40% of them have parents whose level of education is less than a postsecondary degree. Reversely, for students who do receive a RESP, nearly 50% of them have parents who have a university degree. As for government loans, for students who do not receive this financial aid, 45% of their fathers have a university degree, and among those who do receive loans, 45% of their parents have no postsecondary education. Demographics, Language and ImmigrationThe data from the class of 2005 and that of 2010 vary. In fact, the 2013 survey for the 2010 cohort  offers a more elaborate dataset  with a considerable amount of additional independent variables. Therefore, models for both cohorts will differ, the goal of this research is to understand each year individually rather than in comparison to one another. In addition to the SES variables listed above, both surveys include the demographic characteristic of the respondents sex (coded 0=female; 1=male). 
 The coding of parent’s level of education for the 2010 cohort could have been combined as to resemble that of the 52005 cohort, but this would have required the lumping of dissimilar categories such as inserting the “high school degree” from the 2010 cohort into either the “less than high school” category or the “college” category present in the 2005 cohort. As for the “trade certificat” specific to the 2010 cohort, it cannot be included in any of the 2005 categories, doing so would have taken away from the analysis of Table 7 for instance. 22The 2013 survey includes all  variables listed above (resp,  government loans,  parent’s level of education and respondent’s sex) as well as five additional demographic characteristics: First, the age of the respondent at the time of graduation is coded as a dichotomous variable (1=less  than  25;  0=25  and  older).  The  coding  for  this  variable  is  such  since  60%  of  the respondents belong to the first group under 25 years of age. A categorical variable would have complicated the results and the sample spreads too thinly to later categories of 30 to 39 and 40 or more. Other  independent  variables  include  whether  or  not  the  respondent  identifies  him or herself  as  a  member  of  a  visible  (ethnic  or  racial)  minority  (1=does  identify;  0=does  not identify);  the  respondent’s  status  in  Canada  at  the  time  of  the  interview recoded  into  three dummy variables  answering to  whether  or  not  the  student  is  a  Canadian citizen by birth,  a Canadian citizen by naturalization or a landed immigrant (the referent here is Canadian by birth); the  language  the  respondent  first  learned  at  home (1=English  and/or  French;  0=Other);  and finally, the region of the institution from which the student graduated is recoded into four dummy variables,  one  for  each  region:  Atlantic  provinces;  Quebec;  Ontario;  Western  provinces  and territories (the referent being the Atlantic provinces).  Region of institution is included in the analysis because fields of study on offer at different universities, and across different regions, vary. Effectively this means that students are more constrained in some regions, or conversely have more opportunities in other regions, to choose particular fields of study.   Because of this variation, and the possibility that regional choices will be impacted by SES, gender, and race/ethnicity, it seems important to control for this variable.23MODELS & RESULTSThe following section will discuss eight models of analysis addressing each of the four measures of the dependent variable (cultural fields, professional fields, hard sciences and average income) for both 2005 and 2010. Class of 2005 (Survey of 2007)For  the  following  four  tables  concerning  the  2005  cohort,  Model  1  includes  RESP  and government student loans. Model 2 includes the respondent’s sex and Model 3, parent’s level of education .6Hard Sciences vs Soft SciencesFirstly, using the dichotomous measure of hard sciences as the dependent variable, I used logistic  regression  to  estimate  the  effect  of  socioeconomic  factors  and  demographic characteristics  on  the  choice  of  program  type.  Table  1  presents  logistic  regression  models predicting choice of field within the hard sciences including “Physical and life sciences, and technologies”; “Mathematics,  computer and information sciences”; “Architecture, engineering and  related  technologies”;  “Health,  parks,  recreation  and  fitness”;  and  “Agriculture,  natural resources and conservation”. To ensure reliable model estimates, I conducted robustness checks to be certain that results are not influenced by a 6small number of observation(s) or associations in the data. OLS assumptions are not violated in the analyses for Table 4 and collinearity was not found within the logistic regression in Tables 1, 2 and 3 (VIFs < 2; Condition   Index < 5.5)24Table 1: Logistic regression models for hard sciencesLooking first  at  models  within  the  hard sciences  from the  cohort  of  2005 (Table  1), whether or not students receive a Registered Education Savings Plan (RESP) is significant in all three models. In Model 1, for or those who receive RESP, there is a .240 increase in the logit of going into hard sciences as opposed to soft sciences. In model 2, there is a .238 increase in the logit  of  going  into  hard  sciences  and  a  .189  increase  in  Model  3.  Therefore,  even  when controlling for gender in Model 2, the effect of RESP on the type of field pursued remains.When assessing the percentage change in odds, for students who do receive payouts from a RESP, the odds of going into hard sciences are 21% greater as opposed to those who do not receive any RESP funding, suggesting that students from a higher SES background are more likely to choose a field within the hard sciences. These findings support EMI theory, suggesting 
25that students from higher SES backgrounds are more likely to secure qualitatively better “types” of education, thus maintaining social inequalities within the education system. Whether  or  not  a  student  receives  government  student  loans  is  also  consistently significant  throughout  all  three  models.  When  assessing  the  percentage  change  in  odds,  for students who do receive these loans, the odds of going into hard sciences are 15.5% larger than those who do not receive such funds, controlling for parental level of education. These findings are contradictory to results concerning RESP assistance, suggesting that students from lower SES backgrounds (thus receiving government loans) are more likely to go into hard sciences. This suggests  a  more  complicated  relationship  between  SES  and  field  than  argued  in  previous research. Contrary to what EMI may suggest, lower SES students are not underrepresented in these fields and upper class students may not be the only ones securing the more advantageous fields. Students with lower SES may use their government loans to secure higher paying, more prestigious jobs as a means to insure upward mobility or to repay their student debt. In other words, if lower SES students have to carry higher debt loads to attend university, it might be that they are much more likely to enter fields that they perceive will have higher economic rewards. Looking at the following models for cultural fields and professional fields of study will allow further assessment of the accuracy of the EMI theory.Gender  is  also  highly  significant  in  Table  1,  as  was  predicted.  When  measuring  the percentage change in odds, for men the odds of going into hard sciences are 77% greater as opposed to women. This finding suggests that gendered patterns in choice of field of study still prevail, supporting findings by Goyette and Mullen (2006) stating that men have traditionally chosen fields such as business, engineering, and hard sciences while women have rather studied 
26soft sciences. Results further support Charles and Bradley (2009) who find that sex segregation by field of study has not waned as would be expected, but rather that enduring cultural forces of gender-essentialist  ideology  are  still  strongly  at  work.  Looking  at  the  following  models  for cultural  fields  and  professional  fields  of  study  will  allow  to  further  assess  the  potential “degendering” of curricular fields in Canadian undergraduate programs and institutions. Cultural fields vs OtherTable 2 presents  logistic  regression models  predicting choice of  program within the cultural fields included in the “Visual and performing arts, and communications technologies”. Table 2 : Logistic regression models for cultural fields 
