Critical Consciousness Moderates the Relationship Between Transphobia and “Bathroom Bill” Voting

Abstract

In recent years, antitransgender legislation that focuses on gender and the use of public restrooms—so-called bathroom bills—has been considered in many states in the United States. The present study was designed to extend research on transphobic attitudes and elucidate links between religious fundamentalism, social dominance orientation, transphobia, and voting for bathroom bills. Further, we examined the moderating influence of critical consciousness on the relationship between transphobic attitudes and voting on a hypothetical bathroom bill. Results of a moderated mediation multinomial logistic regression path analysis using data from a sample of 282 college students (154 women and 128 men) indicated that religious fundamentalism and social dominance orientation were associated with transphobic attitudes. Transphobia was associated with lower likelihood to vote against or abstain from voting on the bill, compared to voting for it. Critical consciousness was also associated with greater likelihood of voting against the bill rather than for it. The interaction between transphobia and critical consciousness was associated with voting against rather than for the bill, such that at higher levels of transphobia, those higher in critical consciousness were more likely to vote against the bill rather than for it. Implications for incorporating critical consciousness into advocacy efforts to promote transgender rights are discussed.

At the time of this writing, in more than a dozen states in the United States legislation commonly referred to as “bathroom bills” has been considered by state governments. Such bills would bar transgender individuals from being able to use the bathroom that aligns with their gender identity by requiring that all persons must use the bathroom that matches their assigned birth sex listed on their birth certificate. Bathroom bills represent a legal, institutional, and sociopolitical form of stigmatization and marginalization of transgender persons, and also risk placing transgender persons in physical danger. A 2015 study of over 27,000 transgender Americans indicated that in the year before the survey 59% of respondents avoided using public bathrooms, 24% had their presence in a bathroom questioned, 12% were harassed while in a bathroom, and 9% were denied access to a bathroom (National Center for Transgender Equality, 2016). Although research has demonstrated that personality factors such orientation to religion and belief in the legitimacy of social hierarchies are associated with transphobic attitudes (e.g., Kosciw, Greytak, Bartkiewicz, Boesen, & Palmer, 2012; Nagoshi et al., 2008; Tebbe & Moradi, 2012), limited research has explored the association between transphobic attitudes and support for legislative action such as bathroom bills. In particular, theoretically driven research is needed to understand what variables may moderate the relationship between transphobic attitudes and support for antitransgender legislation, so as to uncover strategies for promoting the well-being and safety of transgender persons and preventing the adoption of legal and institutional structures that could place transgender persons in danger. Such work would be consistent with liberation psychology (Moane, 2003) and the social justice emphasis of counseling psychology (Goodman et al., 2004; Ivey & Collins, 2003). The goal of the present study was to assess the relationship between religious fundamentalism (RF) and social dominance orientation (SDO), and voting behaviors on a hypothetical bathroom bill, as mediated by transphobia. Further, we sought to investigate the role of critical consciousness in moderating the relationship between transphobia and bathroom bill voting.

Antitransgender legislation can have significant impacts on the lives of transgender persons (Feinberg, 1996). Despite the colloquial name, bathroom bills extend beyond bathrooms to facilities including locker rooms, changing rooms, and more (National Conference of State Legislatures, 2017). Bathroom bills, if passed and enforced, would discourage or ban transgender individuals from using public restrooms or other similar facilities. In the 2015 U.S. Transgender Survey, which collected data from 27,215 transgender individuals in the United States, 32% of respondents reported limiting their food and drink intake in the past year to avoid using a restroom. Further, 8% reported a urinary tract infection or kidney problem, potentially precipitated by holding urine and avoiding using public bathrooms (National Center for Transgender Equality, 2016). In another study, stress related to the use of public restrooms was associated with elevated cortisol levels upon waking among a sample of 65 transgender men (DuBois, Powers, Everett, & Juster, 2017). Among a sample of 93 transgender persons living in Washington, DC, participants report that being denied access to a public restroom negatively impacted their education; resulted in seeking a new job or poor job performance; impacted their health due to holding urine for a prolonged time; and encouraged avoiding public spaces such as malls, restaurants, gyms, and bars (Herman, 2013). Research on social policy and the health of transgender persons suggests that structural social stigmata, such as policies about discrimination or the nonexistence of public school gay-straight alliances is associated with increased risk for suicide among transgender persons (Perez-Brumer, Hatzenbuehler, Oldenburg, & Bockting, 2015). In similar examinations of the impact of antiminority legislation on the well-being of minority persons, research on amendments to limit marriage to other-sex couples (prior to marriage equality being passed by the U.S. Supreme Court) indicted that such legislation negatively impacts the well-being of sexual minority persons (Hatzenbuehler, McLaughlin, Keyes, & Hasin, 2010; Riggle, Wickham, Rostosky, Rothblum, & Balsam, 2016; Rostosky, Riggle, Horne, Denton, & Huellemeier, 2010; Rostosky, Riggle, Horne, & Miller, 2009). Thus, bathroom bill legislation may impact negatively impact the well-being of transgender persons. Attitudes and beliefs that ultimately lead to antitransgender legislation may have foundations in RF, SDO, and transphobia.

