Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Social rank judgments on Facebook and their emotional consequences as a function of social anxiety Parsons, Carly A. 2016

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

Item Metadata

Download

Media
24-ubc_2016_september_parsons_carly.pdf [ 435.46kB ]
Metadata
JSON: 24-1.0308071.json
JSON-LD: 24-1.0308071-ld.json
RDF/XML (Pretty): 24-1.0308071-rdf.xml
RDF/JSON: 24-1.0308071-rdf.json
Turtle: 24-1.0308071-turtle.txt
N-Triples: 24-1.0308071-rdf-ntriples.txt
Original Record: 24-1.0308071-source.json
Full Text
24-1.0308071-fulltext.txt
Citation
24-1.0308071.ris

Full Text

    Social Rank Judgments on Facebook and Their Emotional Consequences  as a Function of Social Anxiety  by Carly A. Parsons B.Sc., Queen’s University, 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 (Psychology)  The University of British Columbia (Vancouver)  August 2016  © Carly A. Parsons, 2016   iiAbstract Socially anxious individuals are vigilant to social hierarchies, and there is preliminary evidence that they perceive dominant others to be higher in social rank than do those who are not socially anxious.  Other research has indicated that socially anxious individuals spend a great deal of time on the Internet, that they prefer to engage in passive activity on social networking sites, and that both of these behaviours confer emotional risks.  The current study aimed to connect these research areas by determining whether socially anxious individuals engage in a social rank judgment bias on Facebook and whether this might account for the negative emotional consequences they experience.  Two hundred twelve undergraduate students were each presented with 10 randomly selected Facebook profiles of unknown peers and rated each peer on traits related to social rank.  They also completed state measures of self-esteem and affect at the beginning and end of the study.  Participants with higher trait levels of social anxiety experienced greater decreases in state self-esteem and positive affect after having rated the unknown peers’ profiles.  Contrary to expectations, however, these emotional outcomes were not accounted for by a social rank judgment bias, as social anxiety did not predict higher social rank ratings.  These results extend previous research by illustrating the emotional impact of online social judgments for socially anxious individuals.  The findings are discussed in relation to the literatures on social anxiety, social rank judgments, and social networking site use, and other potential mechanisms of this emotional impact are considered.      iii Preface This thesis is an original intellectual product of the author, who was responsible for identification of the research question, study design, data collection, and data analysis.  Lynn E. Alden acted as the project supervisor and assisted with research question development, study design, and data interpretation.  Jeremy C. Biesanz provided study stimuli, facilitated data analysis through the creation of multilevel models, and assisted with analytical design and interpretation.  Furthermore, two of his undergraduate research assistants from the Social Accuracy Lab assisted with the setup and timing of experimental sessions.  The study procedure was approved by the UBC Behavioural Research Ethics Board under the project title “Impression Formation on Facebook” (certificate number H15-00420).  Preliminary results were presented as a poster: Parsons, C. A., Alden, L. E., & Biesanz, J. C. (2016, June). Impression formation on  Facebook as a function of social anxiety. Poster presented at the annual convention of the Canadian Psychological Association, Victoria, BC.            ivTable of Contents Abstract ..................................................................................................................................... ii Preface ..................................................................................................................................... iii Table of Contents ..................................................................................................................... iv List of Tables ........................................................................................................................... vi List of Figures ......................................................................................................................... vii Acknowledgements................................................................................................................ viii  Introduction................................................................................................................................1 Overview................................................................................................................................1 Social Anxiety Disorder ........................................................................................................2 Impression Management in Social Anxiety ...........................................................................3 Social Rank and Affiliation Systems .....................................................................................5 Social rank system in social anxiety. .................................................................................5 Affiliation system in social anxiety. ..................................................................................8 Internet and Social Networking Sites ....................................................................................9 Preference for online social interaction. ..........................................................................10 Positive and negative effects of Internet use. ..................................................................12 Passive social networking site use and social comparison. .............................................15 Social Rank Judgments on Facebook ..................................................................................18 The Current Study................................................................................................................19 Primary analyses. .............................................................................................................20 Secondary analyses. .........................................................................................................21  Method .....................................................................................................................................22 Participants ..........................................................................................................................22 Materials ..............................................................................................................................22 Target Facebook profiles. ................................................................................................22 Measures ..............................................................................................................................23 State Self-Esteem Scale, social self-esteem subscale. .....................................................23 The International Positive and Negative Affect Schedule Short Form, adapted. ............24 Explicit Domain-Specific Social Estimations scale, adapted. .........................................25 Straightforward Social Interaction Anxiety Scale. ..........................................................26 Procedure .............................................................................................................................27   vAnalyses ...............................................................................................................................31 Data preparation...............................................................................................................31 Analytic approach. ...........................................................................................................31 Power analysis and assumptions. .....................................................................................32 Missing data. ....................................................................................................................33  Results .....................................................................................................................................34 Descriptive Statistics ...........................................................................................................34 Preliminary Analyses ...........................................................................................................37 Bivariate correlations. ......................................................................................................37 Between-group comparisons. ..........................................................................................39 Primary Analyses .................................................................................................................42 Social rank ratings. ..........................................................................................................42 Change in state self-esteem. ............................................................................................45 Change in positive affect. ................................................................................................48 Change in negative affect. ...............................................................................................51 Mediation. ........................................................................................................................53 Secondary Analyses .............................................................................................................53 Affiliation ratings.............................................................................................................53 Time per page. .................................................................................................................55   Discussion ................................................................................................................................57 Social Rank Judgment Bias .................................................................................................57 Emotional Consequences of Profile Browsing ....................................................................59 Potential Mechanisms Underlying Emotional Consequences .............................................61 Affiliation Judgment Bias ....................................................................................................63 Information-Seeking Bias ....................................................................................................64 Covariate Effects .................................................................................................................65 Strengths and Future Directions ..........................................................................................67 Conclusions .........................................................................................................................68  References................................................................................................................................70  Appendices ..............................................................................................................................84 Appendix A: Protocol for Preparing Facebook Profile PDF Files ..................................84 Appendix B: State Self-Esteem Scale, Social Self-Esteem Subscale. .............................85 Appendix C: The International Positive and Negative Affect Schedule Short Form, Adapted. ...........................................................................................................................86 Appendix D: Explicit Domain-Specific Social Estimations Scale, Adapted. .................87 Appendix E: Straightforward Social Interaction Anxiety Scale. .....................................88 Appendix F: Demographics Questionnaire. ....................................................................89     viList of Tables Table 1. Proportions of the Total Sample Presented with Each Binder..................................30 Table 2. Descriptive Statistics for Participant-Level Variables..............................................35 Table 3. Descriptive Statistics for Target-Level Variables.....................................................36 Table 4. Bivariate Correlations Between Participant-Level Variables....................................38 Table 5. Difference in Social Anxiety Between Genders........................................................40 Table 6. Differences in Participant-Level Variables Between Cultural Groups.....................41 Table 7. Multilevel Model Predicting Target Social Rank Ratings........................................44 Table 8. Multiple Regression Model Predicting State Self-Esteem at Time 2........................46 Table 9. Multiple Regression Model Predicting Positive Affect at Time 2............................49 Table 10. Multiple Regression Model Predicting Negative Affect at Time 2.........................52 Table 11. Multilevel Model Predicting Target Affiliation Ratings.........................................54 Table 12. Multiple Regression Model Predicting Time Per Page...........................................56             viiList of Figures Figure 1. Effect of Social Anxiety on Change in State Self-Esteem.......................................47 Figure 2. Effect of Social Anxiety on Change in Positive Affect...........................................50                       viii Acknowledgements I would like to thank my supervisor, Dr. Lynn Alden, for her invaluable guidance throughout the development and design of this project.  I also extend my gratitude to my committee members, Drs. Amori Mikami and Jeremy Biesanz.  Both provided valuable feedback and insights at the time of the study proposal, and Jeremy was instrumental in his provision of study materials and assistance with data analysis.  I would like to acknowledge both the Social Sciences and Humanities Research Council and the Faculty of Arts of the University of British Columbia for providing financial resources with which to fund this project.  Finally, I wish to thank my loved ones and fellow UBC Psychology graduate students for their unconditional support throughout both years of my Master’s program.        1Introduction The current study draws on the interpersonal circumplex model to examine how social anxiety impacts social judgments on social networking sites and the emotional consequences that result.  The manuscript begins with a brief overview of the rationale for the study, followed by a detailed consideration of the extant literature on social anxiety, social rank judgments, and problematic use of social networking sites.  A description of the study procedure and results follow.  The manuscript ends with a discussion of the findings, noting research limitations and pointing to future directions for further study. Overview  Socially anxious individuals are theorized to overutilize the social rank axis of the interpersonal circumplex (Trower & Gilbert, 1989).  Research in support of this theory has shown that these individuals perceive others, particularly those who appear dominant, to be higher in social rank than do those who are not socially anxious (Aderka, Haker, Marom, Hermesh, & Gilboa-Schechtman, 2013; Haker, Aderka, Marom, Hermesh, & Gilboa-Schechtman, 2014).  This social rank judgment bias is important to understand, as socially anxious individuals are particularly sensitive to perceptions of others as superior, suffering greater decreases in self-evaluations and affect as a result (e.g., Antony, Rowa, Liss, Swallow, & Swinson, 2005).  As it stands, though, evidence of a social rank judgment bias has been accrued only in relatively artificial contexts (i.e., vignette studies). A separate line of research has shown that individuals with high levels of social anxiety prefer interacting with others online rather than face-to-face (e.g., Caplan, 2003).  Contrary to the ostensible benefits of online social interaction, research has demonstrated that this preference gives way to a host of deleterious outcomes for socially anxious   2individuals, including social isolation, depressive symptoms, and exacerbated social anxiety (e.g., Weidman et al., 2012).  As such, it has become important to explore the online behaviours of socially anxious individuals that may contribute to these negative outcomes.  Online social interaction often takes place over social networking sites (SNS), and of the various activities made possible on these platforms, socially anxious individuals appear to spend more time passively consuming content (e.g., browsing others’ profiles) than actively communicating with others (Shaw, Timpano, Tran, & Joormann, 2015).  Furthermore, passive content consumption may be predominated by social comparison; on Facebook, the two activities are associated with similar negative emotional outcomes, especially for populations who are particularly vulnerable to social comparison (Vogel, Rose, Okdie, Eckles, & Franz, 2015).  As socially anxious individuals represent one such population, the nature of their social comparison and other passive behaviours on SNS needs to be explored.  In light of preliminary evidence of biased social rank judgments in this population, the current study was conducted to test the hypothesis that such judgments underlie socially anxious individuals’ profile browsing on Facebook and thereby account for the negative emotional consequences that result from their Internet and SNS use. Social Anxiety Disorder Social anxiety disorder (SAD) is characterized by an excessive and persistent fear of social or performance situations, particularly those in which there is a possibility of scrutiny by others (American Psychiatric Association, 2013).  An individual with SAD either avoids these situations or endures them with significant anxiety and distress.  SAD is a relatively common anxiety disorder in Western countries, with an estimated lifetime prevalence rate of 13% in the United States (Kessler, Petukhova, Sampson, Zaslavsky, & Wittchen, 2012).    3Individuals with SAD experience distress in numerous forms of interpersonal interactions (Turk, Heimberg, & Hope, 2001) and, as a result, have fewer social relationships than do other people, including those with other anxiety disorders (Mendlowicz & Stein, 2000).  Their relatively few close relationships are also of lower quality: Individuals with SAD report lower perceived social support (Davidson, Hughes, George, & Blazer, 1994), greater dissatisfaction with existing friendships (Rodebaugh, 2009), and lower-quality romantic relationships characterized by less intimacy, emotional expression, and self-disclosure (Sparrevohn & Rapee, 2009).  The interpersonal difficulties of socially anxious individuals result in impairment in many other aspects of their daily lives (e.g., occupational and recreational; e.g., Alden & Taylor, 2004). Importantly, social anxiety is not only experienced in the context of a diagnosable mental disorder.  Arguably, it is more accurately conceptualized as a trait that exists along a continuum (e.g., Heiser, Turner, Beidel, & Roberson-Nay, 2009).  Individuals with moderate to high levels of social anxiety who do not meet SAD criteria demonstrate similar—albeit less severe—impairments and difficulties as compared to those who do meet criteria for diagnosis.  Accordingly, much can be learned about social anxiety from research using non-clinical samples.  This idea is elaborated upon through research cited in the ensuing sections.  