This item is only available as the following downloads:
1 ANTI TRANSGENDER PREJUDICE: A STRUCTURAL EQUATION MODEL OF ASSOCIATED CONSTRUCTS By ESTHER N. TEBBE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2011
2 2011 Esther N. Tebbe
3 To all those who have nurtured my growth, passion, and curiosity and who have helped me find and develop my voice
4 ACKNOWLEDGMENTS I thank many people for the guidance and support they have given me during this process. I thank my friends for supporting and encouraging me throughout this process. I thank Dr. Patricia Devine and Will Cox for all of their assistance and guidance in initially collecting this data. I thank Mike Parent for all of his help reading drafts and conducting statistical analyses I thank my committee for being enthusiastic and supportive I thank my advisor, Dr. Bonnie Moradi, for her hard work in helping me develop and complete my thesis and for all of her teaching, mentoring, and inspiration during this time And finally, I thank Nicole for encouraging me having faith in me, and for her unconditional love.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .............. 4 LIST OF TABLES ................................ ................................ ................................ .......................... 6 LIST OF FIGURES ................................ ................................ ................................ ........................ 7 ABSTRACT ................................ ................................ ................................ ................................ .... 8 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ................... 9 Conceptual Definitions and Empirical Findings ................................ ................................ .... 10 Present Study and Hypotheses ................................ ................................ ............................... 15 2 INSTRUMENTS AND METHODS ................................ ................................ ..................... 17 Pa rticipants ................................ ................................ ................................ ............................ 17 Instruments ................................ ................................ ................................ ............................ 17 Anti transgender A ttitudes ................................ ................................ ............................. 17 Anti l esbian, Gay, and Bisexual Attitudes ................................ ................................ ..... 18 Traditional Gender Role Attitudes ................................ ................................ ................. 19 Need f or Closure ................................ ................................ ................................ ............. 20 Social Dominance Orientation ................................ ................................ ....................... 21 Aggression ................................ ................................ ................................ ...................... 21 Social Desirabilit y ................................ ................................ ................................ .......... 22 Procedure ................................ ................................ ................................ ............................... 22 3 RESULTS ................................ ................................ ................................ .............................. 24 Structure of Anti t ransgender Prejudice Operationalized w ith the T ransphobia S cale ......... 24 C onfirmatory F actor A nalysis of the T ransphobia S cale ................................ ...................... 24 Structural Equation Model of Hypothesized Relations ................................ ......................... 25 Measurement Model ................................ ................................ ................................ ....... 26 Structural Model ................................ ................................ ................................ ............. 27 Exploring Suppressor Effect on Aggression ................................ ................................ .. 28 Comparisons of Models for Women and Men ................................ ............................... 28 4 DISCUSSION ................................ ................................ ................................ ........................ 34 LIST OF REFERENCES ................................ ................................ ................................ .............. 40 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ........ 44
6 LIST OF TABLES Table page 3 1 Confirmatory f actor a nalysis l oadings for the t ransphobia s cale ................................ ...... 30 3 2 Descrip tive statistics and zero order c orrelations a mong variables of i nterest ................. 31 3 3 Factor loadings for the a nti transgender prejudice measurement m odel .......................... 32
7 LIST OF FIGURES Figure page 3 1 Structural model of relations among variables of interest for anti transgender p rejudice. 33
8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science ANTI TRANSGENDER PREJUDICE: A STRUCTURAL EQUATION MODEL OF ASSOCIATED CONSTRUCTS By Esther N. Tebbe August 2011 Chair: Bonnie Moradi Major: Counseling Psychology This study examined the unique relations of a number of theoretically relevant constructs with anti transgender prejudice. Specifically, structural equation modeling was used to test the unique relations of anti lesbian, gay, and bisexual (LGB) prejudice, traditional gender role attitudes, need for closure, social dominance orientation, and aggressi on. Social desirability was controlled as a covariate in the model. Analyses of data from 250 undergraduate students indicated that anti LGB prejudice and traditional gender role attitudes each had positive unique relations with anti transgender prejudice beyond the negative association with social desirability. By contrast, need for closure and social dominance orientation were not associated uniquely with anti transgender prejudice. Aggression proneness yielded a unique negative relation which was the res ult of a social desirability suppressor effect. Additional analyses indicated that women reported lower average anti transgender prejudice than men, but the pattern of relations between the predictor variables and anti transgender prejudice did not differ between women and men. A confirmatory factor analysis also supported the unidimensional structure of a measure of anti transgender prejudice.
9 CHAPTER 1 INTRODUCTION Transgender individuals report widespread exposure to prejudice (Clements Nolle, Marx & Katz, 2006; Lombardi, Wilchins, Priesi ng & Malouf, 2001; Stotzer, 2008 ). For instance, a recent national report documenting the experiences of 6,450 transgender and gender nonconforming respondents revealed that 47% reported an adverse job outcome, 29% rep orted police disrespect or harassment, and 15% of students in either K 12 or higher education left their school as a result of severe harassment (Grant et al., 2011). Furthermore, s ome scholars have postulated that anti transgender prejudice may be importa nt in understanding psychological distress and high rates of suicidality among transgender individuals (Clements Nolle, Marx & Katz, 2006; Bockting, Huang, Ding, Robinson & Rosser, 2005). Thus, understanding and reducing anti transgender prejudice has impo rtant public health and social justice implications. However, research on societal attitudes toward transgender people is only in its early stages Specifically, a few studies have sought to develop measures to assess anti transgender prejudice (e.g., Hill & Willoughby, 2005; Nagoshi, Adams, Terrell, Hill, Brzuzy & Nagoshi, 2008) and these measures have been used in subsequent studies to compare the extent to which different groups ( e.g., men, women, parents, undergraduate students) report anti transgender prejudice (e.g. Gerhardstein & Anderson, 2010; Hill, Menvielle, Sica & Johnson, 2010). An important next step in advancing the u nderstanding of anti transgender prejudice is to elucidate the constellation of theoretically relevant constructs that are associated with such prejudice. Identifying key correlates of anti transgender prejudice can locate such prejudice within larger frameworks of prejudicial attitudes and also reveal correlates that may be fruitful targets for psychoeducational and social justice interventions aiming to reduc e anti transgender prejudice and its harm to transgender communities
10 The present study addresses these needs by using structural equation modeling to evaluate the associations of multiple theoretically relevan t constructs with anti transgender prejudice. In particular, on the basis of the proceeding review of the literature, this study evaluates the associations of anti transgender prejudice with sexual orientation and gender role specific attitudes (i.e., ant i lesbian, gay, and bisexual prejudice, traditional gender role attitudes) as well as with broader correlates of prejudicial attitudes and dispositions (i.e., need for closure, social dominance orientation, aggression). In light of the potential role of socially desirable responding in assessing these attitudes, social desirability is also included in the model. Furthermore, based on potential variability in anti transgender prejudice across women and men (e.g., Hill & Willoughby, 2005; Nagoshi et al., 20 08), patterns of hypothesized relations are compared across women and men. As such, the findings of this study can inform theoretical understanding of anti transgender prejudice and point to a cluster of correlates that may be fruitful targets of intervent ions to reduce anti transgender prejudice Conceptual Definitions and Empirical Findings Hill and Willoughby (2005) define d revulsion to gender nonconforming persons or dressers, transgenderists and/or t ( p. 533). Though the term has been used to discuss the construct of anti transgender prejudice in past research (e.g., Hill & Willoughby, Nagoshi, Adams, Terrell, Hill, Brzuzy, & Nagoshi, 2008) for the purposes of this study we use the phrase prejudicial attitudes (rather than fear or phobia) that comprise this construct We are aware of two research programs that have focused on operationalizing anti transgender prejudice (Hill &W illoughby, 2005;