Religious Fundamentalism

RF represents an orientation toward religious ideology characterized by absolutistic beliefs about the truth of one’s religious beliefs, that evil forces are active in opposition to this set of truths and therefore must be combated, and that everyone following the one true doctrine has a unique relationship with the divine (Altemeyer & Hunsberger, 1992; Hill, Cohen, Terrell, & Nagoshi, 2010). RF has been an influential aspect of research on prejudice toward members of the sexual minority and gender minority communities (Rowatt et al., 2006; Whitley, 2009). In the U.S. and abroad, many mainstream religious organizations have openly positioned themselves in opposition to sexual and gender minority rights (Endsjø, 2005; Weddle & New, 2011; Wilcox & Robinson, 2010). Although prejudice is not caused solely by RF, RF is associated with prejudicial thinking (Hill et al., 2010).

Positive correlations between RF and negative attitudes toward sexual minority and gender minority persons have been demonstrated in numerous studies (Altemeyer & Hunsberger, 1992; Cunningham & Melton, 2013; Hill et al., 2010; Nagoshi et al., 2008). We expected the same relation to hold here, and that RF would be associated positively with transphobia. RF is also related to another predictor of antisexual minority and antigender minority attitudes, SDO.

Social Dominance Orientation

SDO reflects an individual’s desire for their own in-group to be dominant over competing out-groups. The theory underlying SDO posits that specific cases of discrimination (e.g., racism, classism, transphobia, etc.) are aspects of a more general individual difference characteristic reflecting a tendency to desire group-based hierarchies (Sidanius, Pratto, Van Laar, & Levin, 2004). In particular, social dominance theory emphasizes that group-based oppression is established and maintained via the maintenance of social hierarchies that provide access to resources for those in power and remove such access from those in less powerful groups. Although correlated with RF, SDO does not rely on a religious base and thus represents one approach to assessment of a more secular belief in natural social hierarchy. Because the of emphasis on social power and removal of access to resources among subordinate groups, social dominance theory is clearly tied to oppression of transgender persons through the removal of access to basic resources such as use of public bathroom. Further, SDO is related to but distinct from a similar concept, right-wing authoritarianism, in that right-wing authoritarianism principally drives attitudes in situations in which there is active competition against one’s group (i.e., direct threats to group dominance) whereas SDO drives attitudes in situations in which one’s own group has dominance (e.g., attempts by disempowered groups to attain equality with a dominant group; Duckitt & Sibley, 2010). Endorsement of SDO is also associated with preference for social policies that enforce or encourage hierarchies and inequalities among social groups (Pratto, Sidanius, Stallworth, & Malle, 1994), and thus is relevant to the present study’s focus on political and legal implications of transphobia.

SDO has been associated with negative attitudes toward sexual minority and gender minority persons across several studies (e.g., Tebbe & Moradi, 2012; Whitley & Ægisdóttir, 2000; Whitley & Lee, 2000), and appears to function as a generalized attitude toward group-based hierarchies (Kteily, Ho, & Sidanius, 2011). Examining the relationship between SDO and RF, and transphobia, allows for explication of the link between broad attitudes pertaining to society in general, to transphobia. Such model moves from general attitudes to specific prejudices, consistent with theory and research on the development of specific prejudices (Agnew, Thompson, & Gaines, 2000; Duckitt, Wagner, Du Plessis, & Birum, 2002). This offers an advantage over examining personality associates of transphobia or other prejudices with more specific attitude measures (e.g., beliefs in traditional gender roles), which may simply reflect an overlap in the same attitudes rather than a theoretically driven “drilling down” from broad personality constructs, to specific prejudices, to specific social behaviors. Further, while both RF and SDO may uniquely contribute to the development of antitransgender attitudes, it is also important to understand what variables might buffer the relationships between antitransgender attitudes or voting for legislation such as bathroom bills.