27Looking  at  choice  of  program  within  the  cultural  fields,  SES  factors  of  RESP, government loans and father’s level of education, are not significant in this model. This suggests that cultural fields of study are not pursued by any particular type of student and the choice of these  fields  does  not  depend  on  one’s  socioeconomic  status  or  gender.  For  students  whose mothers have a college degree, the odds of going into cultural fields are 38% greater than those whose mothers have no post-secondary education, and 39% larger for those whose mothers have a university degree. This may be that, as shown in Van de Werfhorst’s (2001) study of the impact of the family background on choices of fields of study, children of the cultural elite are likely to choose a cultural field of study. Although available data for the current research does not state parent’s field of study, it may be that students whose mothers have a college or university degree in a cultural field are more likely to pursue such a field themselves. EMI theory does not apply here since students from a lower SES are not more likely to pursue the cultural fields as this theory would suggest. The economic elite seeking to secure more advantageous fields of study, they would most likely be underrepresented in fields such as the arts and humanities. Therefore, in  accordance with the “liberated” theory, Table 2 suggests a weak influence of social class on educational choices and outcomes, as it shows that choice of field of study within the cultural fields is not significantly influenced by level of SES.As for gender, although it is well documented that the distributions of men and women within the higher education systems are extremely uneven (see Bradley 2000; Jacobs 1996; Kelly and Slaughter 1991), Table 2 suggests a progressive elimination of gender differentiation among higher education. Indeed, as seen with previous literature, many U.S and Canadian studies point to persistently high levels of segregation in universities with women strongly underrepresented in science, engineering, and technical programs even in countries with high overall female 
28enrollment rates (Jacobs 2003; Xie and Shauman 2003; England et al. 2007). And despite recent comparative analyses suggesting that some forms of segregation are more pronounced in the most socially or culturally modern societies (Charles and Bradley 2002, 2006; Van Langen and Dekkers 2005), Table 2 hints to an erosion of gender differentiation with respect to cultural fields of study. The potential “degendering” of fields of study will be further explored in the following tables. Professional fields vs Liberal Arts and SciencesTable 3 presents logistic regression models predicting choice of program within the professional fields as opposed to the liberal arts and sciences. This includes fields related to business, health and engineering.Table 3: Logistic regression models for professional fields   29Looking now at choice of field within the professional schools including business, health and engineering, Table 3 shows that whether or not students receive financing via a RESP or government loans is not significant in determining their choice of field within the professional schools rather than the liberal arts. Results are not consistent with EMI theory, suggesting that students of higher SES are not more likely to secure advantageous fields but rather they may be liberated from their family background. Therefore, socioeconomic advantages do not seem to work to effectively and continuously secure for the children profitable locations of their own. As seen with Tables 1 and 2, Table 3 suggests the accuracy of the “liberated” theory which stipulates a waning influence of social class on educational choices and outcomes, as it shows that choice of field of study is not significantly influenced by level of SES. In other words, contrary to what argues EMI theory, students from higher SES backgrounds may not be the only ones seeking to enter fields that they perceive will have higher economic and social rewards. It may be that all students, regardless of their family background, seek to secure such fields. Lower SES students might also be more likely to enter fields with higher payoffs as a way to settle the greater debt loads they accumulated to attend university.Gender is highly significant and is the only variable influencing students choice of field within the professional schools as opposed to the liberal arts and sciences. The odds of going into professional fields are 44% larger for men than for women. As seen with the hard sciences in Table  1,  these  findings  support  the  continuing  “gendering”  of  curricular  fields  in  Canadian undergraduate  programs  and  institutions.  These  findings  concur  with  those  by  Goyette  and Mullen (2006) which state that men have traditionally chosen professional fields such as business and engineering while women have rather studied soft sciences. 
30Average income by fieldTable 4 presents the OLS regression models of the average income per field based on Statistic Canada’s  estimated  gross  annual  earnings  of  2009-2010  graduates  with  a  bachelor  degree working full-time in 2013, by selected fields of study.Looking  at  choice  of  field  based  on  the  average  income  for  each  field  (Table  4), government loans are significant throughout all models. In fact, looking at Model 3, students with government loans are in fields that earn on average 417.720 more than those who do not receive  loans.  Conflicting  with  what  EMI  theory  may  suggest,  these  findings  indicate  that students from lower SES backgrounds (thus receiving loans) earn more than those from higher SES backgrounds who do not receive loans. There are many ways to interpret this, but it may be that students who take risks (i.e., financial loans) pursue fields promising higher earnings in order Table 4: OLS regression models for average income per field of study31to pay their debt or as a way to ensure upward mobility. As found by Yingyi (2009) concerning the influence of SES on choice of college major, this may indicate that once in college, lower SES children and their  families  rationally  utilize  the opportunity  of  college major  choice to enlarge the economic returns from their college education. It may also be that students do not pay as much attention to the potential earnings of a field as they do prestige or reputation when deciding on a program. In addition, it is possible that the coding of average incomes is too coarse to pick up the variations that might in fact influence student choices. For example, law and other social  sciences  are  combined  in  these  models  although  the  pay  differences  between  the occupations to which these degrees lead are likely to be large. Findings support the “liberated” theory in that social class has a weak influence on student’s educational choices and outcomes at the undergraduate level. In this sense, students from lower SES can and do secure fields with high financial rewards.  Clark and Lipset (1991) argued that the modern family determines less the  education  and  jobs  of  individual  family  members  than  before.  Increased  wealth  and government  support  programs  have  expanded  choices  to  individuals,  and  cumulatively transferred more functions than ever away from the family. In fact, social mobility studies show decreasing  effects  of  parents’ education  and  income  in  explaining  childrens’ occupational success.  In  other  words,  according to  Clark  and Lipset,  social  mobility  is  now less  family-determined and more ability and education-determined. As  for  gender,  men  are  in  fields  that  earn  on  average  713.213  more  than  women, supporting  theories  of  a  continuing  “gendering”  of  fields  and,  furthermore,  reproducing  the inequalities in wages among genders. Interestingly, these findings support Davies and Guppy (1997) concerning college enrolment in that (1)males are much more likely to enter fields of study with higher economic returns than are females;(2)socioeconomic factors do not affect 
32chances of  entry into lucrative fields net  of  other  background factors.  Gender differences in major  choice  are  extremely  complex,  and  no  simple  explanation  can  be  provided  for  them. However, one possible interpretation is that women and men typically attach different values to the after-college opportunities associated with each area of study. Recognizing potential family responsibilities, women may prefer more flexible fields in which skills are unlikely to become obsolete (Turner and Bowen, 1999). Considering all four methods for coding fields of study for the class of 2005, it is clear that the effect of gender is robust, with the exception of cultural fields of study.  Whether the dependent variable be coded as hard sciences, professional fields or as income, gender has a statistically  significant  effect  in  the  final  models,  net  of  other  variables.  This  is  clear  and compelling evidence of the continued gender stratification that exists within Canadian higher education.  The effects for SES are mixed and less evident. Effectively there are three separate measures of SES used as independent variables: possession of a RESP, receipt of a government loan, and parental education. Possession of a RESP is significant only when considering fields as hard sciences versus soft sciences (Table 1). This suggests that students from higher SES are more likely to go into fields with higher economic payoffs. However, the receipt of government loans is also significant in Table 1, indicating that students from lower SES also choose to pursue hard sciences, suggesting that the relationship between a students financial status and their choice of field is more complex than what EMI would suggest. The receipt of a government loan is statistically significant in the final models of hard sciences versus soft sciences (Tables 1) and for the coding of the dependent variable by income for each field (Table 4), net of other factors. This suggests that students from less advantaged backgrounds are more likely to pursue hard sciences and fields which command higher income.  These findings question the accuracy of the 