Impression Management in Social Anxiety Fear of negative evaluation is known to be a central feature of SAD (e.g., Leary & Kowalski, 1997; Schlenker & Leary, 1982).  Socially anxious individuals tend to perceive others as inherently critical (Leary, Kowalski, & Campbell, 1988) and to worry about others’ impressions of them both when engaging in and when anticipating social interactions (Gilbert, 2001; Leary, 2010).  In fact, research suggests that socially anxious individuals also   4fear positive evaluation, due to concerns that they will be viewed by others as socially threatening and potentially ostracized as a result (Weeks, Jakatdar, & Heimberg, 2010; Weeks, Rodebaugh, Heimberg, Norton, & Jakatdar, 2009).  Their fears of positive and of negative evaluation are distinct—each one contributing unique variance to social anxiety as a whole—but strongly correlated, suggesting a higher-order fear of evaluation in general (Weeks et al., 2010; Weeks, Heimberg, Rodebaugh, Goldin, & Gross, 2012; Weeks, Heimberg, Rodebaugh, & Norton, 2008; Weeks & Howell, 2012). From a dispositional perspective, social anxiety has been conceptualized on a continuum from shyness to intense fear of negative evaluation (Heiser et al., 2009).  According to some researchers, social anxiety is elicited when individuals are motivated to make particular impressions on others but doubt their ability to do so successfully (e.g., High & Caplan, 2009; Leary et al., 1988).  Individuals who are dispositionally socially anxious are more concerned about the impressions that others make of them, believe these impressions are more important, and expect to make less favourable impressions than do those who are less, or not, socially anxious (Leary et al., 1988).  These low expectations seem to be founded on negative self-perceptions rather than on actual knowledge of others’ perceptions of them (Christensen, Stein, & Means-Christensen, 2003). Although much of the extant literature on social anxiety and SAD has examined the ways in which socially anxious individuals perceive—and attempt to manage—others’ impressions of them, comparatively less has examined the impressions that these individuals form of others.  These latter impressions are important to elucidate, as they likely play a critical role in shaping the cognitions and behaviours of socially anxious people in social interactions.  Researchers have begun to address this gap using the framework of the   5interpersonal circumplex. Social Rank and Affiliation Systems   Two main systems are postulated to underlie the interpersonal world, forming the axes of the interpersonal circumplex.  These axes have been assigned various names, including the dominance, status, or control axis and the nurturance, warmth, or affiliation axis (Gurtman, 1991).  The interpersonal circumplex has proven to be a useful framework for conceptualizing interpersonal behaviours (Kiesler, 1983), personality traits (Gurtman, 1991), and construals of the social world (Gilbert, 2004; Gilbert & Trower, 2001; Irons & Gilbert, 2005; Trower & Gilbert, 1989).  With regard to such construals—and within this manuscript—the systems comprising the axes of the interpersonal circumplex are referred to as the social rank and affiliation systems.  Under the social rank system, the social world is organized into hierarchies (Aderka et al., 2013), which individuals monitor with the goal of achieving a “dominant” position.  Dominance refers either to high amounts of power or of social attractiveness.  From an evolutionary perspective, this system is adaptive as it permits better access to resources and alerts individuals to potential social threats (Trower & Gilbert, 1989).  Furthermore, it seems to be governed by a specialized neural system (Chiao, 2010; Sapolsky, 2005).  Under the affiliation system, the social world is organized into support networks (Aderka et al., 2013), which individuals monitor with the goals of connecting and cooperating with others.  This system is adaptive as it inhibits hostility and encourages resource sharing and skill development (Trower & Gilbert, 1989).   Social rank system in social anxiety. According to Trower and Gilbert’s evolutionary theory (1989), although all individuals use both systems, socially anxious individuals overutilize the social rank system and underutilize the affiliation system.  In other   6words, individuals high in social anxiety show a bias toward viewing the world in hierarchical and competitive terms, where dominant social status is desirable but difficult or impossible to achieve.  Indeed, research has confirmed that socially anxious people view social interactions as more competitive than do non-anxious individuals (Hope, Sigler, Penn, & Meier, 1998) and constantly monitor their environments for signs of social threat (Gilboa-Schechtman, Foa, & Amir, 1999; Rapee & Heimberg, 1997).  This appears not to be a general sensitivity to social rank information, as socially anxious individuals view others as “hostile dominants” (Trower & Gilbert, 1989, p.19) and are particularly attuned to indications of high social rank.  In two recent studies, after reading descriptions of protagonists depicted as either dominant or submissive, participants with SAD rated dominant protagonists as higher in social rank than did control participants, but showed no bias in ratings of submissive protagonists (or for neutral protagonists; Aderka et al., 2013; Haker et al., 2014).  In addition, participants with SAD showed an information-seeking bias in that they sought out less information before indicating their impressions of protagonists, especially with regard to social rank-related traits (Aderka et al., 2013).  Aderka et al. (2013) theorize that social rank information is viewed as threatening by socially anxious individuals and therefore activates defensive behaviours, including less information-seeking, as avoidance strategies.  This theory is consistent with an interpersonal model of SAD, which suggests that socially anxious individuals adopt self-protective strategies (e.g., safety behaviours) in situations that they perceive as threatening (Alden & Taylor, 2010). Haker et al. (2014) speculate that constant monitoring of social hierarchies may have adaptive benefits for socially anxious individuals, allowing for more accurate perceptions of social rank and avoidance of potential harm.  This theory is consistent with those regarding   7the fear of positive evaluation, which is proposed to stem from a desire to avoid interpersonal conflict and ostracism (e.g., Weeks et al., 2009).  Unfortunately, in addition to overestimating others’ social standings, socially anxious individuals have also been shown to underestimate their own, to deleterious effect.  In both clinical and community samples, social anxiety has been associated with lower self-perceptions of social rank above and beyond the effects of depressive symptoms, generalized anxiety symptoms, and comorbid affective disorder diagnoses (Aderka, Weisman, Shahar, & Gilboa-Schechtman, 2009; Weisman, Aderka, Marom, Hermesh, & Gilboa-Schechtman, 2011).  Social anxiety is negatively associated with both implicit and explicit measures of social rank self-esteem, which has been shown to predict social anxiety severity over and above global self-esteem (Gilboa-Schechtman, Friedman, Helpman, & Kananov, 2013).  Contrary to Haker et al.’s theory, socially anxious individuals’ overutilization of the social rank system appears to be maladaptive, as it causes them to view themselves as inadequate, inferior, undesirable, and socially unattractive relative to others (e.g., Hope et al., 1998; Leary & Kowalski, 1997).  Therefore, echoing research previously cited, socially anxious individuals view social interactions as competitive events for which they are poorly equipped. In addition to engaging in more frequent upward social comparisons (i.e., perceptions of others as superior to themselves), socially anxious individuals tend to compare themselves to others on a greater number of dimensions, especially those on which they already believe themselves to be inferior (e.g., social skills; Antony et al., 2005).  As a result, the decrease in self-esteem and affect that typically accompanies upward social comparison (e.g., Swallow & Kuiper, 1988) is especially strong for socially anxious individuals, who form poorer self-evaluations (Mitchell & Schmidt, 2014) and experience greater increases in both anxiety and   8depressive symptoms (Aderka et al., 2009; Antony et al., 2005) as a result.  Furthermore, upward social comparison is likely to maintain social anxiety by confirming these individuals’ maladaptive beliefs (e.g., of inadequacy; Aderka et al., 2013; Antony et al., 2005).  Consequently, doubting their abilities to compete with superior others, socially anxious individuals engage in more submissive and fewer dominant verbal and nonverbal behaviours (Galili, Amir, & Gilboa-Schechtman, 2013; Heerey & Kring, 2007; Walters & Hope, 1998; Weeks, Heimberg, & Heuer, 2011) or complete avoidance of social interaction (Weisman et al., 2011).  All of these responses are known to engender negative reactions from others (Alden & Bieling, 1998).  This body of research extends not only Trower and Gilbert’s (1989) evolutionary theory of social rank system overutilization, but also both cognitive (Clark & Wells, 1995) and interpersonal (Alden & Taylor, 2010) theories of SAD: Socially anxious individuals’ core negative schemas may be based primarily on social rank concerns, and their negative self-perceptions of social rank and interpersonal efficacy may hinder the formation of meaningful relationships (Gilboa-Schechtman et al., 2013).   Affiliation system in social anxiety. In contrast to the support for social rank system overutilization, findings regarding underutilization of the affiliation system in SAD have been less consistent.  On one hand, socially anxious individuals do exhibit impaired affiliation-related abilities.  They have difficulty identifying opportunities to connect with others (Aderka et al., 2009) and perceive less intimacy and closeness in their existing peer, friend, and romantic relationships (Rodebaugh, 2009; Sparrevohn & Rapee, 2009; Weisman et al., 2011).  These perceptions appear to reflect true differences in interpersonal dynamics rather than mere perceptual biases, as socially anxious individuals are regarded by observers and conversational partners as less warm and interested (Alden & Wallace, 1995), and even   9their close relationships are characterized by less emotional expression and self-disclosure (Cuming & Rapee, 2010; Sparrevohn & Rapee, 2009).  More recent research focusing on perceptions of others’ affiliation-related traits has been less conclusive.  Consistent with Trower and Gilbert’s (1989) theory, Haker et al.’s (2014) sample of SAD patients rated protagonists as less friendly than did controls; however, Aderka et al.’s (2013) clinical sample responded in the opposite way, rating protagonists as more friendly than did controls.  Unlike the social rank judgments in these studies, these results were not qualified by the descriptions of the protagonists (i.e., whether or not the protagonists were described as friendly).  Explanations have been offered for both biases (e.g., a halo effect of perceived high social standing; Aderka et al., 2013), but more research is needed to determine which bias is more characteristic of socially anxious individuals, or if one exists at all. Internet and Social Networking Sites Thus far, research examining socially anxious individuals’ social rank- and affiliation-related impressions has relied upon self-report (e.g., Weisman et al., 2011) and descriptions of fictional others (Aderka et al., 2013; Haker et al., 2014; Rodebaugh, Bielak, Vidovic, & Moscovitch, 2016).  At this stage, it is important to corroborate the social rank findings, and clarify the affiliation findings, with behavioural observations and in more realistic contexts.  The Internet provides a particularly interesting place to start.  For people in general, at least in the developed world, the Internet has quickly become an essential part of daily life, particularly as a means of communication and socializing (e.g., Correa, Hinsley, & de Zúñiga, 2010).  Such socializing takes place primarily on social networking sites (SNS), which have become popular platforms for establishing and maintaining friendships and therefore for fulfilling the fundamental need to belong (Baumeister & Leary, 1995;   10Grieve, Indian, Witteveen, Tolan, & Marrington, 2013; Quinn & Oldmeadow, 2013).  They also fulfill the need for self-presentation (Nadkarni & Hofmann, 2012) by allowing for the creation of profiles for “friends” or “followers” to examine.  Recent statistics identify Canadians as the second heaviest Internet users worldwide, with 69% having a profile on at least one SNS in 2013 (Canadian Internet Registration Authority, 2014).  Despite the ongoing proliferation of new and diverse SNS, Facebook remains the most widely used, with 59% of Canadians and 75% of Canadian youth specifically estimated to use it (Forum Research, 2015).  Indeed, SNS are especially popular among adolescents and young adults (e.g., Valkenburg & Peter, 2011). Aside from age, several other individual differences have been identified as predictors of SNS-related behaviours and outcomes.  For instance, people who are more extroverted have more Facebook friends (e.g., Moore & McElroy, 2012), and people with lower levels of trait self-esteem spend more time on Facebook and attribute more meaning to its use (Mehdizadeh, 2010).  Of relevance to this paper, symptoms of social anxiety have been explored as a predictor of the frequency of Internet and SNS use, with mixed results: Some studies have found at least marginal positive relationships (Green, Wilhelmsen, Wilmots, Dodd, & Quinn, 2016; Mo et al., 2014; Pierce, 2009; Shaw et al., 2015), while others have found no relationship (Campbell, Cumming, & Hughes, 2006; Fernandez, Levinson, & Rodebaugh, 2012; Muench, Hayes, Kuerbis, & Shao, 2015) or one moderated by other key variables (Mazalin & Moore, 2004; McCord, Rodebaugh, & Levinson, 2014). Preference for online social interaction. Regardless of frequency, it has been established that social anxiety predicts a preference for online social interaction (POSI; Caplan, 2003) relative to face-to-face interaction.  Several reasons for this preference have   11been highlighted.  In essence, socially anxious individuals find online interactions more comfortable than face-to-face interactions.  In fact, the higher an individual’s trait level of social anxiety, the more comfort they report experiencing online (Pierce, 2009; Weidman et al., 2012) and the greater decrease in social anxiety they demonstrate when interacting online instead of face-to-face (Yen et al., 2012).  This is understandable when the differences between online and face-to-face contexts are considered.  Online contexts offer numerous advantages to the socially anxious individual.  For instance, the interaction partner is not physically present, and therefore physical attractiveness is less salient and the need for eye contact is eliminated; there are fewer nonverbal and contextual cues to pay attention to; the individual is afforded greater anonymity, invisibility, and control over content and timing of messages; errors in communication can be attributed to technological causes (Bonetti, Campbell, & Gilmore, 2010; Green et al., 2016; Shaw et al., 2015; Young & Lo, 2012).  These advantages are not valued only by socially anxious individuals, but social anxiety symptoms are positively correlated with their perceived benefits (Green et al., 2016; Young & Lo, 2012), and socially anxious individuals view online communication as broader, deeper, and more reciprocal than face-to-face communication as a result of them (Lee & Stapinski, 2012; Peter & Valkenburg, 2006).   The central attraction of online communication for these individuals may be the prospect of avoiding negative evaluation (Lee & Stapinski, 2012; Shepherd & Edelmann, 2005).  In fact, low perceived self-presentational skill, a feature of social anxiety that accompanies fear of negative evaluation (Schlenker & Leary, 1982), is a significant predictor of POSI (Caplan, 2003).  For individuals who doubt their abilities to make positive impressions face-to-face, then, formation and maintenance of profiles and social   12relationships can be achieved online as alternative, less threatening ways of satisfying their needs for self-presentation and belonging. Positive and negative effects of Internet use. At a glance, online communication would appear to be an asset for socially anxious individuals, particularly adolescents and young adults, who have been the focus of the majority of the research described above.  For adolescents in general, increased use of online communication has been shown to improve the later ability to initiate real-life friendships (Koutamanis, Vossen, Peter, & Valkenburg, 2013), a benefit that is likely to be more pronounced for those who are socially anxious and presumably have fewer offline friends.  This hypothesis has received some indirect support.  For instance, undergraduate students who interacted briefly with someone online prior to a face-to-face interaction demonstrated reduced anxiety in the face-to-face interaction and were less likely to want to avoid it (Knobeloch, 2013; Markovitzky, Anholt, & Lipsitz, 2012).  Additionally, socially anxious young adults may be more likely to engage in self-disclosure when online (Fernandez et al., 2012; Green et al., 2016), and there is some evidence that online self-disclosure promotes offline self-disclosure through a rehearsal effect (Schouten, Valkenburg, & Peter, 2007; Valkenburg, Sumter, & Peter, 2011).  This is important as self-disclosure is known to promote friendships (Buhrmester & Furman, 1987). Unfortunately, Internet and SNS use may not result in such positive outcomes for individuals with social anxiety.  A great deal of research has linked social anxiety to a construct known as problematic Internet use (PIU; Davis, 2001).  PIU is a multidimensional syndrome consisting of both behavioural and cognitive symptoms that lead to negative psychological, interpersonal, and occupational consequences (Beard & Wolf, 2001; Caplan, 2002).  The behavioural symptoms include excessive and compulsive Internet use (Lee &   13Stapinski, 2012), and the cognitive symptoms include beliefs that one is safer and more confident, comfortable, and socially competent in online than in face-to-face interactions and relationships (Caplan, 2003, 2005).  Arguably, then, PIU is closely tied to—if not inherent in—the POSI previously described. Although PIU was initially identified as a correlate of loneliness (e.g., Caplan, 2002), social anxiety was later found to account for this spurious relationship (Caplan, 2007).  Despite reporting greater confidence in face-to-face interactions with more frequent Internet use (Erwin, Turk, Heimberg, Fresco, & Hantula, 2004), socially anxious individuals become overdependent on the Internet for social interaction.  In one study, participants with higher levels of social anxiety exhibited more online self-disclosure, but less offline self-disclosure, than those with lower levels of social anxiety (Weidman et al., 2012), suggesting that socially anxious individuals may not demonstrate the rehearsal effect previously proposed.  In other words, socially anxious individuals’ online behaviours may not translate to parallel face-to-face social benefits.  Even if they do, there are likely limited opportunities to find out, as the compulsive nature of PIU exacerbates socially anxious individuals’ avoidance of face-to-face interactions, even with family members living in their own homes (Huan, Ang, & Chye, 2014; Lee & Stapinski, 2012; Weidman et al., 2012). Sadly, online social interactions may not be sufficient to satisfy the belonging needs of socially anxious people.  