Critical Consciousness

RF and SDO represent important associates, and theoretical predictors, of transphobia. Yet, these variables would be challenging to alter in an intervention. We are not aware of any studies that have attempted to reduce religious fundamentalism. Such an approach would likely be met with resistance and spur on beliefs that social justice efforts are antireligious, and especially anti-Christian, efforts (French, 2015; Jones, 2017; Rush, 2010). Similarly, we are not aware of intervention-based studies that aim to reduce SDO. Indeed, in studies that attempt to mitigate the effects of SDO the aim has been to find ways to circumvent the influence of SDO rather than reduce it (Danso, Sedlovskaya, & Suanda, 2007). Transphobia may be more malleable as a predictor of voting behavior, as intergroup contact is a primary predictor of antitrans attitudes and facilitating such contact has been found to reduce antitrans attitudes (King, Winter, & Webster, 2009; Norton & Herek, 2013; Walch, Sinkkanen, et al., 2012). However, extant work has either primarily assessed affective reactions to transgender persons rather than real or hypothetical voting behavior, or has not explicated the link between affective attitudes and voting behaviors. Thus, we aimed to extend this work to voting behavior and also to investigate how this relationship might be moderated. One variable that may moderate the relationship between antitrans attitudes and voting behaviors, and which is particularly relevant to counseling psychology, may be critical consciousness.

Critical consciousness focuses on awareness of the experiences of marginalized or oppressed groups (Freire, 1972; Shin, Ezeofor, Smith, Welch, & Goodrich, 2016). The concept of critical consciousness extends beyond affect responses toward minority groups and toward understanding the complex sociopolitical nature of the minority experience. As an example, one could have a positive outlook toward gay and lesbian people based on superficial positive exposure to sexual minorities (e.g., through TV or movies) without grasping the level of discrimination or oppression that such persons may face. Critical consciousness, in contrast, developed from liberation psychology and social justice efforts that aim to address minority health and well-being through the examination of social and political structures that may support oppression (Moane, 2003). The development of critical consciousness has been cornerstone of the field of counseling psychology (Ivey & Collins, 2003; Kenny et al., 2007; Kidd & Kral, 2005; Quintana, 2007). Critical consciousness is also an important construct in that it is a relatively clear and operationalizable aspect of liberation psychology and social justice principles, and integration of critical consciousness into quantitative research been aided by the development of critical-consciousness-related measures (e.g., Shin et al., 2016). In prior research, critical consciousness has been associated negatively with heterosexist, classist, and racist attitudes (Shin et al., 2016). However, the concept has not yet been applied in quantitative inquiries to transgender persons or legal issues that are relevant to transgender persons. Given that the legal rights of transgender persons are in a state of flux throughout much of the United States at the time of this writing, critical consciousness offers an important theoretical bridge between the affective feelings toward transgender persons and willingness to engage in political action to support transgender persons.

Critical consciousness may be an important moderator of the relationship between attitudes related to minority groups and actions that would be damaging to the groups. In behavior models, such as the integrated behavior change model and the theory of planned behavior, links between attitudes and behaviors are moderated by cognitions (Ajzen, 2002; Hagger & Chatzisarantis, 2014). However, as limited research exists on the link between attitudes and political action, limited work also exists on whether critical consciousness might moderate the relationship between transphobia and voting for bathroom bills. Even if transphobic attitudes themselves may be resistant to change, capitalizing on the development of critical consciousness may weaken the relationship between transphobic attitudes and political action that would harm the well-being or risk the safety of transgender persons.

The goal of the present study was to extend work on transphobic attitudes to a public health and sociopolitical topic relevant to transgender persons: bathroom bills. Based on prior literature, we had the following hypotheses:

Hypothesis 1: RF (H1a) and SDO (H1b) would be positively associated with transphobia.