11 Nagoshi et al., 2008). These studies provide the groundwork for empirical examination of theoretically relevant correlates of anti transgender prejudice. One demonstrated correlate of a nti transgender prejudice is anti lesbian and gay pr ejudice. Across studies, a nti transgender prejudice is shown to correlate positively with anti lesbian and gay prejudice ( r s in the .30 to .60 range ; Hill & Willoughby, 2005; Nagoshi et al., 2008). The se correlations may reflect the perception of lesbian, gay, and transgender people as a collective gender tra n s gressive outgroup (e.g., Fassinger & Arseneau, 2007); that is, a group of people that by virtue of their sexual orientation or gender identities and presentations are perceived as violating s ocietal norms of masculinity and femininity. This fusion of groups is also reflected in scientific discourses where sexual orientation and gender identity and presentation issues are frequently discussed together (e.g., Fassinger & Arseneau, 2007; Moradi, Mohr, Worthington, & Fassinger, 200 9 ). Thus, from a conceptual and empirical standpoint, the association between anti transgender prejudice and anti lesbian and gay prejudice may reflect a perceived overlap (or conflation ) of sexual orientation with gender identit ies and presentation s that contradict sex typed gender role prescriptions (Fassinger & Arseneau, 2007 ). While prior research links anti transgender prejudice with anti lesbian and gay prejudice this research has not included anti bisexual prejudi ce when examining the se relations. Large correlations observed between anti bisexual prejudice and anti lesbian and gay prejudice (e.g., Mohr & Rochlen, 1999 ) suggest that anti bisexual prejudice would be correlated with anti transgender prejudice. Moreove r, anti bisexual prejudice may be particularly important to include from a conceptual standpoint Specifically, Mohr and Rochlen (1999) suggested that negative attitudes toward bisexuality arise from the characterization of bisexuality as transitory and il legitimate. Such characterizations have been posited to reflect dichotomous notions of
12 sexual orientation which are rooted in dichotomous notions of sex and gender (Dodge, Reece, & Gebhard, 2008) In other words, a binary view of sex and gender female/woman or male/man is central to a binary view of sexual orientation gay/same sex oriented or heterosexual / other sex oriented. Bisexuality may be a target of prejudice because it challenges thes e binaries (Brewster & Moradi, 2010; Mohr & Rochlen, 1999) In much the same way, transgender identities challenge binary views of gender and sexual orientation. Thus, it seems important to examine the conceptually relevant role of anti bisexual prejudice along with anti lesbian and gay prejudice in relation to anti transgender prejudice. The preceding line of reasoning that anti lesbian, gay, bisexual, and transgender prejudice would be correlated in part because these target groups are perceived to violate gender binaries points to traditional gender role attitudes as another conceptually important correlat e of anti transgender prejudice Indeed, a key correlate of anti gay lesbian and bisexual prejudice is endorsement of traditi onal gender role attitudes ( e.g., Goodman & Moradi, 2008 ; Keiller, 2010 ; Herek, 2002 ) Moreover s cholars have argued that traditional gender role beliefs are the basis of anti transgender prejudice ( e.g., Lombardi, 2009; Nadal, Rivera & Corpus, 2010) Despite this theorized centrality of traditional gender role attitudes to anti transgender prejudice empirical data on the relationship between these two variables has been limited. Specifically, in one study, sexist attitudes which are related to but no t the same as traditional gender role attitudes ( Whitley, 2001 ) were shown to correlate positively and uniquely with anti transgender attitudes above and beyond the association of anti lesbian and gay prejudice (Nagoshi et al., 2008). In the only study th at examined th e link between anti transgender prejudice and traditional gender role attitudes directly, Hill and Willoughby (2005 ) found positive zero order correlations between the two variables ( r = .39 and .65 in two separate samples); but, the extent
13 t o which these correlations reflected shared variance above and beyond the known link of anti lesbian and gay attitudes was not examined. Such an examination is important for clarifying whether traditional gender roles and anti lesbian and gay prejudice exp lain unique aspects of anti transgender prejudice. B eyond the posited roles of sexual orientation and gender role specific attitudes in anti transgender prejudice, other b road er individual difference variables may also be important correlates of anti transgender prejudice. For example, Nagoshi et al. (2008) found that some of the key correlates of anti lesbian and gay prejudice such as right wing authoritarianism and religious fundamentalism are also related to anti transgender prejudice Beyond these previously examined variables, additional literature points to several other individual difference variables that are conceptually important to examine as correlates of anti transgender prejudice: n eed for closure social dominance orientation and aggression proneness These constructs have been linked with anti lesbian, gay, or bisexual prejudice and, as discussed below, are also relevant to anti transgender prejudice from a theoretical standpoint. Need for closur e is defined as a person s desire for order, structure and nonambiguity (Webster & Kruglanski, 1994). More specifically, persons who posse ss a high need for closure are motivated to avoid states of uncertainty (Kruglanski, 1990). As a result, they may be more likely to react negatively to individuals who represent some degree of ambiguity, as this threatens their ability to attain cognitive closure. Thus, individuals with a high need for closure w ould be expected to have negative attitudes toward groups that expand the boundaries (or fall outside) of existing norms. Indeed, need for closure and the associated construct of intolerance for ambiguity have been linked positively with anti lesbian, gay, and bisexual ( LG B ) prejudice (Haslam, Rothschild, & Ernst, 2002 ; Mohr & Rochlen, 1999 ). Just as lesbian, gay, and bisexual
14 individuals challenge norms of heterosexuality and gender traditionality so too do transgender individuals challenge binary notions of sexual orientation and gender identity. Therefore, individual differences on need for closure may be a conceptually important correlate of anti transgender prejudice. Social dominance orientation is another individual difference variable that has bee n linked with anti lesbian and gay prejudice group to dominate and be superior to out Stallworth & Malle, 1994; p. 742). Therefore, social domi nance orientation represents the degree to which individuals desire to maintain social hierarchies As such, researchers have postulated that due to the lower sociopolitical status of women and gay and lesbian individuals in this society, persons high on social dominance orientation are more likely to hold negative attitudes toward these groups ( Pratto et al., 2004; Whitley & Lee, 2000). Indeed, social dominance orientation has been shown to be a ssociated with greater anti lesbian and gay prejudice (Whitley & gisdttir, 2000), and sexist attitudes toward women (Pratto et al., 2004). Similar ly, transgender individuals are members of a stigmatized population that is relegated to a lower social stat us in the United States As such, social dominance orientation may be expected to relate positively to anti transgender prejudice. Finally, aggression proneness is an individual differences variable that has also been considered important to explore as a c orrelate of prejudice toward sexual minority populations. Results have been mixed, however, regarding the link of aggression proneness with prejudice toward LGB groups ( e.g., Adams, Wright & Lohr, 1996; Bernat, Calhoun, Adams & Zeichner, 2001 ; Parrott & Ze ichner, 2005 ) In the only study that examine d the association of aggression proneness with anti transgender prejudice, Nagoshi et al. (2008) found some positive