Among those who interact frequently online, those with higher social anxiety report lower quality of life and higher depressive symptoms (Weidman et al., 2012).  Indeed, although online social behaviour is not associated with well-being for individuals low in social anxiety, it is strongly associated with well-being for those high in social anxiety, with poorer outcomes for those with low offline social behaviour (Indian &   14Grieve, 2014; Koo, Woo, Yang, & Kwon, 2015).  Generally, it is more difficult to establish high-quality relationships online than it is in real life; for instance, frequent online communication does not always result in new friendships (Pierce, 2009).  Exacerbating this problem, adolescents and young adults with social anxiety are more likely to use the Internet to communicate with non-close others (i.e., strangers or acquaintances), whereas those who are more socially comfortable are more likely to use it to communicate—and make plans for valuable face-to-face contact—with existing close friends (Gross, Juvonen, & Gable, 2002). One particularly illustrative study is worth noting when considering socially anxious individuals’ Internet use.  For the purposes of their study, Erwin et al. (2004) posted a survey about social anxiety to the website of an anxiety clinic.  Among respondents to the survey, 92% met criteria for SAD.  As a whole, respondents scored significantly higher on measures of social anxiety symptomatology and impairment than did a sample of SAD patients receiving treatment in the clinic at the time.  Within the online sample, higher frequency of Internet use was associated with greater impairment.  The direction of this relationship is unclear, as those with more severe SAD symptomatology may simply prefer to seek comfort and connection online than to seek treatment face-to-face.  Either way, the PIU characteristic of socially anxious individuals appeared to exacerbate these respondents’ social impairment and isolation in addition to reinforcing their maladaptive behaviours and beliefs (e.g., that others are critical and rejecting).  This latter finding was replicated in a similar study with an entirely clinical sample, in which higher symptoms of social anxiety were associated with greater fear of negative evaluation in online contexts (Mazalin & Klein, 2008).  It appears, then, that socially anxious individuals’ efforts to reap psychological, social, and emotional benefits from the Internet are backfiring.  Thus, it is imperative to better understand the   15nature of their online behaviours and how these might contribute to their psychopathology and other negative outcomes. Passive social networking site use and social comparison. While it is clear that socially anxious individuals favour online over face-to-face social interaction, it is unclear what this online social interaction entails, or how much of their time online is actually spent engaging with others.  Although SNS are typically regarded as platforms for social engagement, a significant proportion of time spent on SNS may be used for non-interactive purposes.  Activity on Facebook, for instance, can be classified according to three main categories: interactive communication, content production, and passive content consumption (Shaw et al., 2015).  Whereas interactive communication (e.g., wall posts, messages, comments, and likes) is associated with stronger social connections and decreased loneliness, passive content consumption (e.g., browsing one’s News Feed or others’ profiles) is associated with weaker social connections, increased loneliness and depressive symptoms, and declines in affective well-being for people in general (Burke, Marlow, & Lento, 2010; Tandoc, Ferrucci, & Duffy, 2015; Verduyn et al., 2015).  With regard to social anxiety, symptoms have been positively associated both with anxiety about using Facebook’s interactive features (McCord et al., 2014) and with time spent passively consuming Facebook content (Shaw et al., 2015).  In Erwin et al.’s (2004) study, too, survey respondents who met criteria for SAD spent less than 20% of their online time interacting with others and over 50% of their time passively browsing the Internet.  Perhaps unsurprisingly, these passive online behaviours do not seem to engender social or emotional benefits for those who are socially impaired.  In contrast to the results of previously described studies involving initial online interaction (Knobeloch, 2013; Markovitzky et al., 2012), highly socially anxious   16individuals who merely viewed their interaction partner’s profile in advance of a face-to-face interaction exhibited greater physiological arousal upon meeting (Rauch, Strobel, Bella, Odachowski, & Bloom, 2014).  In a study of adolescents who had low-quality friendships—as would socially anxious adolescents—those who used the Internet for non-interactive purposes suffered increases in social anxiety and depression (Selfhout, Branje, Delsing, ter Bogt, & Meeus, 2009). Importantly, the aforementioned associations of passive Facebook activity with both depressive symptoms and decreased affective well-being were mediated by increased feelings of envy (Tandoc et al., 2015; Verduyn et al., 2015); this was true even when interactive Facebook activity, non-Facebook SNS use, and in-person social interactions were controlled for (Verduyn et al., 2015).  Feelings of envy presumably stem from social comparison engaged in while browsing others’ Facebook profiles, which has elsewhere been found to mediate the relationship between frequency of Facebook use and depressive symptoms (Lee, 2014; Steers, Wickham, & Acitelli, 2014).  In another study, exposure to upward social comparison was found to decrease state self-esteem and to mediate the inverse relationship between frequency of Facebook use and trait self-esteem (Vogel, Rose, Roberts, & Eckles, 2014).  There is evidence that SNS-specific social comparison is distinct from more general social comparison as a predictor of negative emotional outcomes (Feinstein et al., 2013).  In a longitudinal study, adolescents engaging in more social comparison through technological means including SNS experienced more depressive symptoms one year later after initial depressive symptoms were controlled (Nesi & Prinstein, 2015). Certain individuals may be particularly susceptible to the deleterious effects of social comparison, with a host of individual differences influencing their magnitude.  Individuals   17who are already depressed (Appel, Crusius, & Gerlach, 2015) and who have lower pre-existing levels of life satisfaction (de Vries & Kühne, 2015) are more negatively affected by social comparison, particularly upward social comparison, than are non-depressed or more satisfied others.  In an aforementioned study, the depressive effect of technologically-based social comparison was strongest for females and for those low in popularity (Nesi & Prinstein, 2015).  The tendency to compare oneself to others, known as social comparison orientation (Gibbons & Buunk, 1999), differs between individuals and is related to lower state self-esteem, poorer self-perceptions, and more negative affect balance when engaging in social comparison on Facebook (Vogel et al., 2015).  Social anxiety is moderately correlated with social comparison orientation (r = .31; Gibbons & Buunk, 1999), a finding that is unsurprising given the established social comparison tendencies and sensitivities of socially anxious people (Aderka et al., 2009; Antony et al., 2005). To summarize, individuals higher in social anxiety spend much of their time on Facebook passively consuming content, which may include others’ Facebook profiles (Shaw et al., 2015).  Passive content consumption is likely to be accompanied by upward social comparison, as both activities can be easily linked to profile browsing and predict similar negative emotional consequences (e.g., Verduyn et al., 2015).  Furthermore, individuals higher in social anxiety are known to engage in more frequent upward social comparisons than those who are less socially anxious and to suffer greater negative emotional consequences of these comparisons (Aderka et al., 2009; Antony et al., 2005).  Thus, it is plausible that such frequent and powerful upward social comparisons play a major role in the Internet and SNS use of socially anxious people and the greater negative emotional consequences that result.   18Social Rank Judgments on Facebook  More specifically, given that socially anxious individuals are highly sensitized to social hierarchies, perhaps their upward social comparisons on Facebook are characterized by judgments of others as higher in social rank.  In other words, if it is true that socially anxious individuals overutilize the social rank system, perhaps it is this overutilization that accounts for the negative emotional outcomes that result from their Internet and SNS use.  Theories that socially anxious individuals perceive others as dominant (Trower & Gilbert, 1989)—or higher in social rank—are compelling, but as it stands have only received support in the relatively artificial context of vignette studies (Aderka et al., 2013; Haker et al., 2014).  It is more useful to investigate the existence of this social rank judgment bias in more realistic contexts, particularly those in which impression formation naturally and frequently occurs.  Facebook is an excellent example of such a context.  Do socially anxious individuals demonstrate a social rank judgment bias on Facebook, or do special features of SNS influence their impressions of others in unique ways?  As previously noted, the Internet is characterized by a virtual absence of nonverbal and contextual cues, as well as apparently fewer indications of status (e.g., Bonetti et al., 2010; Shepherd & Edelmann, 2005).  Perhaps these differences make online social rank judgments a more challenging task.  Although people are motivated to ensure that others perceive them accurately and in a way that is consistent with their own self-perceptions (Gergen, 1968; Jones & Pittman, 1982; Schlenker, 1980), it is uncertain whether this is necessarily true on SNS, where creation of an online persona typically involves selection of the most favourable aspects of oneself (Brunskill, 2013; Chou & Edge, 2012).  However, research exploring actual impressions of others’ SNS profiles has shown that profiles   19generally reflect offline dispositions (Gosling, Augustine, Vazire, Holtzman, & Gaddis, 2011; Weidman & Levinson, 2015; Wu, Chang, & Yuan, 2015), even after controlling for the profile owner’s self-reported “ideal” traits (Back et al., 2010). If Facebook profiles reflect true dispositions, then perhaps the cues used by socially anxious individuals to form impressions of others’ social rank traits are just as easily accessed on Facebook as they are in other contexts.  Accordingly, the same social rank judgment bias that emerged in the vignette studies might also emerge when examining such impressions on Facebook.  Furthermore, even if Facebook users do curate their profiles to present only the most positive aspects of themselves, this self-enhancement would presumably only amplify this bias, especially if it is true that the bias exists only when judging others explicitly portrayed as dominant (Aderka et al., 2013; Haker et al., 2014).  Since socially anxious individuals’ social rank judgments have only so far been assessed using an artificial medium, however, assessing them on Facebook has the potential to yield entirely different conclusions.  For example, Facebook profiles are comprised of several sections (e.g., number of Facebook friends, photographs, groups, and “liked” pages), which collectively may provide ambiguous or contradictory information about the individuals presented within.  However, investigating these judgments on Facebook is still a valuable pursuit, as the findings are more directly applicable to real-life contexts. The Current Study  The current study attempts to answer two primary research questions.  First, do socially anxious individuals engage in a social rank judgment bias when forming impressions of others on Facebook?  Second, if they do, does this bias account for greater negative change in state self-esteem and affect after browsing others’ Facebook profiles?  Social   20anxiety is conceptualized as a continuous dispositional variable, such that social rank judgments and emotional changes are explored as a function of participants’ trait levels of social anxiety.  As young adults are the heaviest users of SNS including Facebook (Coyne, Padilla-Walker, & Howard, 2013), these research questions were explored in a sample of undergraduate students for both theoretical and practical reasons. Participants began the study by completing measures of state self-esteem and affect.  They then each browsed the Facebook profiles of 10 unknown peers (targets).  In keeping with previous research (e.g., Fernandez et al., 2012; Lueders, Hall, Pennington, & Knutson, 2014; Weidman & Levinson, 2015), the Facebook profiles were presented to participants in PDF format to allow for standardization of stimuli and greater identity protection of targets.  After browsing each profile, participants rated each target on traits related to both social rank and affiliation.  The total time taken to browse and rate all 10 target profiles was measured and recorded for each participant.  At the end of the experimental session, participants completed the measures of state self-esteem and affect for a second time in addition to a trait measure of social anxiety. Primary analyses. Based on previous research on overutilization of the social rank system, and given that Facebook either mirrors offline contexts or elicits self-enhancement strategies, participants higher in social anxiety were expected to rate targets higher on social rank-related traits after browsing their Facebook profiles.  Thus, the first primary hypothesis was that social anxiety would predict higher target social rank ratings.  Given that upward social comparison leads to greater decreases in self-evaluations and affect for socially anxious individuals (Antony et al., 2005; Mitchell & Schmidt, 2014; Hope et al., 1998), the second primary hypothesis was that social anxiety would predict greater decreases in state   21self-esteem and positive affect, and greater increases in negative affect, over the course of the study.  Finally, a mediation model was hypothesized whereby higher target social rank ratings would mediate the expected relationships between social anxiety and changes in state self-esteem, positive affect, and negative affect.  This mediation model was determined to be reasonable from a temporal perspective: Social anxiety is a preexisting trait that therefore precedes social rank judgments, which would themselves precede any emotional changes that resulted. Secondary analyses. In light of the inconsistent findings on socially anxious individuals’ utilization of the affiliation system, ratings of targets’ affiliation-related traits were explored as a function of participant social anxiety with no hypotheses specified a priori.  Finally, in line with Aderka et al.’s (2013) finding of an information-seeking bias in individuals with SAD, social anxiety was expected to predict less time taken to form and indicate impressions of targets overall.               22Method Participants  A total of 214 undergraduate participants, recruited from the Human Subject Pool (HSP) at the University of British Columbia, participated in the study.  The study advertisement posted to the HSP website specified two inclusion criteria, namely that participants must be between the ages of 17 and 25 and must use Facebook on a regular basis (daily or several times per week) to ensure familiarity with the medium1.  As compensation for their participation, participants received partial course credit toward their introductory psychology course.  The data for two participants were excluded from analyses, as one participant terminated participation halfway through the study and another mislabeled target profiles such that it was impossible to match trait ratings to the appropriate targets.  The remaining sample of 212 participants ranged in age from 17 to 25 years (M = 19.67, SD = 1.76).  Of the sample, 76.89% identified as female.  When asked about their cultural backgrounds, 32.07% of participants self-identified as Caucasian, 30.66% self-identified as Chinese, 16.98% self-identified with another Asian culture2, and 20.28% self-identified with a non-Caucasian and non-Asian culture.  All participants reported speaking English for a minimum of three years. Materials  Target Facebook profiles. A total of 90 Facebook profiles were obtained from a database of 124 profiles that were prepared for previous studies conducted by a separate                                                  1 As 98.58% of participants reported using Facebook at least four days per week, this inclusion criterion was determined to be met.  The three participants who reported using Facebook for fewer than four days per week were not omitted from analyses given the unlikely impact on analytic results. 2 Separating Chinese from other Asian cultures was recommended by a senior graduate student who noted important distinctions between these cultural groups, and made sense given the proportion of Chinese participants in this sample.   23research team.  A Facebook account was created for the aforementioned studies, through which research assistants accepted participants as Facebook friends; all participants had previously provided their consent to become Facebook friends with, and subsequently to have their profiles viewed and downloaded by, the research team.  Research assistants then converted the participants’ profiles into PDF files using screenshots of all visible sections under the “About” tab (see redacted protocol in Appendix A).  As Facebook users vary in their amount of activity and self-disclosure, the total length of the PDF files varied greatly between participants (1-29 pages), with individual sections varying in their presence and length on a given profile.  Participants’ names, their Facebook friends’ names, and their contact information were disguised so as to protect their identities as much as possible.  Each participant was asked to review and approve the PDF created from his or her profile before it was used in research.  For the purposes of the current study, these participants are referred to as targets. Measures State Self-Esteem Scale (SSES; Heatherton & Polivy, 1991), social self-esteem subscale. The SSES is a 20-item self-report questionnaire that measures a participant’s self-esteem at a given moment.  Participants indicate the extent to which each statement is true for them in that moment on a 5-point scale from 1 (not at all) to 5 (extremely).  The scale has demonstrated high internal consistency in multiple undergraduate samples (e.g., Heatherton & Polivy, 1991) and at least one sample of adolescents (Linton & Richard, 1996).  Principal axis factor analyses have revealed three intercorrelated but distinct subcomponents representing performance, social, and appearance self-esteem (Heatherton & Polivy, 1991); the same structure was confirmed in Linton and Richard’s (1996) sample.  Each   24subcomponent shows a different pattern of correlations with related constructs and is differentially sensitive to experimental manipulations and treatment strategies (Heatherton & Polivy, 1991; Linton & Richard, 1996).  In this study, only the 7-item social self-esteem subscale was used, as it is most pertinent to social anxiety (Heatherton & Polivy, 1991).  Three unrelated items (items 1, 4, and 7) were added to the subscale in an effort to conceal its purpose (e.g., “my experience at UBC has been positive so far”), but were not included in data analyses (see Appendix B).  