33EMI theory. There are many ways to interpret these results, among others, it may be that students with lower SES use their government loans to secure higher paying, more prestigious jobs as a means to insure upward mobility or to repay their student debt. It could also be that students from lower SES backgrounds are less risk averse – they may consciously select into fields of study that have clear occupational trajectories with solid economic payoffs.Clearly, in comparison to gender, the effects of social class on student choices of field of study are more complex and thus more difficult  to interpret.  The effect of SES variables on choice of field is not constant throughout all tables for the 2005 cohort. This first set of results hints  towards  a  more  “liberated”  education  for  students  from different  family  backgrounds, suggesting higher social mobility than considered by the EMI theory. The evidence seen so far also implies that relative to the effects that gender has on field of study choices, SES is not as significant a factor.Class of 2010 (Survey of 2013)The following tables concern the cohort of 2010. Model 1 includes SES factors of RESP and government student loans. Model 2 includes mother’s and father’s level of education. Model 3 concerns demographic characteristics of gender, age and whether or not students self-identify as  a  member  of  a  visible  minority.  Model  4  includes  Canadian  Status  (the  referent  being Canadian by birth) and mother tongue as English and/or French relative to another language. Finally, Model 5 adds the Canadian region of the institution in which the student graduated, the referent being the Atlantic provinces 
7 To ensure reliable model estimates, I also conducted robustness checks for the following analyses to be certain that 7results are not influenced by a small number of observation(s) or associations in the data. OLS assumptions are not violated in the analyses of Table 9 and collinearity was not found within the logistic regression in Tables 5, 7 and 8 (highest VIF =4, for mothers with a bachelor’s degree; Condition Index < 18). 34Hard Sciences vs Soft SciencesTable 5 presents logistic regression models predicting choice of field within the hard sciences.Table 5: Logistic regression model for hard sciences35Looking at the logistic regression models for the 2010 cohort concerning the choice of field within hard sciences, RESP is not significant, suggesting that students from a higher SES are not necessarily more likely to go into hard sciences. Government loans, on the other hand, are significant in Models 1 to 3 as well as Model 5 when including parent’s level of education, the respondent’s sex and the region of the institution. In fact, looking at Model 5, for those who receive financial aid from the government, the odds of going into hard sciences are 19% greater as opposed to those who do not receive such loans. Government loans do not remain significant in Model 4 when including factors of Canadian citizenship and language (although the latter variables are not significant either). It may be that landed immigrants and naturalized Canadians are more likely to pursue hard sciences, regardless of financial assistance. As argued by Xie and Goyette  (2003),  ethnicity  does  in  fact  influence  choice  of  a  more  profitable  field  of  study. Specifically among Asian-Americans, many of whom are recent immigrants, and may choose fields with higher possible earnings as a way to ensure upward mobility.These  findings  suggest  that  students  from lower  SES  backgrounds  (thus  eligible  for government loans) are more likely to pursue the hard sciences, further questioning the accuracy of the EMI theory in light of the current data. As also shown in Table 1 for the 2005 cohort, lower SES students are not less likely to pursue these fields and upper class students may not be the only ones securing the more advantageous types of education. Students with lower SES may use their government loans to secure higher paying, more prestigious jobs as a means to insure upward mobility or to repay their student debt. In other words, if lower SES students have to carry higher debt loads to attend university, it might be that they are much more likely to enter fields that they perceive will have higher economic rewards. 
36Gender is again significant as men are more likely to go into hard sciences than women, which concurs with findings from the cohort of 2005 as well as previous literature on the current existence of sex segregation by field in industrialized countries. In fact, for men, the odds of going into hard sciences are 40% larger than for women. As for age at the time of graduation, students under the age of 25 are more likely to graduate from the hard sciences than those 25 and above (Table 5). Table 6 presents the distribution of fields of study by the age of respondents at the time of graduation in 2010. Whereas 44% of students less than 25 years of age graduate from the hard sciences, 36.5% of those older than 25 graduate from similar fields. Finally, looking at Model 5 of Table 5, with regards to the province of Quebec, the region of the institution proves to be an important factor in explaining choice of field of study. In fact, the negative coefficient for the province of Quebec shows that these students are less likely to go into hard sciences than those in the Atlantic provinces.In sum, the logistic regression models are in keeping with the initial research question and reveal that SES factors and gender do in fact influence choice of undergraduate field of study within the hard sciences. The following models will further explore this relationship for various codings of the dependent variable as well as the influence of ethnicity on field of study within the 2010 cohort. 37Table 6: Distribution of field of study by age of respondent for the 2010 cohortField of Study Less than 25 25 or more Total1) Education 28040.178.0141759.8318.30697100.0012.072) Visual and performing arts, communications & technologies 27670.057.9011829.955.18394100.006.823) Humanities 42963.9312.2724236.0710.62671100.0011.624) Social and behavioral sciences, and law 54162.8315.4832037.1714.04861100.0014.915) Business, management and public administration43855.6512.5334944.3515.31787100.0013.636) Physical and life sciences and technologies 51182.9514.6210517.054.61616100.0010.677) Mathematics, computer and information sciences 15556.784.4311843.225.18273100.004.738) Architecture, engineering and related technologies36166.9810.3317833.027.81539100.009.339) Health, parks, recreation and fitness 34951.639.9932748.3714.35676100.0011.7110) Agriculture, natural resources and conservation; transportation services; other15559.624.4310540.384.61260100.004.50Total 3,49560.53100.002,27939.47100.005,774100.00100.0038Cultural fields vs OtherTable 7 presents  logistic  regression models  predicting choice of  program within the cultural fields.Table 7 : Logistic regression models for cultural fields
39Looking at student’s choice of program within the cultural fields (Table 7), measures of SES are not significant with the exception of father’s education at the level of a trade degree, a bachelors degree and a university graduate degree. In fact for students whose fathers have a trade degree, the odds of going into a cultural field are 95% greater than those whose fathers have less than a high school degree. As for students whose fathers have a university graduate degree, the odds of pursuing a cultural field are 88% greater than those whose fathers have less than high school degree. In considering that a higher level of education on the father’s side is an important indicator of higher income and therefore higher SES, results suggest that students from a higher social class are more likely to go into cultural fields. It may be that the financial security felt by students with highly educated fathers leads children to pursue fields based on interest (cultural, social, etc) rather than income or prestige. It may also be possible that for families with higher SES, the cultural fields present elements of prestige, social status or cultural capital greater than that of the hard sciences, for example. Similarly to results in Table 2 for the 2005 cohort, it may be that, as shown in Van de Werfhorst’s (2001) study of the impact of the family background on choices of fields of study, children of the cultural elite are more likely to choose a cultural field of study.Although available data for the current research does not state parent’s field of study, it may be that students whose fathers have a university degree in a cultural field, for instance, are more likely to pursue such a field themselves. EMI theory does not apply here since students from a lower SES are not more likely to go into the cultural fields as this theory would suggest. The economic elite seeking to secure more advantageous fields of study, they would most likely be fewer in fields such as the arts and humanities. Therefore, in  accordance with the “liberated” 