Hypothesis 2: Transphobia would be negatively associated with voting against, and abstaining from voting on, bathroom bill legislation relative to voting for it.

Hypothesis 3: Critical consciousness would be positively associated with voting against the bathroom bill, relative to voting for it.

Hypothesis 4: Critical consciousness would moderate the relationship between transphobia and voting, such that higher levels of critical consciousness would be associated with greater likelihood to vote against the bill with this relationship being stronger for persons who are lower in transphobic attitudes.

Method

Participants

Of the 282 participants in the study, 154 identified as women and 128 identified as men (options for transgender identity and a different identity with a write-in option were not used by any participants). Of the original 299 participants from whom we gathered data, we excluded 17 who identified as sexual minorities (gay, lesbian, bisexual, or another sexual minority identity). Although sexual minorities may hold transphobic attitudes, they may also be more likely to hold more positive attitudes toward the entire LGBT community, and in this study comparison of the heterosexual and sexual minority participants yielded some large effect size differences on the variables in the model (e.g., d = 1.47 on voting against the bathroom bill rather than for it). Participants ranged in age from 18 to 54 (M = 20.10, SD = 3.00, Mdn = 19). Regarding race/ethnicity, participants identified as White (70%), Asian or Asian American (10%), Hispanic or Latino/a (8%), Black or African American (6%), multiracial/ethnic (5%), American Indian or Alaska Native (1%), or selected “other” (1%). Participants completed a single-item measure of perceived socioeconomic status compared to others, ranging from 1 (among the worst off) to 100 (among the best off). Participants rated themselves between 10 and 100 (M = 61.67, SD = 17.31, Mdn = 65). Most participants (70%) identified as Christian; 19% identified as atheist or agnostic, and the remaining 11% were dispersed across different available options for religious identity (e.g., Hindu, Muslim, Buddhist).

Measures

Religious fundamentalism.

RF was assessed using the Religious Fundamentalism Scale (RFS; Altemeyer & Hunsberger, 1992), a frequently used measure intended to reflect a rigid orientation to religiosity and spirituality. The RFS contains 12 items (sample item: “God has given humanity a complete, unfailing guide to happiness and salvation, which must be totally followed”) answered on a 9-point scale (1 = completely disagree, 9 = completely agree). Appropriate items are reverse-scored, and higher scores indicate greater RF. Validity of the RFS has been demonstrated through correlations with negative affect, aggression, and negative thoughts about gay men or lesbians (rs = .50, .39, and .61, respectively) among a sample of undergraduate men. Among a community sample of U.S. participants collected online, RFS scores were associated with rule-based information processing (i.e., resolving moral dilemmas through deference to absolutistic rules as opposed to inclusion of context; Young, Willer, & Keltner, 2013). In those same samples, Cronbach’s alpha internal reliability coefficients for the RF were high (α = .91 in both studies).

Social dominance orientation.

SDO was assessed using the measure of the same name (SDO; Pratto et al., 1994), a frequently used measure that reflects an acceptance of inequality among groups. The SDO contains 16 items (sample item: “If certain groups of people stayed in their place, we would have fewer problems”) answered on a 7-point scale (1 = strongly disagree/disapprove, 7 = strongly agree/approve). Half of the items are reverse-scored, and higher scores reflect greater SDO. Prior research with the SDO has demonstrated its validity and reliability. Among a sample of U.S. undergraduates, SDO was correlated with self-reported conservative political ideology (r = .38), right wing authoritarianism (r = .52), and obedience to authority (r = .33; Kugler, Jost, & Noorbaloochi, 2014). Among another sample of U.S. undergraduates, SDO scores were correlated with intergroup anxiety (r = .43), modern prejudice (r = .44), lower endorsement of direct financial assistance to hypothetical refugees (r = −.25), and lower support for empowerment of hypothetical refugees through actions such as job training (r = −.37) (Costello & Hodson, 2011). Among those two samples, Cronbach’s alpha internal reliability coefficients were also high (α = .91 and .92), as is the case with most studies using the SDO.

Transphobia.