All relevant items, indicated in boldface, were reverse-scored before the total score was computed.  In the current sample, the subscale demonstrated good (α = .87) and excellent (α = .90) internal consistency when administered at the beginning and end of the study, respectively. The International Positive and Negative Affect Schedule Short Form (I-PANAS-SF; Thompson, 2007), adapted. The I-PANAS-SF is a brief measure of positive and negative affect, comprised of 10 emotion word items from the longer Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988).  There are two subscales, with five items reflecting positive affect and five items reflecting negative affect.  These subscales have acceptable internal consistency and have been demonstrated to correlate well with the full PANAS (Thompson, 2007).  When completing the PANAS and its variants, participants indicate the extent to which they feel or have felt each emotion word listed on a 5-point scale, with scale anchors varying as a function of the time span of interest.  To measure momentary affect, participants are asked to what extent they currently feel each emotion from 1 (very slightly or not at all) to 5 (extremely).  Total scores are then computed for each subscale. The content of the I-PANAS-SF has been modified for this study.  The emotion   25words included in the standard I-PANAS-SF have been criticized for overrepresenting arousal (e.g., attentive; Larsen & Diener, 1992) and underrepresenting valence (e.g., cheerful).  Accordingly, the 10 items of the I-PANAS-SF were replaced with 10 items selected from the 60-item PANAS-Expanded Form (PANAS-X; Watson & Clark, 1994) that better represent the valence-type emotions that participants are likely to experience on a momentary basis (see Appendix C).  Items comprising the positive affect subscale are indicated in boldface.  When administered at the beginning of the study, both positive and negative affect subscales demonstrated acceptable internal consistency (α = .73 and .76, respectively); when administered at the end of the study, the internal consistency of each subscale was somewhat higher (α = .78 and .80, respectively). Explicit Domain-Specific Social Estimations scale (EDSSE; Gilboa-Schechtman et al., 2013), adapted.  Gilboa-Schechtman et al. (2013) developed this scale to measure participants’ explicit evaluations of their own social rank- and affiliation-related traits.  There are 20 items in total, with five reflecting high social rank (e.g., dominant), five reflecting low social rank (e.g., submissive), five reflecting high affiliation (e.g., friendly), and five reflecting low affiliation (e.g., hostile).  The researchers conducted analyses supporting a two-factor structure, with good reliability estimates (α = .81) obtained for both the social rank and affiliation subscales (Gilboa-Schectman et al., 2013).  More recently, E. Gilboa-Schechtman (personal communication, March 19, 2015) advised that additional studies have since revealed strong psychometric support for six social rank items and six affiliation items, in addition to the four example items given above.  Therefore, a 16-item scale was used in the current study, with eight items tapping social rank-related traits (indicated in boldface) and eight items tapping affiliation-related traits (see Appendix D).  Participants indicated the   26extent to which they agreed that each trait was descriptive of the target in question on a scale from 1 (disagree strongly) to 7 (agree strongly).  Eight items, identified with asterisks, were reverse-scored before total scores were computed for each subscale.  In order to verify the validity of a two-factor structure for the current sample, three exploratory factor analyses were conducted using principal axis factoring (i.e., one for the 1st, 5th, and 10th targets evaluated by each participant).  Scree plots and eigenvalues resulting from each initial factor extraction suggested the existence of three or four factors, with the first two factors accounting for the majority (45.85-51.98%) of the variance.  The data were then constrained to two, three, and four factors using a direct oblimin rotation.  For each target ordinal position evaluated, the emerging factor structure was cleanest and most parsimonious when constrained to two factors representing social rank and affiliation.  One item was omitted from each factor (italicized in Appendix D): “Cowardly” was excluded from the social rank factor as it consistently cross-loaded, and “distant” was excluded from the affiliation factor as it had consistently low factor loadings (≤ .40) and loaded on the social rank factor in one analysis.  Per best practice guidelines outlined by Costello and Osborne (2005), the remaining items were retained as they all loaded higher than .32 on their respective factors.  The resulting social rank and affiliation subscales demonstrated acceptable (α = .78-.85) and good (α = .82-.89) internal consistency, respectively, across all evaluated targets. Straightforward Social Interaction Anxiety Scale (S-SIAS). The Social Interaction Anxiety Scale (SIAS; Mattick & Clarke, 1998) is a 20-item self-report questionnaire that assesses anxiety in situations involving social interaction.  Participants indicate the extent to which each item statement is characteristic or true of them on a 5-point scale from 0 (not at   27all) to 4 (extremely).  The SIAS has demonstrated high internal consistency and test-retest reliability in clinical samples and samples of both community and undergraduate controls (e.g., Heimberg, Mueller, Holt, Hope, & Liebowitz, 1992; Mattick & Clarke, 1998).  It correlates well with other established measures of social anxiety (Mattick & Clarke, 1998), and in support of discriminant validity, does not correlate well with measures of trait anxiety or general distress (e.g., Osman, Gutierrez, Barrios, Kopper, & Chiros, 1998).  Although not used for this purpose in the current study, the SIAS can discriminate between individuals with SAD and healthy controls (Heimberg et al., 1992; Mattick & Clarke, 1998). Factor analyses have supported a single-factor structure such that the SIAS measures general social interaction anxiety (e.g., Mattick & Clarke, 1998).  However, its reverse-scored items load less strongly on this factor and demonstrate consistently weaker relationships with related measures (Rodebaugh, Woods, & Heimberg, 2007).  Rodebaugh et al. (2007) noted that the psychometric performance (e.g., treatment sensitivity) of the total score is improved when these items are removed.  Based on their recommendations, the 17-item Straightforward SIAS (S-SIAS) was used in this study (see Appendix E).  This scale demonstrated excellent internal consistency (α = .93) in the current sample. Procedure  In advance of data collection, all 90 target Facebook profiles were randomly divided into sets of 5 profiles.  Following this, the sets were randomly paired, and this random pairing procedure was conducted twice such that each set of 5 profiles appeared in two larger sets of 10 profiles.  Through this method, 18 different 10-profile sets were created, with each smaller 5-profile set appearing in two different 10-profile sets.  The PDF versions of the profiles comprising each 10-profile set were stored together in individual binders, with 18   28different binders in total.  Participants who completed the study during pilot testing (n = 26) were presented with either binder 1 or binder 2; the remaining participants (n = 186) were presented with binders in sequential order of their participation (e.g., if participant 140 was presented with binder 14, participant 141 was presented with binder 15).  The proportions of the total sample presented with each binder are presented in Table 1. Up to 10 participants completed the study at each experimental session.  Participants completed the study in the same room, but were seated at least two seats away from one another.  Before providing their consent to participate, participants were informed that they would be viewing the Facebook profiles of a number of unknown peers and would be asked to indicate their impression of each peer by rating them on a variety of traits.  They were also told that they would answer questions about themselves at the beginning and end of the study, meaning that the study was comprised of three segments altogether.  Each experimental session was a maximum of 90 minutes in length, and participants were told to take as much time as needed to feel confident in their impression of each peer before rating them on the trait measure provided.  Finally, participants were asked to raise their hands when they completed each segment of the study, and to notify an experimenter if they already knew one or more of the peers whose profiles were in their assigned binder.  In the rare event that this occurred, the experimenter told the participant to skip these profiles, as the trait ratings were intended to capture first impressions.  All measures were completed through SurveyMonkey using individual iPads provided to each participant.  After providing their consent to participate, participants began the study by completing the SSES and I-PANAS-SF, yielding Time 1 measures of state self-esteem, positive affect, and negative affect.  During this first segment of the study, an experimenter   29started recording the time using an online stopwatch.  After each participant completed the Time 1 state measures, an experimenter provided him or her with a binder of target profiles as outlined above, while another noted his or her start time (in minutes and seconds) for the impression formation segment.  Participants browsed each of the 10 profiles provided to them in turn, completing the EDSSE for each target before proceeding to the next one.  In order to verify that participants evaluated their assigned profiles in the expected order, participants were asked to enter a unique code for each target profile before completing the corresponding EDSSE.  When each participant completed the impression formation segment, an experimenter removed his or her binder while another noted his or her end time. For the final segment of the study, participants completed the SSES and I-PANAS-SF for Time 2, followed by the S-SIAS and a demographics questionnaire (see Appendix F).  These measures were completed in this order in order to get more accurate measurements of participants’ emotional states following the impression formation segment.  Before participants left the experimental session, an experimenter discussed the purpose of the study with them, answered their questions, and provided them with contact information and additional resources.          30Table 1. Proportions of the Total Sample Presented with Each Binder Binder Number Sample Size (Proportion) Number of Pages 1 22 (10.38%) 72 2 25 (11.79%) 84 3 11 (5.19%) 103 4 10 (4.72%) 101 5 11 (5.19%) 82 6 11 (5.19%) 131 7 11 (5.19%) 92 8 11 (5.19%) 88 9 10 (4.72%) 66 10 10 (4.72%) 60 11 10 (4.72%) 100 12 10 (4.72%) 116 13 10 (4.72%) 100 14 10 (4.72%) 114 15 10 (4.72%) 91 16 10 (4.72%) 100 17 10 (4.72%) 61 18 10 (4.72%) 77  Note: Each target Facebook profile appeared in two different binders. Since the minimum number of participants presented with a given binder was 10, a minimum of 20 participants browsed and rated each individual target Facebook profile.                  31Analyses Data preparation. After all data were collected, participants’ scores on each measure were reverse-scored where appropriate and total scale and subscale scores were computed as outlined above.  All recorded times were converted into seconds, and each participant’s start time was subtracted from his or her end time to compute the total number of seconds he or she spent browsing and rating all 10 target profiles.  Further, as profile length varied considerably and the occasional profile was skipped, the total number of seconds was divided by the total number of PDF pages browsed; in this way, the average number of seconds spent browsing each page was computed for each participant.  Prior to conducting analyses, all variables were either standardized or dummy-coded to give them equal weight and enhance the interpretability of results.  Time 2 state self-esteem, positive affect, and negative affect were standardized using the means and standard deviations of their Time 1 counterparts so that change in all three variables could be interpreted according to their respective baseline scales. Analytic approach. Multilevel models were used to investigate participant social anxiety as a predictor of target social rank and affiliation ratings.  Multilevel models account for the interdependence of Level 1 observations within Level 2 groups, and thereby allow for the analysis of data that violate the independent observations assumption of traditional regression analyses.  The current data fit two-level models wherein variables specific to individual targets (i.e., target social rank and affiliation ratings) exist at Level 1 and participant characteristics exist at Level 2.  As the primary predictor, social anxiety, was treated as a trait variable that does not vary within participants, slope was assumed not to vary in the current models.  As such, the data were conceptualized within random intercept   32multilevel models.  Given that each target profile was browsed by at least 20 participants (see Table 1), random effects attributable to individual profiles were accounted for in the models. All other planned analyses, for which both predictors and dependent variables exist at Level 2, were conducted using multiple regression.  For analyses exploring change in state self-esteem, positive affect, and negative affect, the appropriate Time 1 scores were entered as predictors of their corresponding Time 2 scores so that the dependent variables represented residualized change. Power analysis and assumptions. Previous research demonstrating that SAD patients rate dominant protagonists higher in social rank than do controls reported relatively large effect sizes (d =.72, Aderka et al., 2013; d = .70, Haker et al., 2014).  Given a Level 1 sample size of 10 targets, a Level 2 sample size of 35 is sufficient to ensure adequate power (> .80) to detect a medium or large effect (Scherbaum & Ferreter, 2008).  However, sample sizes closer to 100 are needed in order to accurately estimate variance in Level 2 variables (Scherbaum & Ferreter, 2008; Van der Leeden & Busing, 1994).  Furthermore, researchers who conducted a comprehensive meta-analysis concluded that the average effect size in social psychological research is r = .21 (Richard, Bond, & Stokes-Zoota, 2003), an effect size that requires at least 200 participants to detect with reasonable power (.80).  Accordingly, the size of the current sample was sufficient to detect effects of at least small-to-moderate size.  Estimations of statistical power for multilevel models typically assume balanced designs (i.e., the same number of Level 1 observations within each Level 2 group); when unbalanced, specified Level 1 sample sizes may represent averages (Liu, 2003).  As 94.8% of participants completed ratings for all 10 assigned targets, this assumption was determined to   33be met.  Other assumptions are implicated in multilevel modeling analyses (e.g., the independence of residual errors at each level), but research suggests that when sample sizes are sufficient, estimates of regression coefficients and tests of their significance are accurate even when these assumptions are violated (e.g., Van der Leeden & Busing, 1994). Missing data. The greatest number of responses missing from a single scale item was eight (3.77%), with the majority of scale items having no missing responses.  When computing composite total scores, missing data were handled by computing the mean of available responses, which was then multiplied by the total number of items in the scale or subscale.  This method was chosen in lieu of mean imputation in order to avoid artificially altering composite scores as much as possible.  In the rare event that a participant rated fewer than 10 targets, data for the unrated targets were retained as missing given the small amount and the ability of multilevel models to handle such missingness3.                                                            3 Multilevel models are well-equipped to handle data that are missing completely at random (Graham, Taylor, Olchowski, & Cumsille, 2006), as was the case for target ratings as determined by Little’s test (χ2 (10373) = 10472.63, p = .244).   34Results Descriptive Statistics  The means, standard deviations, and ranges for participant-level variables—including predictors, dependent variables, and potential covariates—are presented in Table 2.  The profile of social anxiety in the current sample is comparable to S-SIAS norms established by Rodebaugh et al. (2011) for a sample of university students (M = 23.50, SD = 13.53, range = 1-53).  The same descriptive statistics are presented in Table 3 for all target social rank and affiliation ratings on the EDSSE.  Participants with missing data were excluded separately from each computed statistic, although at least 99.06% of data were included in the computation of each.                  35Table 2. Descriptive Statistics for Participant-Level Variables  Variable  M (SD) Range         Potential Actual S-SIAS 21.54 (13.65) 0-68 0-60 SSES          Time 1          Time 2  23.47 (5.97) 24.38 (6.49) 7-35    7-35 7-35 I-PANAS-SF      PA          Time 1          Time 2      NA          Time 1          Time 2   15.86 (3.36) 15.04 (3.86)  10.04 (3.81) 10.15 (4.13) 5-25      7-25 6-25  5-25 5-24 Time per page 28.41 (8.93)  NA 10.49-68.13 Number of pages  89.24 (18.55)  NA 50-131  Note: S-SIAS = Straightforward Social Interaction Anxiety Scale; SSES = State Self-Esteem Scale, social self-esteem subscale; I-PANAS-SF = The International Positive and Negative Affect Schedule Short Form, adapted; PA = positive affect; NA = negative affect. All descriptive statistics were calculated using appropriate composite scores.               36Table 3. Descriptive Statistics for Target-Level Variables  Target Ordinal Position  Rating  M (SD)       Range         Potential               Actual 1st SR A 31.32 (6.27) 36.67 (5.40) 7-49 7-49 14-45 7-47 2nd SR A 31.12 (7.14) 35.34 (6.16)  9-46 13-49 3rd SR A 30.55 (7.07) 36.55 (5.77)  13-47 13-49 4th SR A 31.13 (7.35) 35.65 (6.31)  12-49 7-49 5th SR A 29.38 (7.06) 35.06 (5.71)  9-47 16-48 6th  SR A 30.04 (7.07) 35.52 (5.83)  12-49 13-49 7th SR A 28.50 (7.11) 36.05 (5.31)  10-48 20-49 8th  SR A 30.66 (6.17) 36.58 (6.23)  13-43 13-48 9th  SR A 30.94 (6.39) 35.64 (6.20)   13-46 15-48 10th  SR A 28.03 (6.24) 35.56 (5.73)  12-42 19-48  Note: SR = target social rank rating; A = target affiliation rating. All descriptive statistics were calculated using total scores for the subscales of the Explicit Domain-Specific Social Estimations scale, adapted.           37Preliminary Analyses  Bivariate correlations. The bivariate correlations between participant-level variables were computed and are presented in Table 4.  Age has been included.  Since age was negatively associated with both Time 1 and Time 2 positive affect on the I-PANAS-SF, it is controlled for in multiple regression analyses assessing change in this construct.  Given the strong negative association between number of pages and time per page, number of pages is controlled for in the secondary analyses involving the latter variable.  Of note, the lack of association between S-SIAS and time per page may provide preliminary evidence that social anxiety does not influence time spent browsing target profiles.               