40theory, Table 7 suggests a waning influence of social class on educational choices and outcomes, as it shows that choice of field of study is not significantly influenced by level of SES.The negative coefficient for males suggests that women are more likely to go into cultural fields.  In  fact,  for  females,  the  odds  of  going  into  “Visual  and  performing  arts,  and communications technologies” are 44% greater as opposed to men. As stated above, this supports existing theories of gendered fields of study, suggesting that sex segregation by field of study has not waned as would be expected and that while men are more likely to go into hard sciences, women enrol in soft sciences and cultural fields. In addition, the negative coefficient for landed immigrants shows that this group is less likely to go into the cultural fields than are Canadians by birth. As seen in previous literature, this may be due to the fact that academic achievement is a primordial goal for Asian immigrant parents, for instance, who perceive education as the only sure path to mobility—a perception that they have passed on to their children (Lee and Zhou 2013; Steinberg 1996; Sue and Okazaki 1990). Indeed, Lee and Zhou (2014) explain that Asian immigrant parents push their children into fields and professions in which they believe their children will experience the least possibility of bias and discrimination. This may explain why landed immigrant students are less likely to pursue cultural fields which are seen as allowing for less social mobility and prestige.  It may also be that for immigrant children who are not as deeply immersed in the values, religion, and deep cultural knowledge of a European tradition, the risks are higher in cultural fields. The deep knowledge of a specific cultural tradition is certainly necessary to do well in some areas of the creative and performing arts. Finally, younger students are more likely to graduate from a program within the cultural fields (the positive coefficient in Table 7 relates to students who are under 25 years old, this 
41category being coded as 1). Indeed, as seen in Table 6, 70% of students who graduate from such fields are under the age of 25 at the time of graduation. This may be that older students enrol in fields  with  higher  payoffs  and  long  term  security,  advantages  that  the  liberal  arts  do  not necessarily promise.  It  may also be that  students  who enrol  in other types of  fields such as professional schools or hard sciences take longer to graduate. Courses may be more laborious, for instance. Completion may also be delayed in these types of programs as tuition in usually higher than in cultural fields, this may require students to take on a part-time employment in addition to  expected course  work,  lengthening the  completion of  their  degree  over  a  longer period  of  time.  Another influential factor in determining whether or not students go into the cultural fields is the region of institution. In fact, for those in the West, the odds of going into cultural fields are 125% greater than those in the Atlantic provinces, and in Quebec the odds are 76% greater of going into cultural fields.  Professional Fields vs Liberal Arts and SciencesTable 8 presents logistic regression models predicting choice of program within the professional fields as opposed to the liberal arts and sciences.Looking at students choice of professional fields as opposed to liberal arts and science, RESP is significant in Model 1 but does not remain so when adding other SES and demographic variables in subsequent models. Government loans do not prove to be significant either when examining factors influencing student’s choice to pursue professional fields. The findings do not support  EMI theory  which  states  that  for  levels  of  education  that  are  universal,  such  as  an undergraduate degree which is increasingly accessible, competition will occur around the type of 
42education attained. As for the 2005 cohort, the 2013 results rather suggest that student’s SES background does not influence choice of field within the professional schools. Table 8: Logistic regression models for professional fields  
43As for gender,  results are consistent with findings from the 2005 cohort.  The odds of going into professional fields are 40% greater for men than for women, supporting  the continuing “gendering” of curricular fields in Canadian undergraduate programs and institutions. These findings concur with those by Goyette  and Mullen (2006) which state  that  men have traditionally  chosen professional  fields  such as  business  and engineering while  women have rather studied soft sciences. In addition, the variable assessing whether or not the respondent is a landed immigrant is significant. Indeed for this group, the odds of going into the professional fields are 96.5% larger than for Canadians by birth. Canadians by naturalization are also more likely to pursue professional fields (Model 5). As argued by Xie and Goyette (2003), ethnicity does  in  fact  influence choice  of  a  more  profitable  field  of  study.  Specifically  among Asian-Americans, many of whom are recent immigrants, and may choose fields with higher possible earnings as a way to ensure upward mobility. Age is significant in Models 3 and 4, but does not remain  so  when  including  the  region  of  the  institution  in  Model  5.  In  fact,  the  negative coefficient on age suggests that students above the age of 25 are more likely to graduate from professional fields as opposed to those under 25. As seen in Table 6, whereas 33% of those under 25 graduate from these fields, 37.5% of those over 25 graduate from these same programs. As for the region of the institution, for students in Quebec, the odds of going into fields such as business and engineering are 45% larger as opposed to the Atlantic provinces and on the other hand, students in Ontario are less likely to go into these fields as opposed to the Atlantic provinces. 44Average income by field Table 9 presents the OLS regression models of the average income per field based on Statistic Canada’s  estimated  gross  annual  earnings  of  2009-2010  graduates  with  a  bachelor  degree working full-time in 2013, by selected fields of study.Table 9: OLS regression models of average income per field of study 
45 Looking at the OLS regression models of average income per field of study (Table 9), RESP, government loans and parent’s level of education are not significant in explaining choice of field based on income. Although it may seem counter-intuitive in light of previous research, student’s SES background does not seem to influence choice of field based on average income. This may be due to the fact that students do not pay as much attention to the potential earnings of a field as they do prestige or reputation when deciding on a program. It may also be that students pursue fields based on interest (cultural, social, etc)  rather than income. Interestingly, contrary to the 2005 cohort (Table 4), in the case of Table 9 there are no significant gender differences in choice of field based on income. These findings do not concur with theories of a continuing “gendering” of fields in which inequalities in wages among genders are said to be reproduced. This does not mean that the gender wage gap has disappeared but rather that men and women both tend to choose fields with higher possible income. It may be that the question of what a student should study and what sort of return on investment it can generate is becoming an an increasingly important question, as families struggle to pay for school and graduates shoulder growing debt. Age is highly significant as students over 25 are more likely to graduate from high paying fields. The negative coefficients in Models 3 through 5 show that students who graduate under the age of 25 earn on average 1,375.591 less than those who graduate being over 25 years old. As seen above, this may be due to the fact that students above the age of 25 are more likely to graduate from professional fields as opposed to those under 25 and that younger students are more likely to graduate from the cultural fields. Students in the province of Quebec are less likely to go into high paying fields. Indeed, they choose fields which earn on average 654.686 less than those in the Atlantic provinces. 