15 correlations between dimensions of aggression proneness (e.g., physical, verbal) and anti tran sgender prejudice However these correlations were significant among men but not among women and the associations were nonsignificant once anti lesbian and gay prejudice was accounted for. One important caveat to these findings is that self reported aggression proneness is shown to be associated with socially desirable responding ( Harris, 1997; D yer, Bell, McCann & Rauch, 2006; Dyer et al., 2009 ). Therefore, controlling for social desirability seems an important step in clarifying the relationship between aggression proneness and anti transgender prejudice. Another important consideration in ex ploring the correlates of anti transgender prejudice is that some gender differences in level of anti transgender prejudice have been observed, with men reporting more negative attitudes than women (Hill & Willoughby, 2005; Nagoshi et al., 2008). But, it i s not clear whether the pattern of correlations between anti transgender prejudice and other constructs differs by gender group. The parallel body of research on anti LGB attitudes suggests gender differences in levels of anti LGB prejudice ( e.g., Herek, 1 988; Herek, 2002; Mohr & Rochlen, 1999; Whitley & gisdttir, 2000) but not in the relationships between such prejudice and other constructs ( e.g., Goodman & Moradi, 2008). Thus, it seems important to attend to potential gender differences in the present s tudy Present Study and Hypotheses The present study advances research on anti transgender prejudice by evaluating the associations of anti transgender prejudice with sexual orientation and gender role specific att itudes as well as with broader correlates of prejudicial attitudes and dispositions. S pecifically, this study uses structural equation modeling to te st the unique relations of anti LGB prejudice, traditional gender role attitudes, need for closure, social dominance orientation, and aggression proneness with anti transgender prejudice. Each of these variables is hypothesized to
16 be associated positively wi th anti transgender prejudice ; but, prior empirical and conceptual literature suggests a pattern of stronger relations for sexual orientation and gender specific correlates (i.e., anti LGB prejudice, traditional gender role attitudes) than for the broader individual difference variables (i.e., need for closure, social dominance orientation, aggression) In light of the potential role of socially desirable responding in assessing these attitudes, social desirability is also included in the model so that th e hypothesized relations are tested above and beyond the role of such response tendencies. Finally, patterns of hypothesized relations are compared across women and men in order to explore potential variability across these groups (e.g., Hill & Willoughby, 2005; Nagoshi et al., 2008). As a preliminary step to testing the aforementioned hypotheses, we use confirmatory factor analysis (CFA) to evaluate the proposed unidimensional structure of anti transgender prejudice as operationalized by Nagoshi and collea (2008) Transphobia Scale Nagoshi et al. (2008) obtained a unidimensional structure using exploratory factor analysis of data from an undergraduate student sample, but the replicability of this structure has not been examined in an independent sample Thus, the present CFA add s to psychometric data regarding measurement of anti transgender prejudice.
17 CHAPTER 2 INSTRUMENTS AND METHODS Participants Data from 25 0 undergraduate students enrolled in an introductory psychology course were analyzed in thi s study. Participants ranged in age from 18 to 26 ( M = 19.06, SD = 0 .99 Mdn = 19 ). Among the participants, 58% identified as female and 42% identified as male In terms of race or ethnicity, 9 0 % of participants identified as White or Caucasian, 4% as Black or African American, 4% as H ispanic 2% as Asian or Pacific Islander, and less than 1% as Americ an Indian. Finally, 99% of participants identified a s straight or heterosexual, less than 1% identified as questioning, and les s than 1% as bisexual Instruments Anti t ransgender a ttitudes The nine item Transphobia Scale (TS; Nagoshi et al. 2008) measure s prejudicial attitudes towards individuals who present differently from socially expected gender roles and identit ies Items are rated on a 7 point scale ranging from 1 = completely disagree to 7 = completely agree A sample item is I believe that a person can never change their gender I tem ratings were averaged, with higher scores indicating more prejudicial attitudes. In terms of reliability for responses to TS items was .82 in Nagoshi (2008) sample of undergraduate students. Validity has been demonstrated with undergraduate student samples through the independence of TS scor es from instrumentality and expressiveness and through positive correlations with religious fundamentalism and right wing authoritarianism (Nagoshi et al., 2008) for responses to TS items was .84.
18 Anti l esbian, G ay, and B i sexual A ttitudes This construct wa s measured using the Attitudes t oward Lesbians and Gay Men Scale (ATLG; Herek, 1988) and the Attitudes r egarding Bisexuality Scale (ARBS; Mohr & Rochlen, 1999). The Attitudes t oward Lesbians and Gay Men (ATLG; Herek, 1988 ) scale is a commonly used measure of anti lesbi an and gay prejudice. The ATLG has 20 items: 1 0 items comprise the Attitudes t oward Lesbians (ATL) subscale, and 10 items comprise the Attitudes t oward Gay men (ATG) subscale. Items are rated on a 9 point sca le ranging from 1 = strongly disagree to 9 = strongly agree A sample item is were reverse scored, and item ratings were averaged, with h igher scores reflecting more prejudicial attitudes. In terms of reliability, .77 for ATL items and .89 for ATG items in a sample of college student s (Herek, 1988) Validity of ATLG scores has been demonstrated through positive correlations with theoretically related constructs such as dogmatism, sexism, religious fundamentalism, and negative correlations with contact with lesbian and gay persons (Herek, 1988) In this st for responses to ATL items and .96 for responses to ATG items The Attitudes r egardi ng Bisexuality Scale Female/Male Version (ARBS FM ; Mohr & Rochlen, 1999) is a measure of attitudes toward bisexual women and men on two dimensions: tolerance and stability. Tolerance reflects views of the legitimacy and stability of bisexuality as an identity label and of the emotional and romantic stability of bi sexual persons themselves. Mohr and Rochlen (1999) also developed the ARBS F and ARBS M two independent measure s of attitudes toward bisexual women and attitudes toward bisexual men though these measures are comprised of a greater number of items, parall eling each other perfectly except in gender specified in each item. Therefore, as we are interested in examining
19 the relation of anti bisexual prejudice as a whole to anti transgender prejudice we used their shorter combined form, the ARBS FM to examine a nti bisexual prejudice as a whole. The ARBS FM has 18 items : 10 items assess stability and 8 items assess tolerance. Items are rated on a 5 point scale ranging from 1 = strongly disagree to 5 = strongly agree bisexuals are afraid to commit to one lifestyle ratings are averaged to produce subscale scores, with higher scores indicative of more negative on the Tolerance subscale, and .91 for responses to items on the Stability subscale in a sample of heterosexual undergraduate students (Mohr & Rochlen, 1999) In nonstudent samples of lesbian, gay, and heterosexual participants and in undergraduate sample s of mostly or exclusively heterosexual participants, validity of ARBS FM scores has been demonstrated through positive correlations with theoretically related constructs such as anti lesbian and gay prejudice and need for cognitive closure (Mohr & Rochlen, 1999) for responses to Stability items and .91 for responses to Tolerance items Traditio nal Gender Role A ttitudes T his construct was measured using the Attitudes Toward Women Scale (AWS; Spence & Helmr eich, 1978) and the Gender Role Beliefs Scale (GRBS; Kerr & Holden, 1999). T he Attitudes t oward Women Scale (AWS; Spence & Helmreich, 1978) is a short form measure of attitudes toward appropriate role behaviors for women relative to men The AWS consists of 15 items rated on a 4 point scale ranging from 0 = a gree strongly to 3 = d isagree strongly A sample Women earning as much as their dates should bear equally the expense when they go out together Appropriate items were reverse scored and item ratings were averaged. F or interpretive clarity we coded AWS items such that higher scores indicate more traditional gender role attitudes for
20 response s to AW S items across s amples (Spence & Helmreich, 1978; Spence & Hahn, 1997) Evidence for validity has been demonstrated through positive correlations of AWS scores with endorsement of feminine gender norms ( Liss, Erchull & Ramsey 2010) In the present study, was 86 for responses to AWS items The Gender Role Belief s Scale (GBRS; Kerr & Holden, 1996) is a measure of beliefs about appropriate behavior s for women and men The GRBS has 20 items rated on a 7 point scale ranging from 1 = s trongly agree to 7 = s trongly disagree more to see a woman who is pushy than a man who is pushy were reverse scored, and item ratings were averaged F or interpretive clarity, in this study item ratings were coded such that hi gher scores indicate more traditional gender role beliefs. In terms of reliability, for response s to GRBS items in samples of non college women and college women and men (Kerr & Holden, 1996). With regards to validity, Kerr and Hol den demonstrated theoretically consistent differences in GRBS scores between college students, women in feminist organizations, and women in a not specifically feminist organization ( i.e., female volunteers), with feminist women reporting lower adherence t o traditional gender role beliefs than the other two groups for response s to GRBS items in the present study was 87 Need f or C losure The Need for Closure Scale (NFCS; Webster & Kruglanski, 1994) is a measure of 6 point scale ranging from 1 = strongly disagree to 6 = strongly agree like to be with people who scored and item ratings were averaged such that higher scores indicate greater need for cognitive alpha was .84 in a sample of undergraduate st udents