38Table 4. Bivariate Correlations Between Participant-Level Variables Variable 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 1. Age - -.05 .07 .07 -.16* -.15* -.05 -.08 .07 -.04 2. S-SIAS  - -.57*** -.56*** -.24*** -.30*** .42*** .40*** .02 -.02 3. SSES T1   - .89*** .19** .21** -.56*** -.49*** -.01 .05 4. SSES T2    - .19** .19** -.57*** -.57*** -.01 .06 5. PA T1     - .76*** -.21** -.21** -.12 .05 6. PA T2      - -.17* -.20** -.11 .05 7. NA T1       - .83*** -.07 .05 8. NA T2         - -.03 .06 9. Time   per page         - -.63*** 10. Number of pages          -  Note: S-SIAS = Straightforward Social Interaction Anxiety Scale; SSES = State Self-Esteem Scale, social self-esteem subscale; PA = positive affect subscale of The International Positive and Negative Affect Schedule Short Form, adapted (I-PANAS-SF); NA = negative affect subscale of the I-PANAS-SF; T1 = Time 1; T2 = Time 2.  *p < .05; **p < .01; ***p < .001.     39Between-group comparisons. Further analyses were conducted to explore potential gender- and culture-based differences in each participant-level variable.  Independent-samples t-tests indicated that female participants scored marginally higher on the S-SIAS than did male participants (t(210) = -1.76, p = .080, d = 0.30); relevant means and standard deviations are presented in Table 5.  A multivariate analysis of variance, with four levels, revealed small but significant differences between select cultural groups on the S-SIAS   (F(3, 208) = 2.72,   p = .045, partial η2 = .04) and the SSES at Time 1 (F(3, 208) = 3.84, p = .011, partial η2 = .05).  Relevant means and standard deviations are presented in Table 6, with significant pairwise differences denoted4.  As a result of these findings, both gender and cultural background will be controlled in all primary and secondary analyses in order to account for differences across groups on the predictor of interest (i.e., social anxiety).                                                             4 Descriptive statistics for other gender- and culture-based differences in participant-level variables are not reported as no other differences were significant (ps > .10).   40Table 5. Difference in Social Anxiety Between Genders Gender M (SD) Range Male 18.55 (11.23) 0-42 1-60 Female 22.43 (14.21)  Note: Social anxiety was measured using the Straightforward Social Interaction Anxiety Scale. The potential range for the scale is 0-68. The difference in the displayed group means approached significance, p = .080.                      41Table 6. Differences in Participant-Level Variables Between Cultural Groups   Variable  Cultural Group  M (SD) Range         Potential                 Actual S-SIAS Caucasian* Chinese Other Asian Other* 17.97 (12.19) 22.70 (13.28) 22.01 (13.14) 25.02 (15.84) 0-68  0-57 4-60 6-56 2-59.5 SSES T1 Caucasian* Chinese* Other Asian Other 25.33 (5.34) 22.02 (5.72) 22.66 (6.27) 23.40 (6.44)  7-35 12-35 9-35 8-33 7-34  Note: S-SIAS = Straightforward Social Interaction Anxiety Scale; SSES T1 = State Self-Esteem Scale, social self-esteem subscale, Time 1; Other = participants of non-Caucasian, non-Asian cultural background.  *means differ at p < .05.                42Primary Analyses  For each primary hypothesis tested, variables identified as covariates in the preliminary analyses were included as predictors5.  This meant that gender and cultural background were always included to control for differences between groups on the S-SIAS.  In addition, age was included in the regression model predicting change in positive affect on the I-PANAS-SF due to the correlations between these variables at both Time 1 and Time 2.  Even though number of pages was found to correlate with only one participant-level variable (i.e., time per page), it was included as a predictor in all primary analyses for two reasons: The length of an individual profile might impact ratings of certain traits and, although the length of all 10 profiles did not correlate with state self-esteem, positive affect, or negative affect at Time 1 or Time 2, it might impact changes between time points in any of these variables (e.g., due to fatigue).  For similar reasons, the ordinal position of a target variable was included as a predictor of target social rank ratings.  As four primary hypotheses were explored, the criterion of significance was reduced a priori to α = .0136.  Social rank ratings. To examine the first hypothesis that participant social anxiety would positively predict target social rank ratings, a multilevel model was created using the lme4 package in R (Bates, Maechler, Bolker, & Walker, 2015).  EDSSE social rank scores for each target were entered as the dependent variable; S-SIAS, gender, cultural background, number of pages, and target ordinal position were entered as predictors.  The main effect of S-SIAS was nonsignificant (β = 0.04, 98.7% CI [-0.03, 0.11], t(204) = 1.32, p = .190)7.  As                                                  5 Cultural background was always entered as a factor such that the output yielded comparisons between Caucasian participants and those of other cultural groups. 6 The Sidak-Bonferroni correction is computed as follows: αA = 1 – (1 – αB)1/C , where αA = the corrected alpha value, αB = the original alpha value, and C = the number of comparisons. 7 Estimates of variance accounted for, by individual predictors or by the overall model, could not be computed for multilevel modeling analyses.   43for the included covariates, the main effect of gender was significant (β = 0.25, 98.7% CI [0.09, 0.40], t(201) = 3.33, p = .001) such that female participants rated targets higher on social rank-related traits than did male participants.  No other covariate effects were significant (p ≥ .038).  The results of this multilevel modeling analysis are presented in  Table 7.                     44Table 7. Multilevel Model Predicting Target Social Rank Ratings Predictor β (SE) t (df)           p           R2 S-SIAS Gender Chinesea Other Asiana Othera Number of pages Target ordinal position 0.04 (0.03) 0.25 (0.07) 0.06 (0.08) 0.01 (0.09) -0.07 (0.09) 0.00 (0.04) -0.02 (0.01) 1.32 (204) 3.33 (201) 0.75 (206) 0.09 (202) -0.76 (205) 0.11 (336) -2.08 (1105)  .190 .001** .454 .890 .448 .910 .038 NA  Note: S-SIAS = Straightforward Social Interaction Anxiety Scale.   aChinese = comparison between Caucasian and Chinese participants; Other Asian = comparison between Caucasian and non-Chinese Asian participants; Other = comparison between Caucasian and non-Caucasian, non-Asian participants.  **p < .013.                             45 Change in state self-esteem. The second hypothesis was that social anxiety would predict greater decreases in state self-esteem between Time 1 and Time 2.  To test this, SSES Time 2 was regressed onto SSES Time 1, S-SIAS, gender, cultural background, and number of pages.  The overall model was significant (adjusted R2 = .79, F(7, 204) = 116.40,  p < .001).  Unsurprisingly, the main effect of SSES Time 1 was significant such that Time 1 positively predicted Time 2 scores (β = 0.91, 98.7% CI [0.82, 1.00], t(204) = 21.55,  p < .001).  Importantly, the main effect of S-SIAS was also significant (β = -0.11, 98.7% CI [-0.20, -0.02], t(204) = -2.66, p = .008) such that greater social anxiety predicted greater decreases in state self-esteem from Time 1 to Time 2, as predicted.  None of the covariate main effects were significant (p ≥ .019).  The results of this multiple regression analysis are presented in Table 8.  The effect of S-SIAS on change in SSES is illustrated in Figure 1.               46Table 8. Multiple Regression Model Predicting State Self-Esteem at Time 2 Predictor β (SE) t (df) p R2 SSES Time 1 S-SIAS Gender Chinesea Other Asiana Othera Number of pages 0.91 (0.04) -0.11 (0.04) 0.10 (0.08) 0.11 (0.09) 0.16 (0.10) 0.24 (0.10) 0.01 (0.03) 21.55 (204) -2.66 (204) 1.18 (204) 1.18 (204) 1.57 (204) 2.37 (204) 0.26 (204)  .000*** .008** .239 .240 .119 .019 .799 .79***  Note: SSES = State Self-Esteem Scale, social self-esteem subscale; S-SIAS = Straightforward Social Interaction Anxiety Scale.  aChinese = comparison between Caucasian and Chinese participants; Other Asian = comparison between Caucasian and non-Chinese Asian participants; Other = comparison between Caucasian and non-Caucasian, non-Asian participants.  **p < .013; ***p < .001.                  47Figure 1. Effect of Social Anxiety on Change in State Self-Esteem   Note: SSES = State Self-Esteem Scale, social self-esteem subscale; S-SIAS = Straightforward Social Interaction Anxiety Scale; Low = 1 SD below the mean score on the S-SIAS (7.89); High = 1 SD above the mean score on the S-SIAS (35.19); Change in SSES = difference between mean SSES Time 1 score (23.47) and computed SSES Time 2 score. SSES Time 2 scores were computed using these S-SIAS and SSES Time 1 values and their corresponding beta weights (e.g., for low S-SIAS, 0.91(23.47) – 0.11(7.89) = 20.49).           -10-8-6-4-20Low HighChange in SSESS-SIAS  48Change in positive affect. Similar analyses were conducted to test the third hypothesis that social anxiety would predict greater decreases in positive affect (PA).  I-PANAS-SF PA Time 2 was regressed onto I-PANAS-SF PA Time 1, S-SIAS, gender, cultural background, number of pages, and age.  The overall model explained significant variance in PA Time 2 scores (adjusted R2 = .59, F(8, 203) = 38.18, p < .001).  As anticipated, the main effect of PA Time 1 was significant (β = 0.83, 98.7% CI [0.72, 0.94], t(203) = 15.53, p < .001) such that Time 1 scores positively predicted Time 2 scores.  Notably, the main effect of S-SIAS was also significant (β = -0.14, 98.7% CI [-0.25, -0.03], t(203) = -2.53, p = .012) such that greater social anxiety predicted greater decreases in positive affect from Time 1 to Time 2, as predicted.  No covariate effects were significant   (p ≥ .049).  The results of this multiple regression analysis are presented in Table 9; the effect of S-SIAS on change in PA is illustrated in Figure 2.              49Table 9. Multiple Regression Model Predicting Positive Affect at Time 2 Predictor β (SE) t (df) p R2 PA Time 1 S-SIAS Gender Chinesea Other Asiana Othera Number of pages Age 0.83 (0.05) -0.14 (0.05) -0.25 (0.12) -0.03 (0.13) 0.09 (0.15) 0.04 (0.15) 0.01 (0.05) -0.05 (0.05) 15.53 (203) -2.53 (203) -1.98 (203) -0.19 (203) 0.58 (203) 0.28 (203) 0.24 (203) -0.86 (203)  .000*** .012** .049 .848 .563 .784 .809 .391 .59***  Note: PA = positive affect subscale of The International Positive and Negative Affect Schedule Short Form, adapted; S-SIAS = Straightforward Social Interaction Anxiety Scale.  aChinese = comparison between Caucasian and Chinese participants; Other Asian = comparison between Caucasian and non-Chinese Asian participants; Other = comparison between Caucasian and non-Caucasian, non-Asian participants.  **p < .013; ***p < .001.                50Figure 2. Effect of Social Anxiety on Change in Positive Affect   Note: PA = positive affect subscale of The International Positive and Negative Affect Schedule Short Form, adapted; S-SIAS = Straightforward Social Interaction Anxiety Scale; Low = 1 SD below the mean score on the S-SIAS (7.89); High = 1 SD above the mean score on the S-SIAS (35.19); Change in PA = difference between mean PA Time 1 score (15.86) and computed PA Time 2 score. PA Time 2 scores were computed using these S-SIAS and PA Time 1 values and their corresponding beta weights (e.g., for high S-SIAS, 0.83(15.86) – 0.14(35.19) = 8.24).            -10-8-6-4-20Low HighChange in PAS-SIAS  51Change in negative affect. To test whether social anxiety predicted greater increases in negative affect, I-PANAS-SF NA Time 2 was regressed onto I-PANAS-SF NA Time 1,  S-SIAS, gender, cultural background, and number of pages.  The overall model was significant (adjusted R2 = .70, F(7, 204) = 71.71, p < .001).  As expected, the main effect of NA Time 1 scores was significant (β = 0.88, 98.7% CI [0.77, 0.99], t(204) = 19.26, p < .001) such that they positively predicted Time 2 scores.  The main effect of S-SIAS was nonsignificant (β  = 0.07, 98.7% CI [-0.04, 0.18], t(204) = 1.54, p = .126).  Unexpectedly, cultural background emerged as a significant predictor whereby Caucasian participants experienced significantly greater increases in negative affect than did non-Chinese Asian participants (β = 0.35, 98.7% CI [0.08, 0.62], t(204) = 2.86, p = .005).  No other covariate effects were significant (p ≥ .357).  The results of this multiple regression analysis are presented in Table 10.              52Table 10. Multiple Regression Model Predicting Negative Affect at Time 2 Predictor β (SE) t (df) p R2 NA Time 1 S-SIAS Gender Chinesea Other Asiana Othera Number of pages 0.88 (0.05) 0.07 (0.05) -0.09 (0.10) -0.01 (0.11) -0.35 (0.12) 0.10 (0.12) 0.02 (0.04) 19.26 (204) 1.54 (204) -0.92 (204) -0.07 (204) -2.86 (204) 0.80 (204) 0.57 (204)  .000*** .126 .357 .941 .005** .427 .567 .70***  Note: NA = negative affect subscale of The International Positive and Negative Affect Schedule Short Form, adapted; S-SIAS = Straightforward Social Interaction Anxiety Scale.  aChinese = comparison between Caucasian and Chinese participants; Other Asian = comparison between Caucasian and non-Chinese Asian participants; Other = comparison between Caucasian and non-Caucasian, non-Asian participants.  **p < .013; ***p < .001.               53 Mediation. Given that social anxiety did not significantly predict the proposed mediator of target social rank ratings, mediation analyses were not pursued. Secondary Analyses The same approach used to test the primary hypotheses was repeated for the secondary hypotheses.  As before, gender, cultural background, and number of pages were always included as predictors, and target ordinal position was added as a predictor of target affiliation ratings.  As these secondary analyses were exploratory in nature, a standard alpha value (α = .05) was used. Affiliation ratings. The first secondary analysis set out to explore whether target affiliation ratings differed as a function of participant social anxiety.  In the multilevel model created, EDSSE affiliation scores for each target were entered as the dependent variable and S-SIAS, gender, cultural background, number of pages, and target ordinal position were entered as predictors.  The main effect of S-SIAS was nonsignificant (β = 0.02, 95% CI  [-0.06, 0.10], t(205) = 0.39, p = .697).  As with social rank ratings, the main effect of gender was significant (β = 0.20, 95% CI [0.00, 0.40], t(203) = 2.00, p = .047) such that female participants rated targets higher on affiliation-related traits than did male participants.  No other covariate effects were significant (p ≥ .052).  The results of this multilevel modeling analysis are presented in Table 11.        54Table 11. Multilevel Model Predicting Target Affiliation Ratings Predictor β (SE) t (df)        p      R2 S-SIAS Gender Chinesea Other Asiana Othera Number of pages Target ordinal position 0.02 (0.04) 0.20 (0.10) -0.17 (0.11) -0.13 (0.12) -0.24 (0.12) 0.02 (0.05) -0.02 (0.01) 0.39 (205) 2.00 (203) -1.59 (207) -1.07 (203) -1.96 (206) 0.31 (295) -1.66 (728)  .697 .047* .113 .288 .052 .757 .098 NA  Note: S-SIAS = Straightforward Social Interaction Anxiety Scale.   aChinese = comparison between Caucasian and Chinese participants; Other Asian = comparison between Caucasian and non-Chinese Asian participants; Other = comparison between Caucasian and non-Caucasian, non-Asian participants.  *p < .05.                55 Time per page. The second and final secondary analysis aimed to determine whether social anxiety predicted time taken to browse target profiles, with the hypothesis that social anxiety would negatively predict seconds spent browsing each page.  Time per page was regressed onto S-SIAS, gender, cultural background, and number of pages.  The overall model was significant (adjusted R2 = .40, F(6, 205) = 24.26, p < .001), but the main effect of S-SIAS was not (β = 0.00, 95% CI [-0.12, 0.12], t(205) = 0.02, p = .981).  However, the main effect of number of pages was significant (β = -0.63, 95% CI [-0.73, -0.53], t(205) =  -11.61, p < .001), suggesting that participants whose assigned binders contained more pages took less time to review each page.  No other covariate effects were significant (p ≥ .145).  The results of this regression analysis are presented in Table 12.                56Table 12. Multiple Regression Model Predicting Time Per Page Predictor β (SE) t (df)  p R2 S-SIAS Gender Chinesea Other Asiana Othera Number of pages 0.00 (0.06) 0.00 (0.13) -0.18 (0.14) -0.24 (0.16) 0.20 (0.16) -0.63 (0.05) 0.02 (205) 0.02 (205) -1.34 (205) -1.46 (205) 1.24 (205) -11.61 (205)  .981 .986 .183 .145 .215 .000*** .40***  Note: S-SIAS = Straightforward Social Interaction Anxiety Scale.   aChinese = comparison between Caucasian and Chinese participants; Other Asian = comparison between Caucasian and non-Chinese Asian participants; Other = comparison between Caucasian and non-Caucasian, non-Asian participants.  ***p < .001.                               57Discussion  The current study explored whether preliminary evidence of a social rank judgment bias in socially anxious individuals would replicate in the context of SNS, specifically Facebook.  Participants browsed the Facebook profiles of 10 random, unknown peers and rated each of them on traits related to social rank.  As participant social anxiety did not predict target social rank ratings, the social rank judgment bias did not emerge in the current sample.  However, as predicted, more socially anxious participants experienced greater decreases in state self-esteem and positive affect as a result of rating peers’ Facebook profiles, a finding that connects research on the detrimental effects of both social judgments and SNS use for this population.  These and other results, in addition to other potential mechanisms of emotional change, are discussed in detail below.  The results of individual analyses are discussed in concert with their associated limitations and future directions.  The section ends with a discussion of broader next steps followed by general conclusions. Social Rank Judgment Bias  The first objective of this study was to examine the relationship between social anxiety and judgments of others’ social rank on Facebook.  Contrary to expectations in light of recent research, participants’ trait social anxiety was not predictive of the ratings they assigned to targets on social rank-related traits.  This means that, when forming impressions of unknown peers on Facebook, socially anxious young adults do not exhibit a social rank judgment bias.   There are a number of methodological differences between earlier studies demonstrating such a bias (Aderka et al., 2013; Haker et al., 2014) and the current one, which may represent limitations and may at least partially account for the lack of effect   58herein.  To begin with, the media through which participants formed impressions of targets differed in many ways.  Facebook profiles—even in PDF format—vary in length and contain more varied types of information (e.g., number of friends and photographs) than do vignettes.  Furthermore, said information is more ambiguous and open to profile-browsers’ own interpretations.  This is especially true as the vignettes used in the previous studies deliberately portrayed protagonists as either dominant or submissive (Aderka et al., 2013; Haker et al., 2014).  In fact, when protagonists were displayed as neither dominant nor submissive (i.e., neutral), no social rank judgment bias emerged (Haker et al., 2014).  The more subjective information contained within Facebook profiles might be likened to these neutral descriptions.  In future studies, target profile PDFs might be manipulated to include more explicit indications of social status—particularly high social status—and examine whether these profiles are perceived differently as a function of participant social anxiety.  Alternatively, if SNS profiles already reflect offline personality traits (e.g., Gosling et al., 2011), profile PDFs may not have to be manipulated at all.  Instead, profiles of targets scoring high on self-reported social rank-related traits could be preferentially selected.  