46Considering all four methods for coding fields of study for the class of 2010, it is clear that the effect of gender is robust (as it also is for the 2005 cohort), with the exception of fields by average income. Whether the dependent variable be coded as hard sciences, cultural fields or professional fields, gender has a statistically significant effect in the final models, net of other variables. Whether it be an overrepresentation of women within the cultural fields or that of men in  the  hard  sciences  and  professional  fields,  this  is  clear  and  compelling  evidence  of  the continued gender stratification that exists within Canadian higher education.  The effects for SES are  mixed  and  less  evident.  Effectively  there  are  three  separate  measures  of  SES  used  as independent  variables:  possession  of  a  RESP,  receipt  of  a  government  loan,  and  parental education.  Possession of  a  RESP is  nonsignificant  throughout  all  models.  This  suggests  that students  from higher  SES (who have the possibility  of  receiving a  RESP) might  not  secure advantageous fields as suggested by EMI theory. The receipt of government loans is significant in  Table  5,  indicating  that  students  from lower  SES choose  to  pursue  hard  sciences,  again suggesting that the relationship between a students financial status and their choice of field is more complexe than what EMI would suggest. These findings question the accuracy of the EMI theory. There are many ways to interpret these results, among others, it may be that students with lower SES use their government loans to secure higher paying, more prestigious jobs as a means to insure upward mobility or to repay their student debt. As for ethnicity, the choice to pursue cultural fields (opted for by born Canadians) as well as professional fields (chosen by landed immigrants and naturalized Canadians) is influenced by the students ethnic background. Hard sciences and fields based on income, however,  are not effected by this demographic characteristic. Recent immigrants may choose fields with higher possible earnings as a way to ensure upward mobility as well as a way to experience the least 
47possibility of bias and discrimination from employers, fellow employees, peers, customers, and clients.Clearly, in comparison to gender and ethnicity, the effects of socioeconomic status on student choices of field of study are more complex and thus more difficult to interpret.  This second set of results concurs with the first from the 2005 cohort, suggesting a more “liberated” education for students from different family backgrounds, presenting higher instances of social mobility than considered by the EMI theory. 48Conclusion This paper set out to answer particular research questions: How does a student’s family background, as measured both by socioeconomic status and ethnicity, influence their choice of undergraduate field of study? And how does a student’s gender influence their choice of undergraduate field of study?   EMI theory suggests that, due to the universalization of education which has become easily accessible to the less socioeconomically advantaged students, the upper class seeks out whatever qualitative differences there are at that level and use their advantages to secure quantitatively similar but qualitatively better education (e.g. fields of study). Indeed, fields of study differ not just in substantive domain areas but also in labor market rewards. Assuming that lucrative majors are advantageous, if families secure for their children “some degree of advantages wherever advantages are commonly possible” (Lucas 2001:1651), it is plausible that college students from higher SES families would choose more lucrative fields than students from modest family origins. However, when using RESP, government loans and parent’s level of education as indicators of a student’s familial SES, results suggest that the relationship between SES and field may not be as simple as suggested by the EMI theory. In fact, the 2005 and 2010 cohorts do not consistently support EMI theory, no matter the coding method for the dependent variable. In the 2005 cohort, students from all SES backgrounds choose hard sciences (Table 1), cultural fields (Table 2) and professional fields (Table 3). And most surprising is that students from lower SES choose fields with a higher potential income (Table 4). As for the 2010 cohort, students from low SES choose to pursue a field within the hard sciences (Table 5), and students 
49from all SES backgrounds go into cultural fields (Table 7) and professional fields (Table 8). Interestingly, student’s SES background in no way influences the choice of a more lucrative field (Table 9).   Although results are consistent with the “liberated” theory in that social class has a weak influence on educational choices and outcomes, predicting the absence of a relationship between choice of field of study and level of SES, there are many possible explanations to these findings. Among others, it may be that lucrative fields do not appeal to every student. Some students and their families may emphasize other rewards beside the financial in their post-secondary education. Studies on adolescent work values show that children from higher social class backgrounds place more emphasis on influence and intrinsic work rewards than extrinsic rewards, when compared to their lower class peers (Kohn and Schooler 1983; Mortimer, Lorence, and Kumka 1986), an outlook that is related to their parents’ positions in the occupational structure (Johnson 2002). These different work values influence what students look for in their education. Relative Risk Aversion (RRA) theory (Breen and Goldthorpe 1997; Goldthorpe 1996) also provides an interesting explanation for such findings. RRA argues that students from various family backgrounds may have differential ability in affording the risks of uncertain job prospects resulting from their choice of field of study. Parents seek to ensure their children acquire a class position at least as advantageous as that from which they started, avoiding the risk of downward mobility. Indeed, the decision of what field to choose entails relative risks associated with job opportunities and financial rewards. Assuming students and their families are rational actors, they would consider the differential risks of fields of study in their decision-making process. Furthermore, assuming that lower SES students would have to 
50avoid risks more than higher SES students, it is likely that lower SES students are prone to choosing lucrative fields, such as computer science and engineering. Goldthrope (1996) applies rational action theory to the explanation of persisting class differentials in educational attainment. He suggests that education may be viewed differently by high and low SES families: Whereas high SES families treat education as a consumption good, low SES families see education as an investment. This may explain differences in choice of field, low SES students pursuing more “secure” fields. In this way, field of study, instead of maintaining the intergenerational transmission of status, as EMI predicts, may well weaken the disadvantageous effect of family background, thereby providing a means for upward social mobility for lower SES students.As for gender,  results suggest that,  to a certain extent,  sex segregation among higher education is still in play today. For both cohorts, men are more likely to go into hard sciences and professional fields than are women, despite the seemingly increasing “degendering” of fields of study and greater female enrolment and graduation. These findings concur with those by Goyette and Mullen (2006) which state that men have traditionally chosen professional fields such as business  and  engineering  while  women  have  rather  studied  soft  sciences.  Recent  research suggests that differences in values might lead women to favour college majors that prepare them to help others,  such as  the  social  sciences  and education,  and lead men to  gravitate  toward lucrative college majors, such as business and those in the technical fields (Marini et. al. 1996). Yingyi  (2009)  offers  the  explanation  that  family  SES  may  have  an  asymmetrical  effect  on college major choice for men and women. Men from a high SES family may still be expected to choose a lucrative career, whereas women from comparable backgrounds may not. On the other 
51hand, men as well as women may be equally motivated to study more lucrative majors as the route to achieve social mobility, for themselves as well as for their families, if they are from lower SES families. Therefore, the role of lucrative college major choice in potentially uplifting students’ and their families’ SES would outweigh the traditional gender role socialization that contributes to the divergent career paths toward which men and women are oriented. Tables 15 through 22 (see Appendix III) look at associations between gender and SES for both cohorts of the NGS data. Findings concur with Yingyi’s previous explanation, as students (men and women) from lower SES families (thus not receiving a RESP and more likely to be granted a student loan) are found to enrol into hard sciences, professional fields and fields with high income, more so than those students who are of a higher SES. Differences in gender, however, do not stand out when considering SES in choice of field. Interestingly, whereas the 2005 cohort presents gender differences  in  choice  of  field  based  on  income  (men  choosing  fields  with  higher  potential payoffs), the 2010 cohort does not present similar results. These tables examine associations in the data and do not show major differences in gendered effects, however, these interactions could be further explored through regression analysis. Davies and Guppy  (2014) provide a plausible explanation among others, arguing that occupations  traditionally  chosen  by  women  such  as  teaching,  nursing  and  social  work,  are professionalizing and increasingly demand university degrees as a minimal requirement. These changes  may  have  shifted  women’s  assessments  of  the  cost  and  benefits  of  seeking  higher education. Since labor market success is more contingent on educational credentials for women than for men (the latter are overrepresented in blue-collar sectors), woman may have higher risks associated with not seeking higher education and therefore higher paying fields. 