21 (Webster & Kruglanski, 1994). Regarding validity, NFCS scores were correlated positively with such constructs as authoritarianism, intolerance for ambiguity, dogmatism, cognitive complexity and impulsivity in samples of undergraduate students and ad ult library patrons (Webster & was .83. Social D ominance O rientation The Social Dominance Orientation scale (SDO; Pratto et al., 1994) is a commonly used measure of group is perceived to be superior to out groups The SDO is comprised of 16 items t hat are rated on a 7 point scale ranging from 1 = very negative to 7 = very positive A sample item is of people are inferior to other groups Appro priate items were reverse scored and item responses were averaged with h igher scores reflect ing greater social dominance orientation. In terms of reliability, alpha for responses to SDO items was .80 in a sample of undergraduate students (Pratto et al., 1994) V alidity evidence includes negative correlations of SDO scores with empathy, altruism and communality in undergraduate samples (Pratto et al., 1994). pha for response to SDO items in this present sample was 94 Aggression The Aggression Questionnaire ( AQ; Buss & Perry, 1992) is a measure of tendency to aggress along four subtraits of aggression. Accordingly, the 29 AQ items are scored along four subscales: Physical Aggression, Verbal Aggression Anger, and Hostility. Items are rated on a 5 point scale ranging from 1 = extremely uncharacteristic of me to 5 = extremely characteristic of me A sample item is hit back were reverse scored and item ratings were averaged, with higher scores indicating a greater tendency responses to Physical
22 Aggression items, .72 for responses to Verbal Aggression items, .83 for responses to Anger items, and .77 for responses to Hostility items in an undergraduate student sample (Buss & Perry, 1992). Validity of AQ scores has been demonstrated through their positive correlations with other perso nality traits such as assertiveness, competitiveness, and emotionality, and their independence from for responses to Physical Aggression items, .74 for Verbal Aggression items, .85 for Anger items, and .75 for Hostility items. Social D esirability The Marlowe Crowne Social Desirability scale (MCSD; Crowne & Marlowe, 1960) is a widely used desirable responses. The MCSD has 33 items to which participants mark either True or False. thoroughly investigate the qualifications of all the candidates In prior literature, overall scores have been obtained both by either summing or averaging item responses. F or this study, a ppropriate it ems were reverse scored and item responses were averaged. Higher scores indicate more socially desirable responding. for responses to items on the MCSD have ranged from .64 to .78 in s amples of undergraduate students (Loo & Thorpe, 2000). In terms of validity, MCSD scores ha ve been shown to correlate with MMPI Lie scores in a sample of undergraduate psychology students (Crowne & Marlowe, 1960). In this present study, KR 20 (equivalent t for responses to MCSD items was 70 Procedure A total of 251 p articipants were recruited from an introductory psychology course at a large Midwestern university and received extra credit in exchange for their participation Participants logged
23 to choose studies in which to participate This study was described as a study of feelings about a variety of topics and their att itudes toward different groups of people. After selecting to participate in this study participants read a consent form and mark ed their consent if they chose to participate Participants completed the survey instruments (presented in randomized order ) an d upon completing the survey were provided a debriefing statement As this study fo cused on non transgender data were excluded from analyses if they self identified as transgender ( n = 1) re sulting in a final data set of 250 participants
24 CHAPTER 3 RESULTS Structure of Anti t ransgender P rejudice O perationalized w ith the T ransphobia S cale Before proceeding to the tests of the hypotheses, we examined the replicability of the posited unidimensional structure of the Transphobia Scale ( TS ) items using confirmatory factor analysis (CFA) The distribution of TS items met guidelines for univariate normality ( i.e., ; Chou & Bentler, 1995; Kline, 2005) Regarding two participants emerged as multivariate outliers with Mahalanobis distances significan t at p < .001 But none of these participants exhibited patterns of random or consistent extreme responding ( e.g., and iteratively removing their data from the dataset did not change fit indices or p arameter estimates. Thus, we included their data in tests of model fit. This seemed reasonable given that ML estimation is robust to moderate multivariate nonnormality (McDonald & Ho, 2002; Muthn & Kaplan, 1985; Weston & Gore, 2006) particularly when univ ariate normality is achieved (Muthn & Kaplan, 1985), Based on prior evidence and recommendations that a sample size of 200 is adequate for CFA (Quintana & Maxwell, 1999 ; Weston & Gore, 2006; Kline, 2005 ), w e deemed our sample size of 250 to be sufficient for the se analyses C onfirmatory F actor Analysis of the T ransphobia S cale We conducted a CFA using maximum likelihood estimation with AMOS 18.0 (Arbuckle, 2006) to test the unidimensional factor structure of the Transphobia Scale as originally proposed by Nagoshi et al. (2008). Criteria for acceptable absolute fit indices such as the Root Mean Square Error of Approximation (RMSEA) and the Standardized Root Mean Residual (SRMR) have ranged from less conservative, RMSEA, SRMR to more conservative, RM
25 & Bentler, 1999; Quintana & Maxwell, 1999; Weston & Gore, 2006). Criteria for incremental fit indices such as the Comparative Fit Index (CFI) have ranged from less Gore, 2006). Researchers, however, have cautioned against applying strict cut offs to these fit indices at the expense of other theoretical considerations, such as model complexity and sample size (Hu & Bentler, 1999; Quintana & Maxwell, 1999; Weston & Gor e, 2006) The unidimensional model yielded acceptable fit to the TS data, 2 (27, N = 250) = 68.0 7 p < .001, RMSEA = .0 8 90% CI: .0 6 .10, SRMR = .05, CFI = .94. W e are unable to compare the factor loadings from the present analysis to the exploratory factor analysis ( EFA ) factor loadings in the study conducted by Nagoshi et al. because the specific factor loadings were not described in that study However, Nagoshi et al. stated male/female dichoto had a low factor loading of .18 in their sample This finding was replicated in the present study, with item 6 loading significantly at .19 ; all other items had significant factor loadings ranging from .39 to .82 (Table 3 1). Thus, the posi ted unidimensional structure of TS items was generally supported in the present data. Structural Equation Model of Hypothesized Relations Bivariate correlations and descriptive statistics for the res on the measures of interest are presented in Table 3 2 To model the latent variables, we constructed three or more indicators per latent construct ( e.g., Weston & Gore, 2006). For latent constructs assessed by three or more scales or subscales, we used those scale and subscale scor es as indicators; for constructs assessed with one or two measures, we created three to four item parcels (depending on the number of items) from those measures. To create these item parcels, we conducted exploratory factor analyses of data from each measu re, rank ordered the items according to their factor loadings, and assigned items to parcels in countervailing order to maximize the equality of
26 average factor loadings between parcels. Using this procedure, we created three parcels each for anti transgend er prejudice, need for closure, and social desirability ; four parcels for social dominance orientation ; and four parcels for traditional gender role attitudes: two parcels of G ender R ole B elief S cale (GRBS) items and two parcels of A ttitudes toward W omen S cale (AWS) items. Importantly, three items on the GRBS are also on the AWS; to reduce item redundancy and multicollinearity between indicators, we removed the three overlapping items from the AWS prior to exploratory factor analysis and item parcel creatio n. For aggression, we used the four A ggression Q uestionnaire (AQ) subscale scores as indicators. For anti lesbian, gay, and bisexual ( LGB ) prejudice we used the A ttitudes toward L esbians (ATL), Attitudes toward G ay men (ATG) A ttitudes regarding Bisexualit y (ARBS) Stability and A RBS Tolerance subscale scores as indicators. Therefore, the final measurement model was comprised of 2 5 indicators and seven latent variables (anti transgender prejudice, anti LGB prejudice, traditional gender role attitudes, need for closure, social dominance orientation, aggression, and social desirability). To test the hypotheses we conducted latent variable structural equation modeling (SEM), using Amos 18.0 (Arbuckle, 2006) with maximum likelihood estimation Following recomme ndations to use a two step procedure for SEM (Kline, 2005; Muthn & Muthn, 2010; Weston & Gore, 2006), we evaluated the adequacy of our observed indicators in measuring their latent constructs by first testing the fit of the measurement model. Next, we te sted the structural model in order to evaluate the hypothesized relations Finally, we conducted multiple group analysis using a nested model comparison to explore potential model differences across data from women and men Measurement M odel In the measurement model, d ata for all indicators met assumptions of univariate normality ( i.e., ).