Equally possible, as noted in the introduction, is that Facebook users engage in self-enhancement, intentionally presenting the best aspects of themselves in their profiles (e.g., Brunskill, 2013).  Such self-enhancement would mean that a majority of Facebook users are motivated to present themselves as high in social status.  If they succeed, then, it is possible that many of the target profiles used in the current study did in fact contain high social rank cues.  In that case, reasons for the absence of a social rank judgment bias in this study would lie elsewhere.   59Other differences between this and previous studies may account for this absence of effect.  Whereas participants in the previous studies assessed social rank using two traits each (e.g., assertiveness and self-confidence; Aderka et al., 2013), composite scores in the current sample were comprised of seven traits.  Although this may have had an impact on results, the smaller number of items used in the previous studies is a relative weakness acknowledged even by the other research team (i.e., ratings may have been insufficiently comprehensive; Aderka et al., 2013).  A more crucial difference is the use of clinical samples by the other research team, as it may be that the social rank judgment bias emerges only above a certain threshold of social anxiety symptoms.  Perhaps an effect of social anxiety would emerge in future studies if participants above and below a clinical cutoff on the S-SIAS were compared.  Finally, perhaps a social rank judgment bias does exist in the current sample but, given the greater complexity and more salient social nature of profile browsing as compared to vignette reading, is characterized or clouded by other social cognitive processes.  This idea is elaborated upon in the discussion about potential mechanisms to follow. Emotional Consequences of Profile Browsing The second objective of this study was to investigate the changes in state self-esteem, positive affect, and negative affect that participants experienced after browsing target profiles, and how these changes were influenced by their social anxiety. The hypothesis was threefold, as social anxiety was expected to predict more pronounced decreases in state self-esteem, decreases in positive affect, and increases in negative affect.  The first part of this hypothesis was supported, as participants higher in social anxiety experienced greater decreases in state self-esteem after rating target Facebook profiles.  This finding is in line with previous research, which has demonstrated that young adults higher in social anxiety or   60social comparison orientation form more negative self-appraisals after engaging in brief social comparison on Facebook (Vogel et al., 2014, 2015) or after reading profiles of high-achieving students (Mitchell & Schmidt, 2014).   The second and third predictions were confirmed and disconfirmed, respectively: Participants higher in social anxiety experienced greater decreases in positive affect, but not greater increases in negative affect, after rating target profiles.  Although this distinction was not anticipated, it does not necessarily represent a deviation from previous findings, as earlier studies in this area did not operationalize affect in the same way.  So far, affect has been measured in terms of overall affect balance (Verduyn et al., 2015; Vogel et al., 2015) or in terms of symptoms of depression, anxiety, or loneliness (Aderka et al., 2009; Antony et al., 2005; Burke et al., 2010; Tandoc et al., 2015; Weidman et al., 2012).  This difference in operationalization may represent a limitation of the current study; alternatively, the study may simply replicate previous results in a more precise manner.  For instance, a decrease in affect balance may very well result from a decrease in positive affect in combination with unchanged negative affect.  Providing further support for this notion, social anxiety has been demonstrated to have significant and distinct relationships with both positive affect and negative affect, a characteristic that sets it apart from other anxiety disorders (Brown, Chorpita, & Barlow, 1998).  Specifically, social anxiety is inversely related to positive affect such that individuals higher in social anxiety experience less frequent, lower-intensity, and less long-lasting positive emotions (r = -.36; Kashdan, Weeks, & Savostyanova, 2011). Collectively, these findings tie together research demonstrating the more pronounced negative emotional impact of social judgments (e.g., Antony et al., 2005) and of Internet or SNS use (e.g., Weidman et al., 2012) for socially anxious individuals.  Although the exact   61nature of participants’ social judgments is uncertain, these results provide preliminary evidence that profile browsing—which may predominate the passive activity preferred by socially anxious individuals on Facebook—is particularly detrimental to these individuals’ emotional well-being, at least for young adults.  This study, therefore, sheds light on the online behaviours that may characterize the problematic Internet use of socially anxious individuals.  Given their emotional impact, these behaviours and their associated cognitions are important to understand for both conceptual and clinical reasons.  Individuals with social anxiety are drawn to SNS in pursuit of comfort, connection, and belonging that they struggle to find in face-to-face interactions; despite these good intentions, their behavioural and cognitive tendencies appear to result in more harm than benefit.  In light of this, mental health professionals are urged to explore the SNS-related behaviours of their socially anxious clients, the cognitive biases that accompany these behaviours, and the resulting impact on their clients’ self-feelings and emotions.  For example, psychologists practicing cognitive-behavioural therapy might encourage their clients with SAD to consider their SNS experiences when completing thought records, in order to challenge cognitive distortions specific to these experiences in the therapy setting.   Potential Mechanisms Underlying Emotional Consequences Assuming that socially anxious individuals both perceive others to be higher in social rank and suffer greater negative emotional consequences from these comparisons, a model was proposed in which higher target social rank ratings would mediate any relationships found between social anxiety and negative emotional change.  As social anxiety did not predict target social rank ratings, however, a social rank judgment bias cannot be said to underlie the greater decreases in state self-esteem and positive affect observed in this study.    62Given that participants rated real peers rather than fictional protagonists, the comparative aspect of the study may have been more salient than it has been in previous vignette studies (i.e., Aderka et al., 2013; Haker et al., 2014), making participants more likely to reflect on their own traits and social standings relative to those of the targets whose profiles they browsed.  In other words, although participants higher in social anxiety did not rate targets higher in social rank-related traits, they may have perceived themselves to be lower on these same traits.  Thus, perhaps the social rank judgment bias as measured in the current study did not mediate the observed relationship between social anxiety and negative emotional change because this relationship is better accounted for by self-judgments than by judgments of others.  This alternative mechanism is plausible in light of extant research.  As previously summarized, social anxiety is negatively related to self-perceptions of social rank (Aderka et al., 2009; Gilboa-Schechtman et al., 2013; Weisman et al., 2011) and positively related to self-perceptions of inferiority (e.g., Hope et al., 1998; Leary & Kowalski, 1997).  Furthermore, there is preliminary evidence that these perceptions mediate the relationship between social anxiety and depressive symptoms (Aderka et al., 2009).  Just as these negative self-perceptions are activated or strengthened when in the presence of others (Antony et al., 2005), perhaps this also occurs when browsing others’ SNS profiles. Alternatively, or in addition, participants’ emotions may have been impacted by judgments about other target characteristics.  For instance, judgments about individual targets’ physical attractiveness may have influenced participants’ feelings of envy (e.g., DelPriore, Hill, & Buss, 2012), which themselves are known to decrease affective well-being (Tandoc et al., 2015; Verduyn et al., 2015).  Judgments of targets may also have been moderated by their degree of similarity to the participant.  People prefer to compare   63themselves to others who are similar in highly salient attributes such as age, sex, and race (Miller, 1982), and more similarity intensifies the felt impact of inferiority in the attribute being compared (Kelley, 1973).  Thus, the emotional impact of browsing or rating target profiles may have been stronger for targets of the same gender and cultural background as the participant.  In future studies, the moderating effects of participant-target match in these and other variables should be explored. Affiliation Judgment Bias A secondary purpose of this study was to clarify contradictory findings regarding socially anxious individuals’ judgments of others’ affiliation-related traits.  Although previous studies have found social anxiety to predict both higher (Aderka et al., 2013) and lower (Haker et al., 2014) ratings of these traits, in the current study social anxiety did not predict these ratings in either direction.  Although this analysis was purely exploratory, these disparate patterns of results warrant conjecture. As discussed in light of the social rank-related findings, methodological differences may be partially responsible.  First, the stimuli upon which participants based their trait judgments were more complex in the current study.  Whereas the information provided in the vignettes was more concise and direct, in one case intentionally presenting protagonists along extremes of the affiliation dimension (i.e., friendly or unfriendly; Aderka et al., 2013), the information contained in the Facebook profiles was more varied and nuanced, likely making judgments more complicated.  Two sources of information within a given profile might present different pictures about the target’s affiliativeness: For example, he or she might have a high number of Facebook friends but no other people pictured in his or her photographs.  Second, the previous studies were conducted using clinical samples, and it may   64be that social anxiety only exerts an effect on affiliation judgments above a clinical threshold. Alternatively, perhaps neither Aderka et al.’s (2013) nor Haker et al.’s (2014) effect of social anxiety on affiliation judgments will replicate reliably.  Their findings might be so inconsistent simply because socially anxious individuals do not share a characteristic bias in their affiliation-related social perceptions.  Overutilization of the social rank system does not necessarily translate to a corresponding underutilization of the affiliation system.  Until further research is conducted on this topic, the existence of an affiliation judgment bias in socially anxious individuals cannot be concluded. Information-Seeking Bias  As an additional exploratory analysis, this study attempted to replicate preliminary evidence of an information-seeking bias whereby socially anxious individuals seek less information before indicating their impressions of others, particularly on social rank-related traits (Aderka et al., 2013).  Information-seeking was operationalized as the amount of time participants spent browsing each page of the target profiles assigned to them.  As social anxiety did not predict less time spent browsing each page, further evidence in support of an information-seeking bias was not found. As before, there are important qualifications to the different result found in this study.  First, information-seeking was operationalized differently than it was in Aderka et al.’s (2013) study, in which participants were given the option to receive up to eight additional pieces of information before rating protagonists on the traits in question.  It is possible that such differences in information-seeking do not translate to differences in time spent reading or browsing.  Alternatively, perhaps the current measure of information-seeking was   65insufficiently sensitive to actual differences in browsing time.  A limitation of this study is that participants were timed for the entire impression formation segment, during which they completed the EDSSE for each target in addition to browsing their profiles.  Hypothetically, if more socially anxious participants took less time to browse Facebook profiles but more time to complete trait measures, this would have cancelled out any information-seeking effect.   As participants in Aderka et al.’s (2013) study were told which traits they would be asked to rate before reading vignettes, perhaps such priming is necessary to activate the information-seeking bias.  However, if this were the case, the bias would presumably have been primed for the 2nd to 10th targets after participants completed the EDSSE for the 1st target.  As only one study has so far found evidence in support of it, it remains possible that the information-seeking bias does not exist or that it does only in clinical samples.  Further research is needed to reach a more definitive conclusion. Covariate Effects Interestingly, although only included in the models to control for differences in social anxiety, gender emerged as a significant predictor of both social rank and affiliation ratings.  Specifically, female participants rated targets higher in both social rank- and affiliation-related traits than did male participants, regardless of other individual differences including social anxiety.  Although gender-related hypotheses were not formulated at the outset of the study, the former finding is somewhat unexpected, as males are traditionally thought to be more concerned with social dominance (e.g., Kiefer & Ryan, 2008).  Perhaps this is truer of social rank-related behaviours than perceptions, though; females might perceive others to be higher in social rank in order to promote interpersonal harmony, in line with their own   66stereotyped gender roles (Rosenberg & Simmons, 1975).  In a similar vein, previous research has found women to engage in more affiliative behaviour and to rate themselves higher on affiliation-related traits than men (Costa, Terracciano, & McCrae, 2001; Moskowitz, Suh, & Desaulniers, 1994).  Although gender differences in perceptions of others’ affiliation-related traits have not previously been explored, it may be that perceptions of others as more affiliative promote more affiliative behaviours.  More research is warranted to determine if these gender differences are consistently found. Although social anxiety did not predict change in negative affect, cultural background emerged as a significant predictor of this variable.  Only two cultural groups differed: Caucasian participants experienced significantly greater increases in negative affect compared only to non-Chinese Asian participants.  The reasons for this culture-based difference are unclear.  One possibility is that the samples of previous studies were primarily composed of Caucasian participants, and individuals from other cultural backgrounds do not all respond to social comparison or passive Facebook use in the same way. The only significant predictor of time per page was the total number of PDF pages browsed by each participant.  This means that the more PDF pages contained in the binder assigned to a participant, the less time he or she spent browsing each individual page.  At the beginning of the study, participants were told that they had a total of 90 minutes to complete all three segments; although encouraged to take as much time as needed to feel confident in their ratings, it is possible that participants remained cognizant of the time limit—or of a prior engagement scheduled for after the study—and adjusted their pace according to the number of pages assigned to them.  Alternatively, as up to 10 participants completed the study in the same room, some participants may have been aware of others’ progress through   67their respective binders and made efforts to match them.  In the future, allowing participants to complete the study alone and with more time scheduled will either eliminate or solidify the effect of number of pages on participants’ browsing speed.  If the effect is eliminated, this may allow for even greater variation in browsing speed and potentially for other effects to emerge. Strengths and Future Directions  In addition to the limitations identified above, there are many strengths to the current study.  For instance, a very large sample size was used, ensuring sufficient power to detect at least small-to-moderate effect sizes.  Similarly, the use of multilevel models was of benefit, allowing for the interdependence of target ratings by the same participant and making it possible to account for both participant- and target-level sources of variance.  Moving forward, further advantage can be taken of these models by including target characteristics (e.g., gender, cultural background) that might account for significant variance in social rank and other judgments.  Finally, by using real Facebook profiles in PDF format, efforts were made to balance both internal and external validity.  The PDF format allowed for standardization of stimuli between participants, and the profiles themselves were minimally altered from those that a participant might see when browsing Facebook ordinarily, evoking impression formation practices that more closely mimic those used in his or her daily life.  As a result, the processes and outcomes explored in this study are more naturalistic than those derived from vignette studies and are therefore more applicable to the actual SNS-related behaviours, cognitions, and emotional outcomes experienced by young adults. If the current study is replicated, alterations to the procedure will help to clarify and expand upon the results yielded.  Several potential changes have already been identified:   68repeating the study with a clinical sample; creating or selecting profiles of targets scoring high on social rank-related traits; allowing participants to complete the study alone and with more time allotted; assessing participant judgments of other target characteristics (e.g., physical attractiveness).  In addition, although the use of university-age participants was an asset to the current study, extending the findings to non-undergraduate community samples would establish greater confidence in these and future conclusions.  Arguably, the most important next step is to identify the mechanism underlying the relationship between social anxiety and decline in state self-esteem and positive affect following profile browsing on Facebook.  Therefore, in future research, participants’ social rank self-ratings should be assessed to determine whether self-perceptions of inferiority—rather than perceptions of others’ superiority—account for these negative emotional consequences. Conclusions  The current study tied together two distinct areas of research pertaining to social anxiety: social rank judgments and problematic Internet use.  As expected, more socially anxious young adults suffered greater decreases in state self-esteem and positive affect after browsing and making judgments of unknown peers’ Facebook profiles.  Previous research has pointed to the potential harms of passive SNS use; the current results provide preliminary evidence that profile browsing, in particular, results in negative emotional consequences in general and especially for socially anxious individuals.  