52Finally,  a student’s ethnic background does in fact  influence choice of field of study. Whereas, landed immigrants are less represented in the cultural fields, they are more likely to choose professional fields of study, as are Canadians by naturalization. As seen with previous literature, for many immigrants, education is seen as a sure path to social mobility which may explain why such students and their  parents choose fields such as engineering and business, specifically  among  Asian-Americans.  Unfortunately,  available  data  does  not  allow to  assess student origins, therefore additional research may shed light on the relationship between specific origins of non domestic students and their choice of field among the cultural and professional undergraduate programs. Indeed, in keeping with the liberated theory, effects of SES on field of study are weak, however; gender and ethnicity are not. This may be due to gender and ethnicity being ascribed and inevitable attributes, whereas the effects of SES are less apparent and more malleable as social  mobility  allows  individuals  to  transcend  their  family’s  social,  cultural  and  financial background. In sum, the relationship between SES and academic field is an interesting one that proves to be more complex than initially expected. Most salient results include the fact that lower SES children are found to favour more lucrative fields and that family SES is found to have differential effects on men and women and for racial/ethnic minorities and whites. Contrary to what EMI would argue, if the choice of a lucrative field bestows advantages for undergraduate students,  then  lower  SES  students  would  seek  those  advantages  more  so  than  higher  SES students. In this way, choice of field could be considered as the unique mechanism to weaken the intergenerational transmission of inequality. Therefore, the increasing importance of what one studies as opposed to where one studies might occur to lower class students as well as higher class ones. It may be that all students, regardless of SES and given that they acquire the 
53necessary resources (through government loans for instance), seek to secure not only a certain level of education, but a type of education. The preliminary results presented in this paper offer an interesting first look at how EMI and the “liberated” theory may lack depth in considering all factors  influencing  choice  of  academic  field  among  students  from  different  socioeconomic backgrounds. Nevertheless, this study is only one step in the understanding of the relationship between SES and field of study. Future research might address this relationship longitudinally, looking at the influence of family background on student’s choice of field throughout different levels of education. This would allow to fully examine whether or not there is in fact a persisting or rather waning influence of class, among other things, on educational choice. Additional research might also be conducted to identify the specific origins of landed immigrants and naturalized Canadians in order to have a deeper understanding of patterns and differences in choice of field among various ethnicities. Based on the current findings, more important would be additional research using more precise data of family income in order to assess subtle differences that the current data might not have grasped. Although RESP and government loans may suggest  a family’s financial situation, the current research found that a large portion of the sample was found within the “middle class” (those who do not receive any financial support) and a more accurate account of their gross income may allow for a deeper understanding of the relationship between SES and choice of field. Finally, findings suggest that class influence may not be as strong as that of gender and ethnicity. In other words, the effect of class may wane over time, but that of ascribed attributes such as gender and ethnicity do not. Although this study only provides speculations concerning the lasting influence of these different variables on the choice of field, a look at 
54associations between class and the latter attributes may contribute to an enhanced understanding of the processes at hand. 55BibliographyAndres, Lesley and Adamuti-Trache, M. 2007. “You’ve Come a Long Way Baby? Persistent Gender Inequality in University Enrolment and Completion in Canada, 1979-2004” Canadian Public Policy, 33(1)Aschaffenburg, Karen, and Ineke Maas. 1997. ‘Cultural and Educational Careers: The Dynamics of Social Reproduction.’ American Sociological Review 62, 4: 573–87  Ball, S. J., Bowe, R., and Gewirtz, S. 1996. 'School choice, social class and distinction: the realization of social advantage in education'. Journal of Education Policy, 11(1): 89-112. Ball, S. J., Davies, J., David, M., and Reay, D. 2002. ''Classification' and 'Judgement': Social Class and the 'Cognitive Structures' of Choice of Higher Education'. British Journal of Sociology of Education, 23(1): 51-72. Ball, S. J. 2010. 'New class inequalities in education'. The International Journal of Sociology and Social Policy, 30(3/4): 155-166. Beck, U., and Beck-Gernsheim, E. 2002. Individualization: institutionalized individualism and its social and political consequences. London: Sage. Bonvillian, Gary and Robert Murphy. 1996. The Liberal Arts College Adapting to Change: The Survival of Small Schools. New York: Garland Publishing.  Bradley, Karen. 2000. “The Incorporation of Women into Higher Education: Paradoxical Outcomes?” Sociology of Education 73:1–18.  Breen, Richard Ruud Luijkx, Walter Müller, and Reinhard Pollak. 2009. ‘Long-term Trends in Educational Inequality in Europe: Class Inequalities and Gender Differences’, European Sociological Review  Breen, Richard and John Goldthorpe. 1997. “Explaining Educational Differentials: Towards a Formal Rational Action Theory.” Rationality and Society 9 (3):275–305.  Brint, Steven. 1998. Schools and Society. Thousand Oaks, CA: Pine Forge Press. Buchmann, Claudia, and Thomas A. DiPrete. 2006. “The Growing Female Advantage in College Completion: The Role of Family Background and Academic Achievement.” American Sociological Review 71:515–41. 
56Charles, Maria, and Karen Bradley. 2002. “Equal but Separate: A Cross-national Study of Sex Segregation in Higher Education.” American Sociological Review 67:573–99.  Charles, Maria, and Karen Bradley. 2006. “A Matter of Degrees: Female Underrepresentation in Computer Science Programs Cross-nationally.” Pp. 183–203 in Women and Information Technology: Research on the Reasons for Underrepresentation, edited by Joanne McGrath Cohoon and Bill Aspray. Cambridge, Mass.: MIT Press. Charles, Maria, and Karen Bradley. 2009. ‘Indulging Our Gendered Selves? Sex Segregation by Field of Study in 44 Countries’, American Journal of Sociology 114, 4: 924–76  Clark, T. N., and Lipset, S. M. 1991. 'Are social classes dying?'. International Sociology, 6(4): 397-410. Davies, Scott. and Neil Guppy. 1997. ‘Field of Study, College Selectivity, and Inequalities in Higher Education’, Social Forces 75, 4: 1417–38 Davies, Scott and Floyd M. Hammack. 2005. “The Channeling of Student Competition in Higher Education: Comparing Canada and the U.S.” The Journal of Higher Education, Vol. 76, No. 1 pp.89-106 Davies, Scott, and David Zarifa. 2012. ‘The Stratification of Universities: Horizontal Inequality in Canada and the United States’. Research in Social Stratification and Mobility 30(2): 143-158  Davies, Scott and Neil Guppy. 2014. The Schooled Society: An Introduction to the Sociology of Education, 3rd edition, Toronto: Oxford University Press, pp. xiv-333.Davies, Scott, Vicky Maldonado, and David Zarifa. 2014. “Effectively Maintaining Inequality in Toronto: Predicting Student Destinations in Ontario Universities” Canadian Sociological Review 51(5): 22-53 Dryer, Helen. 1998. “Parental Role Models, Gender and Educational Choice” The British Journal of Sociology 49(3): 375-398  England, Paula, Paul Allison, Su Li, et al. 2007. “Why Are Some Academic Fields Tipping toward Female? The Sex Composition of U.S. Fields of Doctoral Degree Receipt, 1971–2002.” Sociology of Education 80:23–42. 
57Ethington, C.A. and J.C. Smart. 1986. “Persistence to Graduate Education.” Research in Higher Education 24:287–303.  Ferguson, Sarah Jane and Shunji Wang. 2014. “Graduating in Canada: Profile, Labor Market Outcomes and Student  Debt  of  the  Class  of  2009-2010” Statistics  Canada,  Catalogue no. 81-595-M – No. 2014101Goldthorpe, John H. 1996. “Class Analysis and the Reorientation of Class Theory: The Case of Persisting Differentials in Educational Attainment.” British Journal of Sociology 47 (3):481–505. Goyette, K.A. and A.L. Mullen. 2006. “Who Studies Arts and Sciences? Social Background and the Choice and Consequences of Undergraduate Field of Study.” The Journal of Higher Education 77(3):497–538.Guppy,  Neil,  E.  Grabb,  and  C.  Mollica.  2013.  “The  Canada  Foundation  for  Innovation, Sociology of Knowledge, and the Re-engineering of the University” Canadian Public Policy Vol. 39(1), 1-19, 2013.  Hout, Michael. 1984. “Status, Autonomy, and Training in Occupational Mobility.” American Journal of Sociology 89 (6): 1397–1409.Hout,  Michael.1988.  “More  Universalism,  Less  Structural  Mobility:  The  American Occupational Structure in the 1980s.” American Sociological Review 93 (6): 1358–1400. Hout,  Michael,  Adrian  Raftery,  and  Eleanor  Bell.  1993.  “Making  the  Grade:  Educational Stratification  in  the  United  States,  1925–1989.”  Pp.  25–50  in  Persistent  Inequality: Educational  Stratification  in  Thirteen  Countries,  edited  by  Yossi  Shavit  and  Hans-Peter Blossfeld. Boulder: Westview.Hunt, Jennifer. 2010. “Why do Women Leave Science and Engineering?” National Bureau of Economic Research, Working paper series, Cambridge, MA. Jacobs, J. A. 1986. “The sex-segregation of fields of study: Trends during the college years”. Journal of Higher Education, 57(2), 134-154. Jacobs, J.  A. 1995. “Gender and academic specialities:  Trends among recipients of college degrees in the 1980s”. Cosmology of Education, 68(2), 81-98.Jacobs, Jerry. 1996. "Gender Inequality and Higher Education." Annual Review of Sociology 22:153-85. 