Transphobia was assessed using the Transphobia Scale (TS) developed by Nagoshi et al. (2008) to assess affective attitudes toward transgender persons. The Transphobia Scale contains nine items (sample item: “I don’t like it when someone is flirting with me, and I can’t tell if they are a man or a woman”) answered on a 7-point scale (1 = strongly disagree, 7 = strongly agree). Higher scores reflect more negative attitudes toward transgender persons. In prior research, the TS has demonstrated Cronbach’s alpha internal reliability coefficients in the low .80s in samples of undergraduate men and women (Nagoshi et al., 2008; Tebbe, Moradi, & Ege, 2014). In a sample of undergraduate men and women, the TS demonstrated correlations with measures of negative attitudes toward gay men (r = .68), lesbians (r = .58), and bisexuals (rs = .50 and .68); and SDO (r = .44; Tebbe et al., 2014).

Critical consciousness.

Critical consciousness was assessed using a modification of items on the Contemporary Critical Consciousness Measure (Shin et al., 2016), which was developed to assess racism, classism, and heterosexism. In this study, the items pertaining to sexual orientation minorities were reworded to apply to transgender individuals and only this subset of items was administered to participants. This subscale contains six items (sample item: “I believe the U.S. society generally promotes hatred of transgender individuals”) answered on a 7-point scale (1 = strongly disagree, 7 = strongly agree). Higher scores indicating greater of critical consciousness. In the developmental study, the original sexual orientation items demonstrated a Cronbach’s alpha internal reliability coefficient of .85 among a sample of MTurk participants. In the same sample, scores on the original items were associated negatively with measures of racism (r = −.47), classism (r = −.39), and homophobia (r = −.79; Shin et al., 2016).

“Bathroom bill” voting.

We assessed “bathroom bill” voting using a single item: “Imagine that an election is coming up tomorrow. You have the chance to vote on several state laws. How would you vote on a law that makes sure people have to use the bathroom corresponding to the gender that is on their original birth certificate? You have the option to vote for this, vote against it, or not vote on it at all.” Because individuals have the ability to refrain from voting on issues, three response options were available. These three response options were as follows: vote for, vote against, and I would not vote on this.

Procedure

The present study was approved by the institutional review board at the first author’s institution. Data were collected at a large public university in the Southern United States, in a relatively conservative and predominantly Christian area of the state (City-data.com, 2010). All participants were recruited from undergraduate introduction to psychology courses and consented to take part in the study. Participants were able to see the opportunity to participate in the study via an online research participation portal, and could elect to participate from a range of available studies. Studies on the online system were numbered and not named, meaning that participants did not know the study was about attitudes toward transgender persons until viewing the consent form (though they could withdraw from participation after viewing the consent form if they so chose). Students who elected to participate in the present study completed the measures, hosted online in Qualtrics, in randomized order. Participants were compensated with credit toward a research participation component of coursework; compensation was automated via a link between Qualtrics and the study sign-up software.

Results

A total of 314 participants began the survey. No participants were missing data on the consent form viewing variable, meaning that no participants elected to de-enroll from the study when they read that it was about attitudes toward transgender persons. Fifteen participants were deleted from the data set due to high levels of missing data; all of these participants were missing at least one complete measure, with most (nine of 15) missing all data on more than one measure. An additional 17 participants who reported their sexual orientation as other than heterosexual were also removed from the data set (detailed above). These 32 participants are not included in any results in the present study. Within the data used in this study, a total of 34 of 12408 data points were missing, representing an item level missing data rate of 0.27%. Given this low level of missing data, available item analysis (Parent, 2013) was used to calculate means and Cronbach’s alpha coefficients. Data used in the analyses met guidelines for univariate normality (highest skew = −0.31; highest kurtosis = −0.90). Examination of DeCarlo’s implementation of Mardia’s coefficient of multivariate skew indicated that no cases were identified as multivariate outliers (DeCarlo, 1997). Descriptive statistics for the variables used in the analyses, and available item analysis Cronbach’s alpha for responses to items on the measures, are presented in Table 1. In total, 42% of participants reported that they would vote for a “bathroom bill” that required people to use the bathroom corresponding to their birth certificate, 29% said they would vote against it, and 29% said they would not vote on it.

Table 1.