27 Regarding multivariate normality, 49 92 and three participa nts emerged as multivariate outliers with Mahalanobis distances significant at p < .001 But none of these participants exhibited patterns of random responding and iteratively removing their data from the dataset did not change fit indices or parameter est imates. Thus, we included the ir data in tests of model fit. This seemed reasonable given that ML estimation is robust to moderate multivariate nonnormality (McDonald & Ho, 2002; Muthn & Kaplan, 1985; Weston & Gore, 2006) particularly when univariate norma lity is achieved (Muthn & Kaplan, 1985), Fit index values for the measurement model were 2 (254, N = 250) = 548.03, p < .001, CFI = .93, RMSEA = .07 (90% CI: .06, .07), and SRMR = .06. As discussed previously, these fit indices fell within the range of criteria for adequate model fit particularly considering that CFIs can be depressed in models with smaller sample sizes ( n more complex ( Hu & Bentler, 1999; Quintana & Maxwell, 1999; Weston & Gore, 2006) Additionally, all indicators loaded positively and significantly on their corresponding latent constructs (Table 3 3 ). Therefore, the constructs represented in this model seemed to be adequately measured by the observed variables. Structural M odel Next w e tested the hypothesized structural model to evaluate the unique relationships of anti LGB prejudice, traditional gender role attitudes, need for closure, social dominance orientation, and aggression with anti transgender prejudice, controlling for social desirability (Figure 3 1) Fit index values for the s tructural model were 2 (254, N = 250) = 548.03, p < .001, CFI = .93, RMSEA = .07 (90% CI: .06, .07), and SRMR = .06. These values fell within the range of criteria for adequate model fit (Quintana & Max well, 1999; Weston & Gore, 2006) This model accounted for 74% of the variance in anti transgender prejudice. As shown in Figure 1, with social desirability controlled, anti LGB prejudice and traditional gender role attitudes were
28 positively and uniquely related to anti transgender prejudice. A ggression was negatively related to anti transgender prejudice. Neither need for closure nor social dominance orientation w ere related uniquely to anti transgender prejudice. Exploring Suppressor Effect on A ggressio n The aggression scales had demonstrated nonsignificant or small positive zero order correlation s with anti transgender prejudice but a ggression yielded significant positive unique relation with anti transgender prejudice in the structural equation model T his pattern suggests the presence of a suppression effect (Horst, 1941; Paulhus, Robins, Trzensniewski & Tracy, 2004). To identify the suppress or variable, we conducted a series of structural equation model analyses, including only aggression and one oth er variable until we replicated the suppression effect ( i.e., between aggression and anti transgender prejudice). This iterative testing of the variables in the model reveal ed social desirability to be the suppression variable, having a classical suppress or effect on aggression. That is, when social desirability was controlled, the unique positive relation between aggression and anti transgender prejudice became small and neg ative. C omparisons of Models for Women and M en Consistent with prior literature ( e.g., Hill & Willoughby, 2005; Nagoshi et al., 2008), d ata from the present study yielded a significant difference in mean scores on the Transphobia Scale in an independent samples T test between women and men, M diff = .77, t (248) = 5.87, p < .001. To explore whether the pattern of hypothesized relations between the predictor variables and anti transgender prejudice differed signific antly between women and men in the hypothesized model we conducted a multiple group comparison analys is using a nested model comparison. We first ran a baseline model without cross group equality constraints on the paths between the predictor variables (a nti LGB prejudice, traditional gender role attitudes need for closure, social
29 dominance orientation aggression, and social desirability ) and anti transgender prejudice. We compared this to a model where the predictor criterion paths were constrained to be equal between women and men To evaluate whether there was a significant difference between women and men in the pattern of predictor criterion relations we examined the difference in chi square values between the baseline mod el and the model with cross group equality constrain ts W e found a nonsignificant chi 2 ( 23 N = 250) = 32 64 p = 09 Therefore, constraining the paths to be equal across groups did not result in a significantly worse fit of the data to the model, indicating that the pattern of relations between predictor variables and anti transgender prejudice was similar for women and men
30 Table 3 1 Confirmatory factor analysis l oadings for the t ransphobia s cale Variable Factor Loading Uniqueness .51** .74 2. I think there is something wrong with a person who says that they are neither a man nor a woman .82** .33 revealed to me that they used to be another gender .74** .45 4. I avoid people on the street whose gender is unclear to me .68** .54 5. When I meet someone, it is important for me to be able to identify them as a man or woman .66** .56 6. I believe that the male/female dichotomy is natural .19* .96 traditional gender roles, e.g., aggressive women or emotional men .39** .85 8. I believe that a person can never change their gender .71** .50 penis defines a person as being a man, a vagina defines a person as being a woman. .69** .52 ** p < .01, p < .05
31 Table 3 2. Descriptive statistics and zero order correlations among variables of i nterest Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1. TS -2. ATL .58** -3. ATG .68** .88** -4. ARBS S .50** .55** .55** -5. ARBS T .68** .84** .89** .67** -6. AWS .50** .62** .65** .34** .60** 7. GRBS .62** .68** .71** .41** .67** .78** -8. NFC .30** .22** .22** .20** .24** .16* .22** -9. SDO .44** .57** .60** .39** .59** .57* .52** .20** -10. AQ P .13* .21** .24** .14* .15* .41** .30** .29** .14* -11. AQ V .11 .07 .10 .10 .06 .11 .14* .10 .03 .46** -12. AQ A .03 .17** .12 .18* .07 .24** .16* .19** .01 .68** .55** -13. AQ H .03 .03 .01 .06 .02 .11 .12 .13* .19** .39** .33** .52** -14. MCSD .23** .12 .12 .16* .11 .10 .16* .15* .08 .27** .23** .34** .26** -M 4.24 3.17 3.58 2.80 2.40 1.80 3.41 3.57 2.65 2.25 2.85 2.26 2.70 1.40 SD 1.10 1.68 2.13 .70 .96 .48 .89 .40 1.09 .83 .80 .80 .70 .20 Possible Range 1 7 1 9 1 9 1 5 1 5 0 3 1 7 1 6 1 7 1 5 1 5 1 5 1 5 1 2 .84 .94 .96 .88 .91 .86 .87 .83 .94 .87 .74 .85 .75 .70 Note: Higher scores indicate higher levels of the construct assessed. TS = Transphobia Scale, ATL = Attitudes Toward Lesbians scale ATG = Attitudes Toward Gay Men scale, ARBS Stability = Attitudes Regarding Bisexuality Scale Stability subscale, ARBS T = Attitud es Regarding Bisexuality Scale Tolerance subscale, AWS = Attitudes Toward Women Scale, GRBS = Gender Role Beliefs Scale, NFC = Need For Closure Scale, SDO = Social Dominance Orientation Scale, AQ P = Aggression Questionnaire Physical Aggression scale, AQ V = Aggression Questionnaire Verbal Aggression scale, AQ H = Aggression Questionnaire Hostility scale, AQ A = Aggression Questionnaire Anger scale, MCSD = Marlowe Crowne Social Desirability Scale. p < .05, ** p < .01.