58Jacobs, Jerry. 2003. “Detours on the Road to Equality: Women, Work and Higher Education.” Contexts 2:32–41.Johnson, Monica Kirkpatrick. 2002. “Social Origins, Adolescent Experiences, and Work Value Trajectories during the Transition to Adulthood.” Social Force 80 (4):1307–41. Kohn, Melvin and Carmi Schooler. 1983. Work and Personality: An Inquiry into the Impact of Social Stratification. Norwood, NJ: Ablex. Kelly,  Gail  P.  and  Shelia  Slaughter.  1991.  Women's  Higher  Education  in  Comparative Perspective. Dordrecht,Netherlands:Kluwer. Lee, J., & Jablin, F. M. 1992. A cross-cultural investigation of exit, voice, loyalty and neglect as responses to dissatisfying job conditions. Journal of Business Communication, 29, 203–228. Lee,  J.,  &  Zhou,  M.  2013.  “Frames  of  achievement  and  opportunity  horizons:  Second-generation Chinese, Vietnamese, and Mexicans in Los Angeles”. In D. Card & S. Raphael (Eds.), Immigration, poverty, and socioeconomic inequality (pp. 206–231). New York: Russell Sage Foundation. Lucas,  Samuel.  2001.  ‘Effectively  Maintained  Inequality:  Education  Transitions,  Track Mobility, and Social Background Effects’, American Journal of Sociology 106, 6: 1642–90. Lyons, S. T., Eddy S. W. Ng, L., Schweitzer. 2010. “New Generation, Great Expectations: A Field Study of the Millennial Generation” J Bus Psychol 25:281–292 Ma, Yingyi. 2009. “Family Socioeconomic Status, Parental Involvement, and College Major Choices—Gender, Race/ Ethnic, and Nativity Patterns”, Sociological Perspectives, (52)2, pp. 211-234 Maple, S. A., and Stage, F. K. (1991). “Influences on the choice of math/science major by gender and ethnicity.” American Educational Research Journal, 28(1), 37-60. Mare, Robert, D. (1980) “Social Background and School Continuation Decisions” Journal of the American Statistical Association, Vol. 75, No. 370, pp. 295- 305Mare,  Robert,  D.  (1981)  “Change  and  Stability  in  Educational  Stratification”American Sociological Review, Vol. 46, No. 1, pp. 72-87 Marini, Margaret M., Pi-Ling Fan, Erica Finley, and Ann M. Beutel. 1996. “Gender and Job Values.” Sociology of Education 69:49–65. 
59Mortimer, Jeylan T., Jon Lorence, and Donald Kumla. 1986. Work, Family and Personality: Transition to Adulthood. Norwood, NJ: Ablex. 
Mullen,  A.  L.  (2001).  “Gender,  race  and  the  college  science  track:  An  analysis  of  field concentrations and institutional selectivity.” Journal of Women and Minorities in science and Engineering, 7, 287-302. Mullen, A.L., K.A. Goyette and J.A. Soares. 2003. “Who Goes to Graduate School? Social and Academic  Correlates  of  Educational  Continuation  after  College.”  Sociology  of  Education 76(2):143–69. Ostrovsky, Yuri and Marc Frenette. 2014. “The Cumulative Earnings of Postsecondary Graduates Over 20 years: Results by Field of Study”, Statistics Canada no. 11626-X, Economic Insight, no. 040Pascarella, E. T., & Terenzini, P.T. (1991) How college affects students. San Francisco: Jossey-Bass Raftery, Adrian and Michael Hout. 1993. “Maximally Maintained Inequality: Expansion, Reform, and Opportunity in Irish Education, 1921-75”, Sociology of Education, Vol. 66, No. 1Reay, D., Davies, J., David, M., and Ball, S. J. (2001) 'Choices of Degree or Degrees of Choice? Class, ‘Race’ and the Higher Education Choice Process'. Sociology, 35(04): 855-874.Schofer, Evan, and John W. Meyer. 2005. “The Worldwide Expansion of Higher Education in the Twentieth Century.” American Sociological Review 70:898–920. Shauman, Kimberlee A. 2006. “Occupational Sex Segregation and the Earnings of Occupations: What Causes the Link among College-Educated Workers?” Social Science Research 35:577–619.Shavit, Y. and H.P. Blossfeld, eds. 1993. Persistent Inequality: Changing Educational Attain- ment in Thirteen Countries. Boulder, CO: Westview Press. Shavit, Yossi, Richard Arum, and Adam Gamoran, eds. 2007. Stratification in Higher Education: A Comparative Study. Stanford, Calif.: Stanford University Press. Simpson, J. C. 2001. “Segregated by Subject: Racial Differences in the Factors Influencing Academic Major between European Americans, Asian Americans, and African, Hispanic, and Native Americans” The Journal of Higher Education, 72(1) 63-100 
60Somerville, K., & Walsworth, S. (2009). Vulnerabilities of highly skilled immigrants in Canada and the United States. American Review of Canadian Studies, 39, 147–161. Statistic Canada “National Graduate Survey, 2013 (Class of 2009-2010)” Statistics Canada, “Survey of Approaches to Educational Planning, 2013”, The Daily, Wednesday, October 29, 2014 Steinberg, L. (1996). “Ethnicity and Adolescent Achievement”. American Educator Summer, 20(2), 28–48. Sue, S., & Okazaki, S. (1990). “Asian-American educational achievements: A phenomenon in search of an explanation”. American Psychologist, 45(8), 913–920. Torche, Florencia. 2011. “Is a College Degree Still the Great Equalizer? Intergenerational Mobility across Levels of Schooling in the United States”, American Journal of Sociology, 117(3), pp. 763-807 Turner, S., E., and Bowen, W. G. 1999. “Choice of Major: The Changing (Unchanging) Gender Gap”, Industrial and Labor Relations Review, 52(2) Van de Werfhorst, Herman G. 2001. Field of Study and Social Inequality: Four types of Educational Resources in the Process of Stratification in the Netherlands,  205p.Van Langen, Annemarie, and Hetty Dekkers. 2005. “Cross-national Differences in Participating in Tertiary Science, Technology, Engineering, and Mathematics Education.” Comparative Education 41:329–50.Wilson, K.L (1978) “Toward an improved explanation of income attainment: recalibrating education and occupation”. American Journal of Sociology, 84(3), 684-697.Wilson, K.L and Smith-Lovin, L. (1983). Scaling the prestige, authority, and income potential of college curricula”. Social Science Research, 12(2), 159-186.Wright, Laura, David Walters, and David Zafira. 2013. “Government Student Loan Default: Differences between Graduates of the Liberal Arts and Applied Fields in Canadian Colleges and Universities”, Canadian Sociological Review 50(1): 89-115 Xie, Y., and Goyette, K. (1999). “Education Expectations of Asian American Youths: Determinants and Ethnic Differences” Sociology of Education 72(1), 22-36 
61Xie, Y., and Goyette, K. (2003). “Social mobility and the educational choices of Asian Americans”. Social Science Research, 32(3), 467-498. Xie, Yu, and Kimberlee A. Shauman. 2003. Women in Science: Career Processes and Outcomes. Cambridge, Mass.: Harvard University Press.  Zafar, Basit. 2013. “College Major Choice and the Gender Gap”, Journal of Human Resources,  48(3), pp.545-595 Zarifa, David. 2012. “Persistent Inequality or Liberation from Social Origins? Determining Who Attends Graduate and Professional Schools in Canada’s Expanded Postsecondary System”, Canadian Sociological Review, 49(2): 109-137 62Appendix IStatistic Canada’s estimated gross annual earnings of 2009-2010 graduates with a bachelor degree working full-time in 2013, by selected fields of study63Appendix IIThe following cross-tables present the descriptive statistics of the association between parent’s level of education (mother’s and father’s are presented separately) and whether or not the student receives a RESP and a governmental loan.Table 11: Father’s level of education and whether or not student receives a RESP and governmental loans for the 2005 cohort.Table 12: Mother’s level of education and whether or not student receives a RESP and governmental loans for the 2005 cohort.Resp No postsecondary College University TotalNo 2,68542.7393.001,25019.8988.032,34937.3883.656,284100.0088.32Yes 20224.317.0017020.4611.9745955.2316.35831100.0011.68Gov. LoansNo 1,26736.6143.8963418.3244.651,56045.0755.563,461100.0048.64Yes 1,62044.3356.1178621.5155.351,24834.1544.443,654100.0051.36Total 2,88740.58100.001,42019.96100.002,80839.47100.007,115100.00100.00Resp No postsecondary College University TotalNo 2,76443.9892.531,51824.1687.492,00231.8683.666,284100.0088.32Yes 