Correlations and Descriptive Statistics

Variable	1	2	3	4	5	6	7	M	SD	Possible range	Alpha
1. RFS	—	.11	.39*	−.27*	.31*	−.01	−.32*	5.07	1.76	1–9	.92
2. SDO		—	.41*	−.46*	.19*	.02	−.22*	2.91	1.07	1–7	.92
3. TS			—	−.46*	.56*	−.12	−.49*	4.05	1.23	1–7	.89
4. CCCM				—	−.44*	.06	.41*	4.82	1.27	1–7	.89
5. Vote for					—	−.54*	−.55*	.42	.49	0–1
6. Abstain						—	−.41*	.29	.45	0–1
7. Vote against							—	.29	.46	0–1

Note. TS = Transphobia Scale; RFS = Religious Fundamentalism Scale; SDO = social dominance orientation; CCCM = Contemporary Critical Consciousness Measure.

^*p < .01.

Analyses were conducted in Mplus version 7.2 (Muthén & Muthén, 2012). The model was defined as a moderated mediation multinomial logistic regression model, with choosing to vote for the “bathroom bill” as the reference group for the dependent variable. Variables used in the analysis were centered prior to data transfer for analysis and the interaction between transphobia and critical consciousness was defined in Mplus. Transphobia was regressed onto RF and SDO and voting was regressed onto transphobia, critical consciousness, and the interaction between transphobia and critical consciousness. Participant gender was used as a control variable. We tested both partial (i.e., including direct relationships from SDO and RF to voting) and full (i.e., excluding those paths) mediation models.

Multinomial logistic regressions use categorical dependent variables and do not produce fit statistics that are present in path analyses or structural equation models that use continuous dependent variables. Rather, models can be evaluated based on path parameter values, and competing models can be compared using the Bayesian information criterion (BIC). For the partial mediation model, BIC was 4247.97. For the full mediation model, BIC was 4236.67. The lower BIC value, by 11.30 points, indicates strong support for the superiority of the full mediation model (Kass & Raftery, 1995). Practically, in the full mediation model, only one of the four added paths was significant (from RF to voting against the bill rather than for it (B = −0.23, SE = 0.10, p = .024; odds ratio [OR] = 0.720, 95% CI [0.559, 0.928]).

We proceeded with analyses using the full mediation model. Results of the analysis are presented in Figure 1. For paths that statistically predict the dichotomous voting variables, ORs are presented as well as standardized path coefficients. Recommendations about interpreting the size of ORs vary, but ORs over 3.0 may be considered to represent strong relationships (Haddock, Rindskopf, & Shadish, 1998).

Figure 1.

Path analysis model. Values reported are unstandardized regression weights and standard errors. For the dichotomous voting variables, OR (odds ratios) with 95% confidence intervals are reported. The reference group for the voting variable was choosing voting in favor of the bathroom bill. Values attached to the lower left of boxes represent the regressions for the variables on the gender control variable. Gender was coded as 0 = women, 1 = men, and thus higher values for the covariates indicate higher scores on the corresponding variable for men compared with women. * p < .05. ** p < .01. *** p < .001.

RF and SDO were both positively associated with transphobia, consistent with hypothesis 1a and 1b. The 95% CI for the odds ratio for the relationship between transphobia and voting against the bathroom bill was below 1, meaning that, consistent with Hypothesis 2a, transphobia was associated with 5.13 times greater likelihood of voting for the bathroom bill rather than voting against it (i.e., for each standardized unit decrease in transphobia, participants were 5.13 times more likely to vote against the bathroom bill rather than for it, corresponding to a strong relationship). Transphobia was also associated with abstaining on the bathroom bill vote, again with an odds ration under 1. Thus, consistent with Hypothesis 2b, endorsement of transphobia was associated with 3.00 times greater odds of abstaining from the vote rather than voting for the bathroom bill (i.e., for each standardized unit decrease in transphobia, participants were 3.00 times more likely to abstain from a vote than for vote the bathroom bill, corresponding to a strong relationship). These result supported Hypothesis 2.

Critical consciousness was associated with voting against the bathroom bill, consistent with Hypothesis 3. For each unit increase in critical consciousness odds of voting against the bill, increased 2.00 times, compared to voting for it, corresponding to a moderate relationship. Critical consciousness was not related to voting to abstain on the bathroom bill, compared to voting for it.