Contrary to expectations, these heightened consequences were not accounted for by a social rank judgment bias in the form of higher perceptions of peers’ social rank.  Future research should explore other social cognitive processes that might account for these heightened consequences, including lower self-perceptions of social rank, in order to develop a clearer understanding of the emotional   69risks of SNS-based social judgments for socially anxious people.  Given the popularity of SNS as a means of connecting with others, particularly among young adults, this line of research has important implications for understanding the nature of SNS use in at-risk populations and for targeting relevant cognitions and behaviours in the treatment setting.     70References  Aderka, I. M., Haker, A., Marom, S., Hermesh, H., & Gilboa-Schechtman, E. (2013).  Information-seeking bias in social anxiety disorder. Journal of Abnormal Psychology, 122(1), 7-12. doi:10.1037/a0029555  Aderka, I. M., Weisman, O., Shahar, G., & Gilboa-Schechtman, E. (2009). The roles of the  social rank and attachment systems in social anxiety. Personality and Individual Differences, 47(4), 284-288. doi:10.1016/j.paid.2009.03.014  Alden, L. E., & Bieling, P. (1998). Interpersonal consequences of the pursuit of safety.  Behaviour Research and Therapy, 36(1), 53-64. doi:10.1016/S0005-7967(97)00072-7  Alden, L. E., & Taylor, C. T. (2004). Interpersonal processes in social phobia. Clinical  Psychology Review, 24(7), 857-882. doi:10.1016/j.cpr.2004.07.006  Alden, L. E., & Taylor, C. T. (2010). Interpersonal processes in social anxiety disorder. In J.  G. Beck (Ed.), Interpersonal processes in the anxiety disorders: Implications for understanding psychopathology and treatment (pp. 125-152). Washington, DC: American Psychological Association. doi:10.1037/12084-005  Alden, L. E., & Wallace, S. T. (1995). Social phobia and social appraisal in successful and  unsuccessful social interactions. Behaviour Research and Therapy, 33(5), 497-505. doi:10.1016/0005-7967(94)00088-2  American Psychiatric Association (2013). Diagnostic and statistical manual of mental  disorders (5th ed.). Arlington, VA: American Psychiatric Association. doi:10.1176/appi.books.9780890425596  Antony, M. M., Rowa, K., Liss, A., Swallow, S. R., & Swinson, R. P. (2005). Social  comparison processes in social phobia. Behavior Therapy, 36(1), 65-75.  doi:10.1016/S0005-7894(05)80055-3  Appel, H., Crusius, J., & Gerlach, A. L. (2015). Social comparison, envy, and depression on  Facebook: A study looking at the effects of high comparison standards on depressed individuals. Journal of Social and Clinical Psychology, 34(4), 277-289. doi:10.1521/jscp.2015.34.4.277  Back, M. D., Stopfer, J. M., Vazire, S., Gaddis, S., Schmukle, S. C., Egloff, B., & Gosling,  S. D. (2010). Facebook profiles reflect actual personality, not self-idealization. Psychological Science, 21(3), 372-374. doi:10.1177/0956797609360756   Bates, D., Maechler, M., Bolker, B., & Walker, S. (2015). Fitting linear fixed-effects models  using lme4. Journal of Statistical Software, 67(1), 1-48. doi:10.18637/jss.v067.i01     71Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal  attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497-529. doi:10.1037/0033-2909.117.3.497  Beard, K. W., & Wolf, E. M. (2001). Modification in the proposed diagnostic criteria for  Internet addiction. CyberPsychology & Behavior, 4(3), 377-383. doi:10.1089/109493101300210286  Bonetti, L., Campbell, M. A., & Gilmore, L. (2010). The relationship of loneliness and social  anxiety with children’s and adolescents’ online communication. Cyberpsychology, Behavior, and Social Networking, 13(3), 279-285. doi:10.1089/cyber.2009.0215  Brown, T. A., Chorpita, B. F., & Barlow, D. H. (1998). Structural relationships among  dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal. Journal of Abnormal Psychology, 107(2), 179-192. doi:10.1037/0021-843X.107.2.179  Brunskill, D. (2013). Social media, social avatars and the psyche: Is Facebook good for us?  Australasian Psychiatry, 21(6), 527-532. doi:10.1177/1039856213509289  Buhrmester, D., & Furman, W. (1987). The development of companionship and intimacy.  Child Development, 58(4), 1101-1113. doi:10.2307/1130550  Burke, M., Marlow, C., & Lento, T. (2010). Social network activity and social well-being.  Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 1909-1912. doi:10.1145/1753326.1753613  Campbell, A. J., Cumming, S. R., & Hughes, I. (2006). Internet use by the socially fearful:  Addiction or therapy? CyberPsychology & Behavior, 9(1), 69-81.  doi:10.1089/cpb.2006.9.69  Canadian Internet Registration Authority (2014). The Canadian Internet. Retrieved 16th June  2016 from https://cira.ca/factbook/2014/the-canadian-internet.html  Caplan, S. E. (2002). Problematic Internet use and psychosocial well-being: Development of  a theory-based cognitive-behavioral measurement instrument. Computers in Human Behavior, 18(5), 553-575. doi:10.1016/S0747-5632(02)00004-3  Caplan, S. E. (2003). Preference for online social interaction: A theory of problematic  Internet use and psychosocial well-being. Communication Research, 30(6), 625-648. doi:10.1177/0093650203257842  Caplan, S. E. (2005). A social skill account of problematic Internet use. Journal of  Communication, 55(4), 721-736. doi:10.1111/j.1460-2466.2005.tb03019.x     72Caplan, S. E. (2007). Relations among loneliness, social anxiety, and problematic Internet  use. CyberPsychology & Behavior, 10(2), 234-242. doi:10.1089/cpb.2006.9963  Chiao, J. Y. (2010). Neural basis of social status hierarchy across species. Current Opinions  in Neurobiology, 20(6), 803-809. doi:10.1016/j.conb.2010.08.006  Chou, H. T. G., & Edge, N. (2012). “They are happier and having better lives than I am”:  The impact of using Facebook on perceptions of others’ lives. Cyberpsychology, Behavior, and Social Networking, 15(2), 117-121. doi:10.1089/cyber.2011.0324  Christensen, P. N., Stein, M. B., & Means-Christensen, A. (2003). Social anxiety and  interpersonal perception: A social relations model analysis. Behaviour Research and Therapy, 41(11), 1355-1371. doi:10.1016/S0005-7967(03)00064-0  Clark, D. M., & Wells, A. (1995). A cognitive model of social phobia. In R. G. Heimberg,  M. R. Liebowitz, D. A. Hope, & F. R. Schneider (Eds.), Social phobia: Diagnosis, assessment, and treatment (pp. 69-93). New York: Guilford Press.  Correa, T., Hinsley, A. W., & de Zúñiga, H. G. (2010). Who interacts on the Web? The  intersection of users’ personality and social media use. Computers in Human Behavior, 26(2), 247-253. doi:10.1016/j.chb.2009.09.003  Costa, P. T., Jr., Terracciano, A., & McCrae, R. R. (2001). Gender differences in personality  traits across cultures: Robust and surprising findings. Journal of Personality and Social Psychology, 81(2), 322-331. doi:10.1037/0022-3514.81.2.322  Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four  recommendations for getting the most from your analysis. Practical Assessment, Research & Evaluation, 10(7). Retrieved 26th March 2016 from http://pareonline.net/getvn.asp?v=10&n=7  Coyne, S. M., Padilla-Walker, L. M., & Howard, E. (2013). Emerging in a digital world: A  decade review of media use, effects, and gratifications in emerging adulthood. Emerging Adulthood, 1(2), 125-137. doi:10.1177/2167696813479782  Cuming, S., & Rapee, R. M. (2010). Social anxiety and self-protective communication style  in close relationships. Behaviour Research and Therapy, 48(2), 87-96. doi:10.1016/j.brat.2009.09.010  Davidson, J. R. T., Hughes, D. C., George, L. K., & Blazer, D. G. (1994). The boundary of  social phobia: Exploring the threshold. Archives of General Psychiatry, 51(12), 975-983. doi:10.1001/archpsyc.1994.03950120047008  Davis, R. A. (2001). A cognitive-behavioral model of pathological Internet use. Computers  in Human Behavior, 17(2), 187-195. doi:10.1016/S0747-5632(00)00041-8    73de Vries, D. A., & Kühne, R. (2015). Facebook and self-perception: Individual susceptibility  to negative social comparison on Facebook. Personality and Individual Differences, 86, 217-221. doi:10.1016/j.paid.2015.05.029  DelPriore, D. J., Hill, S. E., & Buss, D. M. (2012). Envy: Functional specificity and sex- differentiated design features. Personality and Individual Differences, 53(3), 317-322. doi:10.1016/j.paid.2012.03.029  Erwin, B. A., Turk, C. L., Heimberg, R. G., Fresco, D. M., & Hantula, D. A. (2004). The  Internet: Home to a severe population of individuals with social anxiety disorder?  Journal of Anxiety Disorders, 18(5), 629-646. doi:10.1016/j.janxdis.2003.08.002  Feinstein, B. A, Hershenberg, R., Bhatia, V., Latack, J. A., Meuwly, N., & Davila, J. (2013).  Negative social comparison on Facebook and depressive symptoms: Rumination as a mechanism. Psychology of Popular Media Culture, 2(3), 161-170. doi:10.1037/a0033111   Fernandez, K. C., Levinson, C. A., & Rodebaugh, T. L. (2012). Profiling: Predicting social  anxiety from Facebook profiles. Social Psychological and Personality Science, 3(6), 706-713. doi:10.1177/1948550611434967  Forum Research (2015). 2015 Canadian social media usage statistics. Retrieved 16th June  2016 from canadiansinternet.com/2015-canadian-social-media-usage-statistics  Galili, L., Amir, O., & Gilboa-Schechtman, E. (2013). Acoustic properties of dominance and  request utterances in social anxiety. Journal of Social and Clinical Psychology, 32(6), 651-673.  Gergen, K. J. (1968). Personal consistency and the presentation of self. In C. Gordon & K. J.  Gergen (Eds.), The self in social interaction (Vol. 1, pp. 299-308). New York: Wiley.  Gibbons, F. X., & Buunk, B. P. (1999). Individual differences in social comparison:  Development of a scale of social comparison orientation. Journal of Personality and Social Psychology, 76(1), 129-142. doi:10.1037/0022-3514.76.1.129   Gilbert, P. (2001). Evolution and social anxiety: The role of attraction, social competition,  and social hierarchies. The Psychiatric Clinics of North America, 24(4), 723-751. doi:10.1016/S0193-953X(05)70260-4  Gilbert, P. (2004). Evolutionary approaches to psychopathology and cognitive therapy. In P.  Gilbert (Ed.), Evolutionary theory and cognitive therapy. New York: Springer. Retrieved 8th April 2015 from http://site.ebrary.com/lib/ubc/detail.action?docID=10265296      74Gilbert, P., & Trower, P. (2001). Evolution and process in social anxiety. In W. R. Crozier &  L. E. Alden (Eds.), International handbook of social anxiety: Concepts, research and interventions relating to the self and shyness (pp. 259-279). New York: John Wiley & Sons.  Gilboa-Schechtman, E., Foa, E. B., & Amir, N. (1999). Attentional biases for facial  expressions in social phobia: The face-in-the-crowd paradigm. Cognition and Emotion, 13(3), 305-318. doi:10.1080/026999399379294  Gilboa-Schechtman, E., Friedman, L., Helpman, L., & Kananov, J. (2013). Self-evaluations  of social rank and affiliation in social anxiety: Explicit and implicit measures. International Journal of Cognitive Therapy, 6(3), 208-220. doi:10.1521/ijct.2013.6.3.208  Gosling, S. D., Augustine, A. A., Vazire, S., Holtzman, N., & Gaddis, S. (2011).  Manifestations of personality in online social networks: Self-reported Facebook-related behaviors and observable profile information. Cyberpsychology, Behavior, and Social Networking, 14(9), 483-488. doi:10.1089/cyber.2010.0087  Graham, J. W., Taylor, B. J., Olchowski, A. E., & Cumsille, P. E. (2006). Planned missing  data designs in psychological research. Psychological Methods, 11(4), 323-343.  doi:10.1037/1082-989X.11.4.323  Green, T., Wilhelmsen, T., Wilmots, E., Dodd, B., & Quinn, S. (2016). Social anxiety,  attributes of online communication and self-disclosure across private and public Facebook communication. Computers in Human Behavior, 58, 206-213. doi:10.1016/j.chb.2015.12.066  Grieve, R., Indian, M., Witteveen, K., Tolan, G. A., & Marrington, J. (2013). Face-to-face or  Facebook: Can social connectedness be derived online? Computers in Human Behavior, 29(3), 604-609. doi:10.1016/j.chb.2012.11.017  Gross, E. F., Juvonen, J., & Gable, S. L. (2002). Internet use and well-being in adolescence.  Journal of Social Issues, 58(1), 75-90. doi:10.1111/1540-4560.00249  Gurtman, M. B. (1991). Evaluating the interpersonalness of personality scales. Personality  and Social Psychology Bulletin, 17(6), 670-677. doi:10.1177/0146167291176010  Haker, A., Aderka, I. M., Marom, S., Hermesh, H., & Gilboa-Schechtman, E. (2014).  Impression formation and revision in social anxiety disorder. Journal of Anxiety Disorders, 28(2), 133-139. doi:10.1016/j.janxdis.2013.05.001  Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for measuring  state self-esteem. Journal of Personality and Social Psychology, 60(6), 895-910. doi:10.1037/0022-3514.60.6.895    75Heerey, E. A., & Kring, A. M. (2007). Interpersonal consequences of social anxiety. Journal  of Abnormal Psychology, 116(1), 125-134. doi:10.1037/0021-843X.116.1.125  Heimberg, R. G., Mueller, G. P., Holt, C. S., Hope, D. A., & Liebowitz, M. R. (1992).  Assessment of anxiety in social interaction and being observed by others: The Social Interaction Anxiety Scale and the Social Phobia Scale. Behavior Therapy, 23(1), 53-73. doi:10.1016/S0005-7894(05)80308-9  Heiser, N. A., Turner, S. M., Beidel, D. C., & Roberson-Nay, R. (2009). Differentiating  social phobia from shyness. Journal of Anxiety Disorders, 23(4), 469-476. doi:10.1016/j.janxdis.2008.10.002  High, A. C., & Caplan, S. E. (2009). Social anxiety and computer-mediated communication  during initial interactions: Implications for the hyperpersonal perspective. Computers in Human Behavior, 25(2), 475-482. doi:10.1016/j.chb.2008.10.011  Hope, D. A., Sigler, K. D., Penn, D. L., & Meier, V. (1998). Social anxiety, recall of interpersonal information, and social impact on others. Journal of Cognitive Psychotherapy, 12(4), 303-322. Retrieved 10th April 2015 from http://www.ingentaconnect.com/content/springer/jcogp/1998/00000012/00000004/art00003  Huan, V. S., Ang, R. P., & Chye, S. (2014). Loneliness and shyness in adolescent  problematic Internet users: The role of social anxiety. Child & Youth Care Forum, 43(5), 539-551. doi:10.1007/s10566-014-9252-3  Indian, M., & Grieve, R. (2014). When Facebook is easier than face-to-face: Social support  derived from Facebook in socially anxious individuals. Personality and Individual Differences, 59, 102-106. doi:10.1016/j.paid.2013.11.016  Irons, C., & Gilbert, P. (2005). Evolved mechanisms in adolescent anxiety and social rank  systems. Journal of Adolescence, 28(3), 325-341. doi:10.1016/j.adolescence.2004.07.004  Jones, E. E., & Pittman, T. S. (1982). Toward a general theory of strategic self-presentation.  In J. Suls (Ed.), Psychological perspectives on the self (Vol. 1, pp. 231-262). Hillsdale, NJ: Lawrence Erlbaum Associates.  Kashdan, T. B., Weeks, J. W., & Savostyanova, A. A. (2011). Whether, how, and when  social anxiety shapes positive experiences and events: A self-regulatory framework and treatment implications. Clinical Psychology Review, 31(5), 786-799. doi:10.1016/j.cpr.2011.03.012  Kelley, H. H. (1973). The process of causal attribution. American Psychologist, 28(2), 107- 128. doi:10.1037/h0034225    76Kessler, R. C., Petukhova, M., Sampson, N. A., Zaslavsky, A. M., & Wittchen, H. U. (2012).  Twelve-month and lifetime prevalence and lifetime morbid risk of anxiety and mood disorders in the United States. International Journal of Methods in Psychiatric Research, 21(3), 169-184. doi:10.1002/mpr.1359  Kiefer, S. M., & Ryan, A. M. (2008). Striving for social dominance over peers: The  implications for academic adjustment during early adolescence. Journal of Educational Psychology, 100(2), 417-428. doi:10.1037/0022-0663.100.2.417   Kiesler, D. J. (1983). The 1982 interpersonal circle: A taxonomy for complementarity in  human transactions. Psychological Review, 90(3), 185-214. doi:10.1037/0033-295X.90.3.185  Knobeloch, M. R. (2013). The effects of timing and visual cues on social anxiety in online  communication (Doctoral dissertation). Retrieved 10th June 2016 from ProQuest Dissertations and Theses database. (Publication No. 3604438).  Koo, H. J., Woo, S., Yang, E., & Kwon, J. H. (2015). The double meaning of online social  space: Three-way interactions among social anxiety, online social behavior, and offline social behavior. Cyberpsychology, Behavior, and Social Networking, 18(9), 514-520. doi:10.1089/cyber.2014.0396  Koutamanis, M., Vossen, H. G. M., Peter, J., & Valkenburg, P. M. (2013). Practice makes  perfect: The longitudinal effect of adolescents’ instant messaging on their ability to initiate offline friendships. Computers in Human Behavior, 29(6), 2265-2272. doi:10.1016/j.chb.2013.04.033  Larsen, R. J., & Diener, E. (1992). Promises and problems with the circumplex model of  emotion. Thousand Oaks, CA: Sage Publications.  Leary, M. R. (2010). Social anxiety as an early warning system: A refinement and extension  of the self-presentation theory of social anxiety. In S. G. Hoffman & P. M. DiBartolo (Eds.), Social anxiety: Clinical, developmental, and social perspectives (pp. 471-486). San Diego, CA: Elsevier. doi:10.1016/B978-0-12-375096-9.00018-3  Leary, M. R., & Kowalski, R. M. (1997). Social anxiety. New York: Guilford Press.   Leary, M. R., Kowalski, R. M., & Campbell, C. D. (1988). Self-presentational concerns and  social anxiety: The role of generalized impression expectancies. Journal of Research in Personality, 22(3), 308-321. doi:10.1016/0092-6566(88)90032-3  Lee, B. W., & Stapinski, L. A. (2012). Seeking safety on the internet: Relationship between  social anxiety and problematic internet use. Journal of Anxiety Disorders, 26(1), 197-205. doi:10.1016/j.janxdis.2011.11.001     77Lee, S. Y. (2014). How do people compare themselves with others on social network sites?:  The case of Facebook. Computers in Human Behavior, 32, 253-260. doi:10.1016/j.chb.2013.12.009  Linton, K. E., & Richard, G. M. (1996). Self-esteem in adolescents: Validation of the State  Self-Esteem Scale. Personality and Individual Differences, 21(1), 85-90. doi:10.1016/0191-8869(96)83741-X  Liu, X. (2003). Statistical power and optimum sample allocation ratio for treatment and  control having unequal costs per unit of randomization. Journal of Educational and Behavioral Statistics, 28(3), 231-248. doi:10.3102/10769986028003231  Lueders, A., Hall, J. A., Pennington, N. R., & Knutson, K. (2014). Nonverbal decoding on  Facebook: Applying the IPT-15 and the SSI to personality judgments. Journal of Nonverbal Behavior, 38(4), 413-427. doi:10.1007/s10919-014-0195-z  Markovitzky, O., Anholt, G. E., & Lipsitz, J. D. (2012). Haven’t we met somewhere before?  The effects of a brief internet introduction on social anxiety in a subsequent face to face interaction. Behaviour Research and Therapy, 50(5), 359-365. doi:10.1016/j.brat.2012.02.002  Mattick, R. P., & Clarke, J. C. (1998). Development and validation of measures of social  phobia scrutiny fear and social interaction anxiety. Behaviour Research and Therapy, 36(4), 455-470. doi:10.1016/S0005-7967(97)10031-6  Mazalin, D., & Klein, B. (2008). Social anxiety and the Internet: Positive and negative  effects. Sensoria: A Journal of Mind, Brain and Culture, 4(2), 43-50. doi:10.7790/ejap.v4i2.137  Mazalin, D., & Moore, S. (2004). Internet use, identity development and social anxiety  among young adults. Behaviour Change, 21(2), 90-102. doi:10.1375/bech.21.2.90.55425  McCord, B., Rodebaugh, T. L., & Levinson, C. A. (2014). Facebook: Social uses and  anxiety. Computers in Human Behavior, 34, 23-27. doi:10.1016/j.chb.2014.01.020  Mehdizadeh, S. (2010). Self-presentation 2.0: Narcissism and self-esteem on Facebook.  