22326.847.4721726.1112.5139147.0516.34831100.0011.68Gov. LoansNo 1,34238.7744.9382323.7847.441,29637.4554.163,461100.0048.64Yes 1,64545.0255.0791224.9652.561,09730.0245.843,654100.0051.36Total 2,98741.98100.001,73524.39100.002,39333.63100.007,115100.00100.0064                                                                                                                                                                Table 13: Father’s level of education and whether or not student receives a RESP and governmental loans for the 2010 cohort.Table 14: Mother’s level of education and whether or not student receives a RESP and governmental loans for the 2010 cohort.Resp Less than High SchoolHigh School Trade CertificatCollege Bachelor’s University grad.TotalNo 63613.7589.961,21526.2684.384439.5881.4359012.7581.721,08023.3574.2866214.3172.994,626100.0080.12Yes 716.1810.0422519.6015.621018.8018.5713211.5018.2837432.5825.7224521.3427.011,148100.0019.88Gov. LoansNo 30410.5343.0067523.3746.882538.7646.5135512.2949.1777226.7353.0952918.3258.322,888100.0050.02Yes 40313.9657.0076526.5153.1229110.0853.4936712.7250.8368223.6346.9137813.1041.682,886100.0049.98Total 70712.24100.001,44024.94100.005449.42100.0072212.50100.001,45425.18100.0090715.71100.005,774100.00100.00Resp Less than High SchoolHigh School Trade CertificatCollege Bachelor’s University grad.TotalNo 4259.1994.031,46331.6384.962194.7386.9090119.4879.731,20626.0774.264128.9169.364,626100.0080.12Yes 272.355.9725922.5615.04332.8713.1022919.9520.2741836.4125.7418215.8530.641,148100.0019.88Gov. LoansNo 2047.0645.1380827.9846.921164.0246.0352218.0746.1989831.0955.3034011.7757.242,888100.0050.02Yes 2488.5954.8791431.6753.081364.7153.9760721.0753.8172625.1644.702548.8042.762,886100.0049.98Total 4527.83100.001,72229.82100.002524.36100.001,13019.57100.001,62428.13100.0059410.29100.005,774100.00100.0065Appendix IIIThe following tables present the descriptive statistics of the association between gender and whether or not the student receives a RESP and a government student loan.2005 cohortTable 15: Gender and financial aid for students within the hard sciences in the 2005 cohortTable 16: Gender and financial aid for students within the cultural fields in the 2005 cohort                Soft Sciences                                 Hard Sciences Hard SciencesRESP         Yes           |          No                                 Yes           |          NoFemale          310                    2,450             179                    1,291Male          144                    1,165             182                    1,218Government Loans         Yes           |          No                                 Yes           |          NoFemale          1,422                 1,338             758                     712Male           657                    652             718                     682                Cultural fields OtherRESP         Yes           |          No                                 Yes           |          NoFemale           48                      266             441                    3,475         Male           14                      119             312                    2,264                Cultural fields OtherGovernment Loans         Yes           |          No                                 Yes           |          NoFemale          167                     147                 2,013                 1,903Male           78                       55                           1,297                 1,279   66Table 17: Gender and financial aid for students within the professional fields in the 2005 cohortTable 18: Gender and financial aid by average income in the 2005 cohort             Professional fields OtherRESP         Yes           |          No                                 Yes           |          NoFemale          138                    1,134             351                    2,607Male          142                     911             184                    1,472             Professional fields OtherGovernment loans          Yes           |          No                                 Yes           |          NoFemale          623                     649            1,557                  1,401Male          525                     528             850                     806$38,000 - $48,000 $50,000 - $53,000 $56,000 - $65,000RESP    Yes       |        No                        Yes       |        No               Yes       |       NoFemale     172             1,044       227              1,943         90               754Male      77               579       114                937        135              867$38,000 - $48,000 $50,000 - $53,000 $56,000 - $65,000Government loans    Yes       |        No                  Yes       |        No              Yes       |       NoFemale     617               599       1,106            1,064        457              387Male     340               316         516               535        519              483672010 cohortTable 19: Gender and financial aid for students within the hard sciences in the 2010 cohortTable 20: Gender and financial aid for students within the cultural fields in the 2010 cohort                Soft Sciences                                 Hard Sciences RESP         Yes           |          No                                Yes           |          NoFemale          404                   1,828            259                     944Male          226                     952            259                     902Government Loans         Yes           |          No                                Yes           |          NoFemale         1,100                   1,132            662                     541Male          563                      615            561                     600                 Cultural Fields                                               OtherRESP         Yes           |          No                                Yes           |          NoFemale           51                      215            612                    2,557Male           23                      105            462                    1,749Government Loans         Yes           |          No                                Yes           |          NoFemale          136                     130           1,626                  1,543Male           68                       60           1,056                  1,15568Table 21: Gender and financial aid for students within the professional fields in the 2010 cohortTable 22: Gender and financial aid by average income in the 2010 cohortProfessional Fields OtherRESP         Yes           |          No                                Yes           |          NoFemale         192                      902            471                    1,870Male         203                      705            282                    1,149Government Loans         Yes           |          No                                Yes           |          NoFemale          543                     551           1,219                   1,122Male          421                     487             703                      728$38,000 - $48,000 $50,000 - $53,000 $56,000 - $65,000RESP      Yes       |        No                        Yes       |        No               Yes       |       NoFemale       225               827          298              1,373       140               572Male      136                493         180                754       169               607$38,000 - $48,000 $50,000 - $53,000 $56,000 - $65,000Government loans      Yes       |        No                       Yes       |        No              Yes       |       NoFemale       557               495          811               860       394              318Male       310               319         434                500       380              39669

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:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0216544/manifest

Comment

Related Items