The interaction between transphobia and critical consciousness was associated significantly with voting against the bathroom bill rather than for it, consistent with hypothesis 4. The interaction corresponded to a moderate relationship and is displayed in Figure 2. Higher values on the vertical axis of the figure indicate greater probability of voting against the bathroom bill, compared with voting for it. At low levels of transphobia, those high and low in critical consciousness were more likely to vote against, not for, the bill. As transphobia increased among the sample, those high in critical consciousness experienced a weaker relationship between transphobia and voting. At higher levels of transphobia, though voting against the bill declined, it did so at a lower rate for those at levels of critical consciousness. The interaction was not associated with abstaining from voting on the bill, compared to voting for it.

Figure 2.

Simple slopes interaction between transphobia and critical consciousness and the probability of voting against a bathroom bill.

Discussion

The goal of the present study was to extend research on attitudes toward transgender persons to a specific sociopolitical issue relevant to transgender persons in the United States. Specifically, we examined associations among RF, SDO, transphobia, and critical consciousness, and voting for or against a “bathroom bill” similar to actual legislation being discussed in the United States at the time of this writing that would force individuals to use restrooms consistent with the sex identified on their birth certificate. The findings of the present study elucidate important relations among variables associated with antitransgender prejudice and potential for voting for, or against, legal changes that would negatively impact the welfare and safety of transgender persons. Consistent with past research (Nagoshi et al., 2008; Tebbe et al., 2014), RF and SDO were associated positively with transphobia. Transphobia was associated positively with voting against and abstaining from voting on a “bathroom bill” (compared to voting for it). Critical consciousness moderated the relationship between transphobia and voting against the bill rather than for it.

Prior findings of links between RF and SDO with transphobia were replicated via high-medium correlations between RF and SDO, and transphobia. Further, the bivariate correlations between RF, SDO, transphobia, and voting for or against a “bathroom bill” were significant and in the expected directions. Thus, the broad assessments of prejudice-related attitudes (RF and SDO) had significant bivariate relations with hypothetical voting behaviors. However, consistent with research on prejudice (Agnew et al., 2000; Duckitt et al., 2002), RF and SDO were associated with transphobia and transphobia fully mediated the association between RF/SDO and hypothetical voting behaviors. when all the variables included in the present study were assessed in a multinomial logistic regression. RF and SDO did not demonstrate clear direct relations with voting either against or abstaining on the “bathroom bill” compared with voting for it. Thus, while RF and SDO may be important correlates of transphobia reflecting broad social attitudes, they do not appear to uniquely add to antitransgender voting attitudes beyond specific antitransgender attitudes.

Critical consciousness moderated the relationship between transphobic attitudes and voting for the bathroom bill. When transphobia was lower, individuals high and low on critical consciousness were quite likely to vote against rather than for the bathroom bill. As transphobia increased, those at higher and lower levels of critical consciousness diverged, with the relationship between transphobia and voting against rather than for a bathroom bill having a sharper negative slope for those lower in critical consciousness. These results speak to the importance of counteracting negative affective attitudes toward transgender persons and the development of critical consciousness about the situations faced by transgender persons in the United States This result also supports the tenets of liberation psychology and counseling psychology’s emphasis on social justice (Goodman et al., 2004; Ivey & Collins, 2003; Moane, 2003) that a lack of negative attitudes may not be sufficient to promote active engagement in the promotion of sociopolitical action that would benefit minority groups (in the present sample, transgender persons). Rather, interventions and advocacy designed to foster the growth of critical consciousness are vital to promoting positive institutional change (Dean et al., 2000).

The results of the present study have implications for research and advocacy. It is important to extend research beyond attitudes toward minorities and toward sociopolitical behaviors that can threaten or provide basic freedoms and equal access to minority persons. Because attitudes toward minorities, particularly sexual and gender minority persons, can be strong influenced by affective reactions (Terrizzi, Shook, & Ventis, 2010), it is important to understand how affect is linked to sociopolitical behaviors. Further research might explore the relationships between general affective attitudes toward sexual orientation minorities and attitudes toward specific sociopolitical issues, such as coverage of gender transitioning in military health care, same-sex adoption, or antidiscrimination laws. At the same time, critical consciousness appears to be a crucial and malleable factor related to political support for transgender rights.