32 Table 3 3. Factor l oading s fo r the anti transgender prejudice measurement m odel Latent variable and indicator Unstandardized Loading SE Z Standardized Loading Anti transgender attitudes TS Parcel 1 .92 .07 13.20** .74 TS Parcel 2 1.05 .08 13.13** .78 TS Parcel 3 .97 .07 13.86** .75 Anti LGB attitudes ATG 2.04 .10 20.40** .96 ATL 1.52 .08 19.00** .91 ARBS T .90 .05 18.00** .93 ARBS S .44 .04 11.00** .62 Traditional gender role attitudes AWS Parcel 1 .39 .03 13.00** .73 AWS Parcel 2 .23 .02 11.50** .62 GRBS Parcel 1 .83 .05 16.60** .87 GRBS Parcel 2 .83 .05 16.60** .91 Need for closure NFC Parcel 1 .33 .03 11.00** .77 NFC Parcel 2 .38 .03 12.67** .81 NFC Parcel 3 .41 .03 13.67** .85 Social dominance orientation SDO Parcel 1 1.08 .06 18.00** .89 SDO Parcel 2 1.05 .06 17.50** .91 SDO Parcel 3 1.03 .06 17.17** .90 SDO Parcel 4 1.07 .06 17.83** .91 Aggression AQ P .63 .05 12.60** .76 AQ V .48 .05 9.60** .60 AQ H .39 .04 9.75** .90 AQ A .72 .04 18.00** .56 Social desirability MCSD Parcel 1 .16 .02 8.00** .58 MCSD Parcel 2 .16 .02 8.00** .59 MCSD Parcel 3 .14 .02 7.00** .54 Note : TS = Transphobia Scale, ATL = Attitudes Toward Lesbians subscale, ATG = Attitudes Toward Gay Men subscale, ARBS S = Attitudes Regarding Bisexuality Scale Stability subscale, ARBS T = Attitudes Regarding Bisexuality Scale Tolerance subscale, AWS = Attitudes Toward Women scale, GRB S = Gender Role Beliefs Scale, NFC = Need for Closure, SDO = Social Dominance Orientation, AQ P = Aggression Questionnaire Physical Aggression Scale, AQ V = Aggression Questionnaire Verbal Aggression scale, AQ H = Aggression Questionnaire Hostility scale, AQ A = Aggression Questionnaire Anger scale MCSD = Marlowe Crowne Social Desirability scale ** p < .01
33 Figure 3 1. Structural model of relations among variables of interest for anti transgender p rejudice. Dashed lines indicate nonsignificant paths. p < .05, ** p < .0
34 CHAPTER 4 DISCUSSION This study contributes to research on anti transgender prejudice by examining the relations of such prejudice with a number of theoretically relevant constructs. R esults from this study suggest that anti transgender prejudice is more closely related to sexual orientation and gender role specific attitude s ( i.e., anti lesbian, gay, and bisexual prejudice and traditional gender role attitudes) than to bro ader individual difference correlates of prejudicial attitudes ( i.e., need for closure social dominance orientation) By identifying key correlates of anti transgender prejudice, these findings can advance theoretical and empirical understanding of anti t ransgender prejudice and also inform psychoeducational and social justice intervention programs aimed at reducing anti transgender prejudice and its negative impact on transgender individuals and communities. Consistent with our hypotheses and with prior literature (Hill & Willoughby, 2005; Nagoshi et al., 2008), after controlling for social desirability, anti lesbian, gay, and bisexual (LGB) prejudice was positively and significantly linked to anti transgender prejudice Furthermore, among the set of var iables in the model, anti LGB prejudice emerged as the construct with the largest unique relation with anti transgender prejudice This strong association may reflect the perception that sexual orientation and gender identity and presentation minority persons as contradicting sex typed gender role prescriptions (Fassinger & Arseneau, 2007). Therefore, based on the present results anti LGB prejudice may be of particular interest in the understanding of anti transgender prejudice. In designing and implementing intervention could work with their clients to explore more fully their feelings not only toward transgender individuals, but also toward gay men, lesbian women, and bisexual persons, as these attitudes are
35 so strongly tied together Moreover, counselors could work with clients to understand the common thread ( i.e., traditional gender role attitudes) potentially underlying prejudice towards both LGB and transgender populations. Such exploration may also be helpful for counselors in training who are seeking to increase competency in working with transgender clients. It is unsurprising, therefore, that a dherence to traditional gender role attitudes wa s found to relate positively and significantly to anti transgender prejudice. While the unique relation of adherence to traditional gender role beliefs with anti transgender prejudice was less than that of anti LGB attitudes, it was nonetheless one of the stronger correlates of anti transgender prejudice in this model replicating results from prior literature ( e.g., Hill & Willoughby, 2005). As a result, in addition to anti LGB prejudice, counselors engaging in prejudice reduction interventions might aim t o assist clients in further understanding their gender role ideology and how this might impact their views of transgender individuals. Relatedly, counselors in training could be encouraged to explore their beliefs and attitudes regarding gender roles when working with a transgender client. Taken together, the results of this study suggest that addressing anti LGB prejudice and traditional gender roles may be helpful in intervention programs aimed at reducing anti transgender prejudice. Among the broad indiv idual difference variables included in the model, only aggression proneness yielded a significant unique link with anti transgender prejudice; and, with social transgender prejudice w a opposite to the direction hypothesized. Specifically, i n the present study, the zero order correlations of the aggression subscales with anti transgender prejudice revealed that only physical aggression had a significant small positive correlation with a nti transgender prejudice, and the other aggression subscale scores were not correlated significantly with anti transgender
36 prejudice. But, in the structural model: a significant small negative relation emerged between aggression and anti transgender preju dice, after controlling for social desirability. The zero order correlations involving aggression in the present study were consistent with a pattern of small positive zero order correlations observed in prior research (Nagoshi et al., 2008) But the prese nt results revealed that social desirability acted as a suppressor on aggression in t his study, such that before social desirability was entered into the structural equation model, it suppressed criterion irrelevant variance in aggression. With social desi rability controlled for, aggression was able to more efficiently predict variance in anti transgender prejudice (Paulhus, Robins, Trzesniewski & Tracy, 2004). These findings underscore the importance of controlling for social desirability in future studies of the association between self reported aggression proneness and anti transgender prejudice. Furthermore, researchers could employ experimental study designs to measure behavioral manifestations of aggression and its relation to anti transgender prejudic e. Finally, contrary to hypotheses, neither need for closure nor social dominance orientation was related uniquely to anti transgender prejudice. While Mohr and Rochlen (1999) demonstrated a positive zero order correlation between need for closure and a nti bisexual prejudice, the relation of need for closure with anti transgender prejudice has, until now, been account for unique variance in anti transgender prejudice beyond the other variables included in the model Additionally, though social do minance orientation has been linked to various forms of prejudice in past research (Pratto et al., 1994), it also did not correlate uniquely to anti transgender prejudice in this study A similar finding was discovered in a study of the key correlates of anti LGB prejudice (Goodman & Moradi, 2008); after controlling for other variables, social dominance orientation did not relate uniquely to anti LGB prejudice. Taken