Cyberpsychology, Behavior, and Social Networking, 13(4), 357-364. doi:10.1089/cyber.2009.0257  Mendlowicz, M. V., & Stein, M. B. (2000). Quality of life in individuals with anxiety  disorders. American Journal of Psychiatry, 157(5), 669-682. doi:10.1176/appi.ajp.157.5.669      78Miller, C. T. (1982). The role of performance-related similarity in social comparison of  abilities: A test of the related attributes hypothesis. Journal of Experimental Social Psychology, 18(6), 513-523. doi:10.1016/0022-1031(82)90070-1  Mitchell, M. A., & Schmidt, N. B. (2014). An experimental manipulation of social  comparison in social anxiety. Cognitive Behavior Therapy, 43(3), 221-229. doi:10.1080/16506073.2014.914078   Mo, R., Leung, L., Hao, Y., Wu, X., Xi, R., & Zhang, S. (2014). Examining the mediating  roles of microblog use in the relationships between narcissism, social anxiety, and social capital. International Journal of Cyber Behavior, Psychology and Learning, 4(2), 58-75.  Moore, K., & McElroy, J. C. (2012). The influence of personality on Facebook usage, wall  postings, and regret. Computers in Human Behavior, 28(1), 267-274. doi:10.1016/j.chb.2011.09.009  Moskowitz, D. S., Suh, E. J., & Desaulniers, J. (1994). Situational influences on gender  differences in agency and communion. Journal of Personality and Social Psychology, 66(4), 753-761. doi:10.1037/0022-3514.66.4.753  Muench, F., Hayes, M., Kuerbis, A., & Shao, S. (2015). The independent relationship  between trouble controlling Facebook use, time spent on the site and distress. Journal of Behavioral Addictions, 4(3), 163-169. doi:10.1556/2006.4.2015.013  Nadkarni, A., & Hofmann, S. G. (2012). Why do people use Facebook? Personality and  Individual Differences, 52(3), 243-249. doi:10.1016/j.paid.2011.11.007  Nesi, J., & Prinstein, M. J. (2015). Using social media for social comparison and feedback- seeking: Gender and popularity moderate associations with depressive symptoms. Journal of Abnormal Child Psychology, 43(8), 1427-1438. doi:10.1007/s10802-015-0020-0   Osman, A., Gutierrez, P. M., Barrios, F. X., Kopper, B. A., & Chiros, C. E. (1998). The  Social Phobia and Social Interaction Anxiety Scales: Evaluation of psychometric properties. Journal of Psychopathology and Behavioral Assessment, 20(3), 249-264. doi:10.1023/A:1023067302227  Peter, J., & Valkenburg, P. M. (2006). Research note: Individual differences in perceptions  of internet communication. European Journal of Communication, 21(2), 213-226. doi:10.1177/0267323105064046  Pierce, T. (2009). Social anxiety and technology: Face-to-face communication versus  technological communication among teens. Computers in Human Behavior, 25(6), 1367-1372. doi:10.1016/j.chb.2009.06.003    79Quinn, S., & Oldmeadow, J. A. (2013). Is the igeneration a “we”generation? Social  networking use among 9- to 13-year-olds and belonging. The British Journal of Developmental Psychology, 31(1), 136-142. doi:10.1111/bjdp.12007  Rapee, R. M., & Heimberg, R. G. (1997). A cognitive-behavioral model of anxiety in social  phobia. Behaviour Research and Therapy, 35(8), 741-756. doi:10.1016/S0005-7967(97)00022-3  Rauch, S. M., Strobel, C., Bella, M., Odachowski, Z., & Bloom, C. (2014). Face to face  versus Facebook: Does exposure to social networking web sites augment or attenuate physiological arousal among the socially anxious? Cyberpsychology, Behavior, and Social Networking, 17(3), 187-190. doi:10.1089/cyber.2012.0498   Richard, F. D., Bond, C. F., Jr., & Stokes-Zoota, J. J. (2003). One hundred years of social  psychology quantitatively described. Review of General Psychology, 7(4), 331-363. doi:10.1037/1089-2680.7.4.331   Rodebaugh, T. L. (2009). Social phobia and perceived friendship quality. Journal of Anxiety  Disorders, 23(7), 872-878. doi:10.1016/j.janxdis.2009.05.001  Rodebaugh, T. L., Bielak, T., Vidovic, V., & Moscovitch, D. A. (2016). The effects of social  anxiety on interpersonal evaluations of warmth and dominance. Journal of Anxiety  Disorders, 38, 68-78. doi:10.1016/j.janxdis.2016.01.002  Rodebaugh, T. L., Heimberg, R. G., Brown, P. J., Fernandez, K. C., Blanco, C., Schneier, F.  R., & Liebowitz, M. R. (2011). More reasons to be straightforward: Findings and norms for two scales relevant to social anxiety. Journal of Anxiety Disorders, 25(5), 623-630. doi:10.1016/j.janxdis.2011.02.002  Rodebaugh, T. L., Woods, C. M., & Heimberg, R. G. (2007). The reverse of social anxiety is  not always the opposite: The reverse-scored items of the Social Interaction Anxiety Scale do not belong. Behavior Therapy, 38(2), 192-206. doi:10.1016/j.beth.2006.08.001  Rosenberg, F. R., & Simmons, R. G. (1975). Sex differences in the self-concept in  adolescence. Sex Roles, 1(2), 147-159. doi:10.1007/BF00288008  Sapolsky, R. M. (2005). The influence of social hierarchy on primate health. Science,  308(5722), 648-652. doi:10.1126/science.1106477  Scherbaum, C. A., & Ferreter, J. M. (2008). Estimating statistical power and required sample  sizes for organizational research using multilevel modeling. Organizational Research Methods, 12(2), 347-367. doi:10.1177/1094428107308906  Schlenker, B. R. (1980). Impression management: The self-concept, social identity, and  interpersonal relations. Monterey, CA: Brooks/Cole.    80Schlenker, B. R., & Leary, M. R. (1982). Social anxiety and self-presentation: A  conceptualization model. Psychological Bulletin, 92(3), 641-669. doi:10.1037/0033-2909.92.3.641  Schouten, A., Valkenburg, P., & Peter, J. (2007). Precursors and underlying processes of  adolescents’ online self-disclosure: Developing and testing an “Internet-Attribute-Perception” model. Media Psychology, 10(2), 292-315. doi:10.1080/15213260701375686  Selfhout, M. H. W., Branje, S. J. T., Delsing, M., ter Bogt, T. F. M., & Meeus, W. H. J.  (2009). Different types of Internet use, depression, and social anxiety: The role of perceived friendship quality. Journal of Adolescence, 32(4), 819-833. doi:10.1016/j.adolescence.2008.10.011  Shaw, A. M., Timpano, K. R., Tran, T. B., & Joormann, J. (2015). Correlates of Facebook  usage patterns: The relationship between passive Facebook use, social anxiety symptoms, and brooding. Computers in Human Behavior, 48, 575-580. doi:10.1016/j.chb.2015.02.003  Shepherd, R. M., & Edelmann, R. J. (2005). Reasons for internet use and social anxiety.  Personality and Individual Differences, 39(5), 949-958. doi:10.1016/j.paid.2005.04.001  Sparrevohn, R. M., & Rapee, R. M. (2009). Self-disclosure, emotional expression and  intimacy within romantic relationships of people with social phobia. Behaviour Research and Therapy, 47(12), 1074-1078. doi:10.1016/j.brat.2009.07.016  Steers, M. L. N., Wickham, R. E., & Acitelli, L. K. (2014). Seeing everyone else’s highlight  reels: How Facebook usage is linked to depressive symptoms. Journal of Social and Clinical Psychology, 33(8), 701-731. doi:10.1521/jscp.2014.33.8.701  Swallow, S. R., & Kuiper, N. A. (1988). Social comparison and negative self-evaluations:  An application to depression. Clinical Psychology Review, 8(1), 55-76. doi:10.1016/0272-7358(88)90049-9  Tandoc, E. C., Jr., Ferrucci, P., & Duffy, M. (2015). Facebook use, envy, and depression  among college students: Is facebooking depressing? Computers in Human Behavior, 43, 139-146. doi:10.1016/j.chb.2014.10.053  Thompson, E. R. (2007). Development and validation of an internationally reliable short- form of the Positive and Negative Affect Schedule (PANAS). Journal of Cross-Cultural Psychology, 38(2), 227-242. doi:10.1177/0022022106297301  Trower, P., & Gilbert, P. (1989). New theoretical conceptions of social anxiety and social  phobia. Clinical Psychology Review, 9(1), 19-35. doi:10.1016/0272-7358(89)90044-5    81Turk, C. L., & Heimberg, R. G., & Hope, D. A. (2001). Social anxiety disorder. In D. H.  Barlow (Ed.), Clinical handbook of psychological disorders: A step-by-step treatment  manual (3rd ed., pp. 114-154). New York: Guilford Press.  Valkenburg, P. M., & Peter, J. (2011). Online communication among adolescents: An  integrated model of its attraction, opportunities, and risks. Journal of Adolescent Health, 48(2), 121-127. doi:10.1016/j.jadohealth.2010.08.020  Valkenburg, P. M., Sumter, S. R., & Peter, J. (2011). Gender differences in online and offline  self-disclosure in pre-adolescence and adolescence. British Journal of Developmental Psychology, 29(2), 253-269. doi:10.1348/2044-835X.002001  Van der Leeden, R., & Busing, F. (1994). First iteration versus IGLS/RIGLS estimates in  two-level models: A Monte Carlo study with ML3. Unpublished manuscript, Leiden University, Leiden, Netherlands.  Verduyn, P., Lee, D. S., Park, J., Shablack, H., Orvell, A., Bayer, J., … Kross, E. (2015). Passive Facebook usage undermines affective well-being: Experimental and longitudinal evidence. Journal of Experimental Psychology: General, 144(2), 480-488. doi:10.1037/xge0000057   Vogel, E. A., Rose, J. P., Okdie, B. M., Eckles, K., & Franz, B. (2015). Who compares and  despairs? The effect of social comparison orientation on social media use and its outcomes. Personality and Individual Differences, 86, 249-256. doi:10.1016/j.paid.2015.06.026  Vogel, E. A., Rose, J. P., Roberts, L. R., & Eckles, K. (2014). Social comparison, social  media, and self-esteem. Psychology of Popular Media Culture, 3(4), 206-222. doi:10.1037/ppm0000047   Walters, K. S., & Hope, D. A. (1998). Analysis of social behavior in individuals with social  phobia and nonanxious participants using a psychobiological model. Behavior Therapy, 29(3), 387-407. doi:10.1016/S0005-7894(98)80039-7  Watson, D., & Clark, L. E. (1994). The PANAS-X: Manual for the Positive and Negative  Affect Schedule—Expanded Form. Iowa City, IA: University of Iowa. Retrieved 10th April 2016 from http://ir.uiowa.edu/cgi/viewcontent.cgi?article=1011&context=psychology_pubs  Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief  measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063-70. doi:10.1037//0022-3514.54.6.1063  Weeks, J. W., Heimberg, R. G., & Heuer, R. (2011). Exploring the role of behavioral  submissiveness in social anxiety. Journal of Social and Clinical Psychology, 30(3), 217-249. doi:10.1521/jscp.2011.30.3.217   82Weeks, J. W., Heimberg, R. G., Rodebaugh, T. L., Goldin, P. R., & Gross, J. J. (2012).  Psychometric evaluation of the Fear of Positive Evaluation Scale in patients with social anxiety disorder. Psychological Assessment, 24(2), 301-312. doi:10.1037/a0025723  Weeks, J. W., Heimberg, R. G., Rodebaugh, T. L., & Norton, P. J. (2008). Exploring the  relationship between fear of positive evaluation and social anxiety. Journal of Anxiety Disorders, 22(3), 386-400. doi:10.1016/j.janxdis.2007.04.009  Weeks, J. W., & Howell, A. N. (2012). The Bivalent Fear of Evaluation model of social  anxiety: Further integrating findings on fears of positive and negative evaluation. Cognitive Behaviour Therapy, 41(2), 83-95. doi:10.1080/16506073.2012.661452  Weeks, J. W., Jakatdar, T. A., & Heimberg, R. G. (2010). Comparing and contrasting fears  of positive and negative evaluation as facets of social anxiety. Journal of Social and Clinical Psychology, 29(1), 68-94. doi:10.1521/jscp.2010.29.1.68  Weeks, J. W., Rodebaugh, T. L., Heimberg, R. G., Norton, P. J., & Jakatdar, T. A. (2009).  “To avoid evaluation, withdraw”: Fears of evaluation and depressive cognitions lead to social anxiety and submissive withdrawal. Cognitive Therapy and Research, 33(4), 375-389. doi:10.1007/s10608-008-9203-0  Weidman, A. C., Fernandez, K. C., Levinson, C. A., Augustine, A. A., Larsen, R. J., &  Rodebaugh, T. L. (2012). Compensatory Internet use among individuals higher in social anxiety and its implications for well-being. Personality and Individual Differences, 53(3), 191-195. doi:10.1016/j.paid.2012.03.003  Weidman, A. C., & Levinson, C. A. (2015). I’m still socially anxious online: Offline  relationship impairment characterizing social anxiety manifests and is accurately perceived in online social networking profiles. Computers in Human Behavior, 49, 12-19. doi:10.1016/j.chb.2014.12.045  Weisman, O., Aderka, I. M., Marom, S., Hermesh, H., & Gilboa-Schechtman, E. (2011).  Social rank and affiliation in social anxiety disorder. Behaviour Research and Therapy, 49(6), 399-405. doi:10.1016/j.brat.2011.03.010  Wu, Y. C. J., Chang, W. H., & Yuan, C. H. (2015). Do Facebook profile pictures reflect  user’s personality? Computers in Human Behavior, 51, 880-889. doi:10.1016/j.chb.2014.11.014   Yen, J. Y., Yen, C. F., Chen, C. S., Wang, P. W., Chang, Y. H., & Ko, C. H. (2012). Social  anxiety in online and real-life interactions and their associated factors. Cyberpsychology, Behavior, and Social Networking, 15(1), 7-12. doi:10.1089/cyber.2011.0015      83Young, C. M. Y., & Lo, B. C. Y. (2012). Cognitive appraisal mediating relationship between  social anxiety and internet communication in adolescents. Personality and Individual Differences, 52(1), 78-83. doi:10.1016/j.paid.2011.09.001                                              84Appendices Appendix A: Protocol for Preparing Facebook Profile PDF Files 1. Log into the lab Facebook account and accept the study participant as a friend 2. Click on the “About” Tab of the participant’s profile  There are different sections to the “About” tab (i.e., About, Photos, Friends, Likes, Music, etc.)  Starting with whichever section is at the top, check to see if you can click “See All” to expand the section  Note that for certain sections like Likes or Music, if there are multiple tabs within that section, you want to screenshot whichever of those tabs has the most things 3. Click on the “Awesome Screenshot” application and select “Capture Selected Area”  Note: Make sure you are using Google Chrome 4. Click and drag the section of the profile that you want to save and click on the “Capture” button 5. Another window will automatically open with the section of the profile that was just cropped 6. Click on the “Blur” tool (it’s the blue waterdrop in the toolbar at the top) and blur out any information that could identify the participant (e.g., name, friends’ names, birthday, email address, phone number, home address, etc.) 7. Click on “Done” at the top beside the toolbar and you will be taken to another page 8. Click on “Save” which is on the right hand side under the box “Save Local” 9. Go back to the Facebook profile and repeat steps 4-9 for all other sections under the “About” tab 10. Once you have all the parts of the profile, open the program “Combine PDFs” and select all the files you want to combine and then click on the “Merge” button  Save the document as the participant ID number 11. Show the PDF to the participant at the end of the study and ask them to verify that they are comfortable with all the information that is in the document  If the participant wishes to have additional information blurred out, make a note of this and correct the PDF accordingly   85Appendix B: State Self-Esteem Scale (SSES),  Social Self-Esteem Subscale  This is a questionnaire designed to measure what you are thinking at this moment. There is of course no right answer for any statement. The best answer is what you feel is true of yourself at the moment. Be sure to answer all of the items, even if you are not certain of the best answer. Again, answer the statements as they are true for you RIGHT NOW.   Not at all A little bit Somewhat Very much Extremely 1. My experience at UBC has been positive so far. 1 2 3 4 5 2. I am worried about whether I am regarded as a success or failure. 1 2 3 4 5 3. I feel self-conscious. 1 2 3 4 5 4. I am happy with my current living arrangement. 1 2 3 4 5 5. I feel displeased with myself. 1 2 3 4 5 6. I am worried about what other people think of me. 1 2 3 4 5 7. I feel uncertain about which major to choose. 1 2 3 4 5 8. I feel inferior to others at this moment. 1 2 3 4 5 9. I feel concerned about the impression I am making. 1 2  3 4 5 10. I am worried about looking foolish. 1 2 3 4 5         86Appendix C: The International Positive and Negative Affect Schedule Short Form      (I-PANAS-SF), Adapted  This scale consists of a number of words that describe different feelings and emotions. Please indicate to what extent you currently feel:   Very slightly or Not at all  A little  Moderately  Quite a bit  Extremely 1. Content 1 2 3 4 5 2. Anxious 1 2 3 4 5 3. Interested 1 2 3 4 5 4. Sluggish 1 2 3 4 5 5. Confident 1 2 3 4 5 6. Sad 1 2 3 4 5 7. Cheerful 1 2 3 4 5 8. Irritable 1 2 3 4 5 9. Relaxed 1 2 3 4 5 10. Distressed 1 2 3 4 5              87Appendix D: Explicit Domain-Specific Social Estimations Scale (EDSSE), Adapted Please indicate the extent to which you agree that each of the following traits is likely to be descriptive of the person whose Facebook profile you just viewed.   Disagree strongly Disagree Disagree a little Neither agree nor disagree Agree a little Agree Agree strongly 1. Strong 1 2 3 4 5 6 7 2. *Distant 1 2 3 4 5 6 7 3. *Passive 1 2 3 4 5 6 7 4. Friendly 1 2 3 4 5 6 7 5. Dominant 1 2 3 4 5 6 7 6. *Cold 1 2 3 4 5 6 7 7. *Cowardly 1 2 3 4 5 6 7 8. Pleasant 1 2 3 4 5 6 7 9. Authoritative 1 2 3 4 5 6 7 10. *Hostile 1 2 3 4 5 6 7 11. *Submissive 1 2 3 4 5 6 7 12. Generous 1 2 3 4 5 6 7 13. Assertive 1 2 3 4 5 6 7 14. *Mean 1 2 3 4 5 6 7 15. *Weak 1 2 3 4 5 6 7 16. Considerate 1 2 3 4 5 6 7      88Appendix E: Straightforward Social Interaction Anxiety Scale (S-SIAS) Please indicate the degree to which you feel each statement is characteristic or true of you.  Not at all Slightly Moderately Very Extremely 1. I get nervous if I have to speak with someone in authority (teacher, boss, etc.). 0 1 2 3 4 2. I have difficulty making eye contact with others. 0 1 2 3 4 3. I become tense if I have to talk about myself or my feelings. 0 1 2 3 4 4. I find it difficult to mix comfortably with the people I work with. 0 1 2 3 4 5. I tense up if I meet an acquaintance in the street. 0 1 2 3 4 6. When mixing socially, I am uncomfortable. 0 1 2 3 4 7. I feel tense if I am alone with just one other person. 0 1 2 3 4 8. I have difficulty talking with other people. 0 1 2 3 4 9. I worry about expressing myself in case I appear awkward. 0 1 2 3 4 10. I find it difficult to disagree with another’s point of view. 0 1 2 3 4 11. I have difficulty talking to attractive persons of the opposite sex. 0 1 2 3 4 12. I find myself worrying that I won’t know what to say in social situations. 0 1 2 3 4 13. I am nervous mixing with people I don’t know well. 0 1 2 3 4 14. I feel I’ll say something embarrassing when talking. 0 1 2 3 4  15. When mixing in a group, I find myself worrying I will be ignored. 0 1 2 3 4 16. I am tense mixing in a group. 0 1 2 3 4 17. I am unsure whether to greet someone I know only slightly. 0 1 2 3 4   89Appendix F: Demographics Questionnaire Please complete the following information about yourself as appropriate. Your age: ______  Your gender: Male ___ Female ___  Your marital status:  ___ Single ___ Cohabitating ___ Married   ___ Separated ___ Divorced ___ Other (please specify) _______________  Your cultural background:  ___ African or Carribean/West Indian ___ Central Asian/Middle Eastern (e.g., Israeli, Palestinian, Iranian) ___ Caucasian (e.g., European, Australian) ___ Chinese ___ Japanese ___ Korean ___ Native (e.g., First Nations, Metis, Inuit) ___ South Asian (e.g., East Indian, Sri Lankan, Pakistani) ___ South or Latin American (e.g., Mexican, Brazilian, Chilean) ___ Southeast Asian (e.g., Vietnamese, Cambodian, Thai) ___ Other (please specify) __________________________  Your place of birth: _____________________ If you were not born in Canada, how many years have you been in Canada? _____ years   Your parents’ place(s) of birth: _____________________________________  Your first language: ____________________  If English is not your first language, how long have you spoken English? _____ years  How many years of university have you completed? (please round to the nearest half year you’ve completed) _____ years  Your major (indicate if undecided): __________  In an average week, how many days do you use Facebook? (please respond with a single number) _____     

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.24.1-0308071/manifest

Comment

Related Items