We found critical consciousness to moderate the relationship between transphobic attitudes and hypothetical voting behaviors in the expected direction. That is, when transphobia was higher, higher levels of critical consciousness were associated with increased likelihood to vote against the bathroom bill rather than for it. Thus, critical consciousness raising may be important to moving voters toward engaging in trans-affirming political actions. As with all attitudes toward minorities, contact with transgender persons is a strong predictor of more positive attitudes toward transgender people and more willingness to support trans rights. However, given that transgender identity does not have as high an occurrence as, for example, lesbian or gay identity, reliance on transgender people to come out is an inefficient means by which to promote contact and intergroup awareness. For example, in one study of undergraduates enrolled in a human sexuality class, only three of 42 participants reported having a transgender friend (Walch, Sinkkanen, et al., 2012). In the same study, student exposure to a panel of transgender individuals was associated with decreased transphobia, more efficiently than when a lecture class was delivered before the panel. However, in that study participants were students in a human sexuality class and thus may have self-selected into a sample more open to sexuality diversity. In another study, door-to-door canvassers, both transgender and cisgender, were successful in reducing transphobia among a community sample. This door-to-door intervention increased support for a nondiscrimination law, with effects enduing at 3-month follow-up (Broockman & Kalla, 2016).

Despite these positive findings of the effectiveness of critical consciousness raising, the scopes of the two mentioned studies present challenges to the implementation of critical consciousness raising. In the first example, participants were students enrolled in a human sexuality class and, generally, may well have been uniquely open to learning more about human sexuality. The second study offers more promising results, but door-to-door canvassing is costly and time-consuming. Further, the intervention was conducted in Miami and it is unclear what challenges such an intervention might face in areas where canvassers may be met with more hostility. Continued work might explore the effectiveness of support for grassroots consciousness raising within communities be establishing partnerships with critical community members and organizations (Altman, 1995; Viswanathan et al., 2004). Alternatively, means could be identified through which to harness social media effectively in awareness campaigns (Guo & Saxton, 2014; Obar, Zube, & Lampe, 2012) so as to reach a wider audience in a most cost-effective manner while decreasing the potential for hostility and maintaining effectiveness.

The results of the present study must be interpreted in light of its limitations. First, we sampled college students. Aside from the standard caveats about college student sampling, this imposes some limitations on the practical implications of the results. The sample was composed mostly of young persons who historically have low voting rates (McDonald, 2016). Further, younger persons tend to be more politically liberal (DeSilver, 2014), even in the Southern United States (Saad, 2017), and specifically to be more likely than older persons to express neutral or affirming attitudes toward transgender persons (Moore, 2015). Thus, while the present study extends extant findings on attitudes toward transgender persons among college samples (Tebbe et al., 2014; Walch, Ngamake, Francisco, Stitt, & Shingler, 2012; Woodford, Silverschanz, Swank, Scherrer, & Raiz, 2012), the present results could be usefully extended by intentionally sampling individuals who typically have high voter turnout rates, especially in “off-year” elections where bills such as these may be proposed. Second, it is possible that some participants were not as aware of transgender issues as may have been optimal for this study. Though transgender bathroom bills were in news articles at the time of data collection, some participants may not have been aware of these news stories and some may not have understood what was meant by “original birth certificate.” Third, we assessed gender as a binary variable (for participant), rather than as more complex interplays of gender role ideologies (Moradi & Parent, 2013). We also assessed attitudes toward “transgender” persons broadly, and future experimental research may examine specific effects of the direction of gender transition on bathroom bill attitudes (e.g., fears about “men in girls’ rooms”) or the effects of gender presentation (e.g., transgender individuals who easily “pass” vs. those who do not) on attitudes toward bathroom bills. Finally, the present data were cross-sectional and the voting variable was hypothetical, imposing limitations on the ability to interpret causality of the present findings and the concreteness of the voting behavior variable.

In sum, the present study extends current work on attitudes toward transgender persons to understanding voting behaviors related to a relevant topic for transgender persons in the United States The present study has the potential to inform further research that extends attitude research into the domain of sociopolitical action, and also supports the value of the work being done by counseling psychologists and others to raise critical consciousness on the political and legal marginalization and oppression of transgender persons.

Public Significance Statement.

The present study demonstrates relations between religious fundamentalism, social dominance orientation, transphobia, and critical consciousness and hypothetical voting for on a “bathroom bill” that would force individuals to use bathrooms corresponding to the sex on their birth certificates. The study highlights associations important to the health of transgender persons in terms of restriction of access to public bathrooms.