37 together, the resu lts of this study point to a nti LGB prejudice and adherence to traditional gender role attitudes as key correlates of anti transgender prejudice Consistent with prior literature ( e.g., Hill & Willoughby, 2005; Nagoshi et al., 2008) results of this study demonstrate differences between women and men in level of anti transgender prejudice, with men reporting more negative attitudes toward transgender persons than women. However, r esults of this study also suggest that the pattern of relationships between anti transgender prejudice and other constructs does not differ between women and men. These findings are similar findings from other studies that have found no differences between women and men in the relationships of anti LGB prejudice with other constructs ( e.g., Goodman & Moradi, 2008). Thus, findings from this study suggest that psychoeducational and social justice programs aimed to reduce anti transgender prejudice can be designed for working with women and men togethe r. Results of this study also provide additional psychometric support for the Transphobia Scale (Nagoshi et al., 2008), by replicating e in a sample independent of the instrument development study. Thus, researchers and practitioners may have greater confidence in using the TS as a unidimensional measure of anti transgender prejudice. However, findings from the present confirmatory factor a nalysis ( CFA ) and from the exploratory factor analysis ( EFA ) conducted by Nagoshi et al. suggest revisiting the wording or use of item 6 in the measure. Despite the contributions of this study to the literature on anti transgender prejudice, i ts findings should be interpre ted in light of a number of limitations. The demographic composition of the present sample forms the boundaries for interpretation of the results. All participants in this study were undergraduate students, the vast majority of whom identified as White. This
38 limits the generalizeability of the findings to the broader population T hus future research should examine the correlates of anti transgender prejudice with samples that represe nt greater diversity in terms of age, race/ethnicity, sexual orientation, level of educational, socioeconomic status, and other backgrounds. Future research should also examine the effect of contextual variables on anti transgender prejudice. The present study focused on indi vidual differences in attitudes and cognitive style. Previous research suggests that contextual and interpersonal factors such as personally knowing a transgender individual (Hill & Willoughby, 2005; Hill, Menvielle, Sica & Johnson, 2010) may also be important additional factors to consider in relation to anti transgender prejudice. Future research, therefore, could explore the concomitant roles of individual differences along with intrapersonal and contextual factors in relation to anti transgender prejudice. The present findings suggest that anti LGB prejudice and traditional gender role attitudes would be key individual difference factors to include in such research. Furthermore, contextual factors related to these individual diffe rence variables might be useful places to start the exploration of contextual correlates of anti transgender prejudice. For example, exposure to ge nder non experiential correlates to explore. Though this study focused on anti transgender prejudice future research could also examine the relation s of key variables to transgender affirming attitudes as well as to transgender affirming and rejecting behaviors. T here is a substantial body of research that has their gender identity or presentation ( e.g., Clements Nolle et al., 2006; Grant et al., 2011; Lombardi et al., 20 01), and research that seeks to understand the basis of these acts is sorely
39 needed. To our knowledge, only one study to date has sought to explore the relation between anti transgender attitudes and behaviors. In an experimental study of anti transgender prejudice transsexual characters, participants transgender attitudes were correlated negatively with their judgments of transsexual characters Specifically, participants rated transsexual characters as less att ractive, in greater need of mental health services and more negatively in general than they did non transsexual characters (Gerhardstein & Anderson, 2010) Future research could examine factors that facilitate or interrupt the translation of anti transgen der prejudice into anti transgender behaviors, such as hate crimes and school bullying. Similarly, research could explore factors that promote transgender affirming behaviors. Finally, though we controlled for socially desirable responding in our model, it is important to note that there was a negative correlation between social desirability and anti transgender prejudice. That is, more social desirability was associated with lower anti transgender prejudice. Future research could continue to focus on wa ys to minimize the potential for social desirab ility such as conducting Int ernet based self report studies where anonymity is greater. Additionally, researchers could measure anti transgender prejudice in other ways that do not rely upon self reported data ( i.e., physiological measures, implicit bias measures, etc) in order to reduce the potential for socially desirable responding. Despite the limitations outlined above, the findings from this study can further our understanding of anti transgender pre judice by locating such prejudice within a broader framework of prejudicial attitudes. By highlight ing those constructs most closely associated with anti transgender prejudice, then findings can also inform intervention efforts aiming to reduce anti transg ender prejudice.
40 LIST OF REFERENCES Adams, H. E., Wright, L. W. & Lohr, B. A. (1996). Is homophobia associated with homosexual arousal? Journal of Abnormal Psychology, 105, 440 445. Bernat, J. A., Calhoun, K. S., Adams, H. E. & Zeichner, A. (2001). Homop hobia and physical aggression toward homosexual and heterosexual individuals. Journal of Abnormal Psychology, 110, 179 187. doi: 10.1037/0021 843X.110.1.179 Bockting, W., Huang, C. Y., Ding, H., Robinson, B. & Rosser, B. R. S. (2005). Are transgender persons at higher risk for HIV than other sexual minorities? A comparison of HIV prevalence and risks. International Journal of Transgenderism, 8 123 131. doi: 10.1300/J485v08n02_11 Brewster, M. L. & Moradi, B. (2010). Perceived experiences of anti bisexual prejudice: Instrument development and evaluation. Journal of Counseling Psychology, 57, 451 468. Buss, A. H. & Perry, M. (1992). The Aggression Questionnaire. Personality Processes and Individual Differences, 63, 452 459. Chou, C. P. & Bentler, P. M. (1995). Estimates and tests in structural equation modeling. In R. H. Hoyle (Ed.), Structural equation modeling: Concepts, issues and applications (pp. 37055). Th ousand Oaks, CA: Sage. Clements Nolle, K., Marx, R., Katz, M. (2006). Attempted suicide among transgender persons: The influence of gender based discrimination and victimization. Journal of Homosexuality, 51 53 69. Crowne, D. P. & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349 354. Dodge, B., Reece, M., & Gebhard, P. H. (2008). Kinsey and beyond: Past, present, and future considerations for research on male bisexuality. Jo urnal of Bisexuality, 8, 175 189. doi:10.1080/15299710802501462 Dyer, K. F., Bell, R., McCann, J. & Rauch, R. (2006). Aggression after traumatic brain injury: Analysing socially desirable responses and the nature of aggressive traits. Brain Injury, 20, 116 3 1173. DOI: 10.1080/02699050601049312 Dyer, K. F., Dorahy, M. J., Hamilton, G., Corry, M., Shannon, M., MacSherry, A., McRobert, G., Elder, & Elhill, B. (2009). Anger, aggression, and self harm in PTSD and complex PTSD. Journal of Clinical Psychology, 65, 1099 1114. DOI: 10.1002/jclp.20619 Fassinger, R. E., & Arseneau, J. R. (2007). 'I'd rather get wet than be under that umbrella': Differentiating the experiences and identities of lesbian, gay, bisexual, and transgender people. In K. J. Bieschke, R. M. Perez, & K. A. DeBord (Eds.), Handbook of counseling and p sychotherapy with lesbian, gay, bisexual, and transgender clients (2nd ed.). (pp. 19 49). Washington, DC US: American Psychological Association.
41 Gerhardstein, appearance and evaluation s of transsexual people. Sex Roles, 62, 361 373. Goodman, M. B & Moradi, B. (2008). Attitudes and behaviors toward lesbians and gay men: Critical correlates and mediated relations. Journal of Counseling Psychology, 55, 371 384. Grant, J. M., Mottet, L. A., Tanis, J., Harrison, J., Herman, J. L. & Keisling, M. (2011). Injustice at every turn: A report of the national transgender discrimination survey. Washington: National Center for Transgender Equality and National Gay and Lesbian Task Force. Harris, J. A. (1997). A further evaluation of The Aggression Questionnaire: Issues of validity and reliability. Behaviour Research & Therapy, 35, 1047 1053. doi: 10.1016/S00 05 7967(97)00064 8 Haslam, N., Rothschild, L. & Ernst, D. (2002). Are essentialist beliefs associated with prejudice? British Journal of Social Psychology, 41, 87 100. gender differences. The Journal of Sex Research, 25, 451 477. States. The J ournal of Sex Research, 39, 264 274. Hill, D. B. & Willoughby, B. L. B. (2005). The development and validation of the Genderism and Transphobia scale. Sex Roles, 53, 531 544. Hill, D. B., Menvielle, E., Sica, K. M. & Alisa, J. (2010). An affirmative interv ention for families with gender variant children: Parental ratings of child mental health and gender. Journal of Marital Therapy, 26, 6 23. Horst, P. (1941). The role of the predictor variables which are independent of the criterion. Social Science Resear ch Council, 48, 431 436. Hu, L. T. & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 8 205 223. Keiller, S. W. (2010). Masculine norms men and lesbian women. Psychology of Men & Masculinity, 11, 38 52. Kerr, P. S. & Holden, R. R. (1996). Development of the Gender Role Beliefs Scale. Journal of Social Behavior & Personality, 11 3 1 6. Kline, R. B. (2005). Principles and practice of structural equation modeling (2 nd ed.). New York: Guilford.
42 Kruglanski, A. W. (1990). Motivations for judging and knowing: Implications for causal attribution. In E. T. Higgins & R. M. Sorrentino (Eds.). The handbook of m otivation and cognition: foundation of social b ehavior (Vol. 2, pp. 333 368). New York: Guilford Press. Liss, M., Erchull, M. J. & Ramsey, L. R. (2010). Empowering or oppressing? Development and exploration of th e Enjoyment of Sexualization Scale. Personality and Social Psychology Bulletin, 37, 55 68. Lombardi, E. L. (2009). Varieties of transgender/transsexual lives and their relationship with transphobia. Journal of Homosexuality, 56 977 992. DOI: 10.1080/0091 8360903275393 Lombardi, E. L., Wilchins, R., Priesing, D. & Malouf, D. (2001). Gender violence: Transgender experiences with violence and discrimination. Journal of Homosexuality, 42, 441 450. doi:10.1300/J082v42n01_05 Loo, R. & Thorpe, K. (2000). Confirm atory factor analyses of the full and short versions of the Marlowe Crowne Social Desirability Scale. The Journal of Social Psychology, 140, 628 635. McDonald, R. P. & Ho, M. H. R. (2002). Principles and practice in reporting structural equation analyses. Psychological Methods, 7, 64 82. doi: 10.1037//1082 989X.7.1.64. Mohr, J. J., & Rochlen, A. B. (1999). Measuring attitudes regarding bisexuality in lesbian, gay male, and heterosexual populations. Journal of Counseling Psychology, 46, 353 369. Moradi B., Mohr, J. J., Worthington, R. L. & Fassinger, R. E. (2009). Counseling psychology research on sexual (orientation) minority issues: Conceptual and methodological challenges and opportunities. Journal of Counseling Psychology, 56, 5 22. Muthn, B. & K aplan, D. (1985). A comparison of some methodologies for the factor analysis of non normal Likert variables. British Journal of Mathematical and Statistical Psychology, 38, 171 189. Muthn, L. K. & Muthn, B. O. (1998 ion. Los Angeles, CA: Muthn & Muthn. Nadal, K., Rivera, D. & Corpus, M. (2010). Sexual orientation and transgender microaggressions; Implications for mental health and counseling. In Derald Wing Sue (Ed.) Microaggressions and marginality: Manifestation, dynamics, and impact (pp. 217 240) Hoboken, NJ: John Wiley & Sons, Inc. Nagoshi, J. L., Adams, K. A., Terrell, H. K., Hill, E. D., Brzuzy, S., & Nagoshi, C. T. (2008). Gender differences in correlates of Homophobia and Transphobia. Sex Roles, 59 521 531 Parrott, D. J. & Zeichner, A. (2005). Effects of sexual prejudice and anger on physical aggression toward gay and heterosexual men. Psychology of Men & Masculinity, 6, 3 17.
43 Paulhus, D. L., Robins, R. W., Trzesniewski, K. H. & Tracy, J. L. (2004). Two replicable suppressor situations in personality research. Multivariate Behavioral Research, 39 301 326. Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance orientation: A personality variable predicting social and politic al attitudes. Journal of Personality and Social Psychology, 67, 741 763. Quintana, S. M. & Maxwell, S. E. (1999). Implications of recent developments in structural equation modeling for counseling psychology. The Counseling Psychologist, 27, 485 527. Spen ce, J. T. & Hahn, E. D. (1997). The Attitudes Toward Women Scale and attitude change in college students. Psychology of Women Quarterly, 21, 17 34. Spence, J. T. & Helmreich, R. L. (1978). Masculinity & femininity: Their psychological dimensions, correlates & antecedents. Austin: University of Texas Press. Stotzer, R. L. (2008). Gender identity and hate crimes: Violence against transgender people in Los Angeles County. Sexuality Research & Social Policy 5 43 52. Webster, D. M. & Kruglanski, A. W. (1994). Individual differences in need for cognitive closure. Journal of Personality and Social Psychology, 67, 1049 1062. Weston, R. & Gore, P. A. (2006). A brief guide to structural equation modeling. The Counseling Psychologist, 34, 719 751. Whitley, B. E. (2001). Gender role variables and attitudes toward homosexuality. Sex Roles, 45, 691 721. DOI: 10.1023/A:1015640318045 Whitley, B. E. & gisdttir, S. (2000). The gender belief system, authoritarianism, social Sex Roles, 42, 947 967. DOI: 10.1023/A:1007026016001 Whitley, B. E. & Lee, S. E. (2000). The relationship of authoritarianism and related constructs to attitudes towar d homosexuality. Journal of Applied Social Psychology, 30, 144 170.
44 BIOGRAPHICAL SKETCH Esther Tebbe was born in Nicosia, Cyprus and moved to Ma dison, W isconsin in 2001 She graduated from James Madison Memorial High School in 2005 and completed her undergraduate work at the University of Wisconsin Madison where she studied psychology and social welfare. After graduating with her B achelor of A rts in 2009, she moved to Gainesville, Florida to begin graduate studies in counseling psychology at the Un iversity of Florida.