Abstract
Prior work suggests that some cisgender people do not consider gender identity when thinking about what being transgender means, and that exposure to a definition of transgender can reduce negative attitudes toward transgender people. In two studies, we sought to integrate these lines of research by examining whether anti-transgender attitudes are associated with how cisgender people define the term transgender. In Study 1, 293 participants (132 female and 157 male; four participants did not report their sex/gender) recruited via Mechanical Turk were randomly assigned to one of two conditions. Participants in one condition first read a definition of “transgender” that referred to gender identity and then reported their attitudes toward transgender people. Participants in the other condition did not read the definition and instead simply reported their attitudes toward transgender people. Contrary to predictions, exposure (vs. lack thereof) to the definition did not affect attitudes. However, participants who identified gender identity as a central component of the definition of transgender on a manipulation check endorsed more positive attitudes. In Study 2, 295 cisgender participants (165 women and 130 men; recruited via Mechanical Turk) wrote down their own definitions of transgender. Participants who spontaneously included (vs. did not include) gender identity in their definitions reported distinctively positive attitudes. Observed effects held over and above individual differences in cognitive style, gender identification, contact with transgender people, and participant gender. How cisgender people think about gender identity may thus index their attitudes toward transgender people.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
As the proportion of the US population identifying as lesbian, gay, bisexual, or transgender continues to increase (Jones, 2022), so does societal conversation about gender identity and its relationship to birth-assigned sex (Parker et al., 2022). Sex refers to biological and physiological systems involving the X and Y chromosomes, sex hormones, and sexual (including genital and reproductive) differentiation (Hyde et al., 2019). By contrast, gender refers to socially constructed expectations and roles for women and men (Hyde et al., 2019) and includes both external facets such as gender presentation and internal facets such as gender identity (Morgenroth & Ryan, 2021). Sex and gender are distinct, albeit often intertwined, constructs. This distinction is exemplified by transgender people, whose gender identity and/or expression do not coincide with the sex they were assigned at birth. Although the noncoincidence of sex and gender is pivotal to the definition of transgender, there is some variation in other dimensions that may be included in this definition. For instance, some institutional definitions focus solely on gender identity (e.g., GLAAD, n.d.), whereas others also refer to gender expression and gendered behavior (e.g., APA, 2015, n.d.). Similar lack of consensus is evident among laypeople: Whereas some cisgender people (i.e., people whose gender identity and assigned sex coincide) define “transgender” as denoting a specific gender identity, others’ definitions instead rely on gender expression or assigned sex (Anderson, 2022; Buck, 2016). The goal of the present work was to examine how the content of cisgender people’s definitions of “transgender” relates to their attitudes toward transgender people, and whether this relationship holds over and above associations with other established predictors of these attitudes.
Much of existing research on attitudes toward transgender people has focused on identifying individual-difference predictors of transphobia. This research has revealed that stronger anti-transgender attitudes are expressed by heterosexual cisgender people who endorse more socially conservative views, as indexed by variables such as political conservatism, religiosity and religious fundamentalism, and right-wing authoritarianism (Buck, 2016; Hill & Willoughby, 2005; Nagoshi et al., 2008; Norton & Herek, 2013; Tee & Hegarty, 2006). However, over and above social conservatism, how cisgender people think about the social category of gender appears to be a more specific predictor of anti-transgender bias. In particular, cisgender people who endorse gender essentialism (i.e., the belief that gender is binary, biological, and innate) tend to express greater anti-transgender bias (e.g., Rad et al., 2019). Moreover, exposure to information about an anti-essentialist interpretation of gender increases cisgender people’s support for transgender rights (Wilton et al., 2019), whereas exposure to a biological interpretation of gender heightens anti-transgender prejudice (Ching & Xu, 2018). Cisgender people’s anti-transgender attitudes can therefore be rooted in the perception that transgender people threaten conventional gender categories, insofar as transgender people are perceived as not conforming to the binary view of gender that is central to gender essentialist beliefs (Ching & Xu, 2018; Garelick et al., 2017).
Given that beliefs about binary gender help explain anti-transgender views (e.g., Garelick et al., 2017; Prusaczyk & Hodson, 2020), attitudes toward transgender people may also be associated with how cisgender people think what the term transgender means and how this term is situated in the overall concept of gender. Conceptual models of gender have suggested that “gender” involves at least five dimensions (Hyde et al., 2019; Tate et al., 2014), including birth-assigned sex category, current gender identity (i.e., individuals’ internal experience of their own gender), gender roles (i.e., the set of social expectations associated with one’s gender), gender presentation or expression (i.e., the ways in which one presents one’s gender interpersonally, including via clothing and name), and gendered evaluation of the social world (including gender-based attitudes and behaviors). Research has underscored the importance of gender identity over and above birth-assigned sex in determining people’s gendered life experiences (for a review, see Hyde et al., 2019). Accordingly, whether cisgender people consider gender identity when they think about what “transgender” means may be a key correlate of their attitudes toward transgender people. For example, going beyond biology or genetics (e.g., birth-assigned sex) in one’s understanding of gender indicates lesser endorsement of gender essentialism (for a review, see Dar-Nimrod & Heine, 2011) and thus potentially more positive attitudes toward transgender people as well.
Some prior research is consistent with this reasoning. Specifically, Buck (2016) found that when asked to define “transgender” in their own words, some cisgender respondents evoked notions of birth-assigned sex and gender presentation while omitting mentions of gender identity. For instance, some participants defined “transgender” as referring to a person who has changed their sex or behaves in a way that does not match their assigned sex. Cisgender people who did not mention gender identity when defining transgender also expressed more anti-transgender attitudes (Buck, 2016; for a recent replication, see Anderson, 2022; see also Adams et al., 2016, who found that transphobic attitudes were correlated with discomfort with violations of gender identity norms). These findings thus indicate that defining “transgender” without reference to a person’s current gender identity is correlated with anti-transgender views among cisgender people.
Furthermore, experimental work has provided causal support for the notion that exposure to a definition of the term transgender can affect cisgender individuals’ attitudes toward transgender people. Specifically, Flores et al. (2018) found that participants who were (vs. were not) provided with a definition of transgender that referred to gender identity subsequently reported lower transphobia; however, discomfort with transgender people and support for transgender rights were not impacted by exposure to the definition. Flores et al. (2018) interpreted their findings as suggesting that transphobic attitudes may stem from unfamiliarity with transgender people (e.g., from not fully understanding what being transgender means). Supporting this interpretation, previous work by the same research team revealed that cisgender people who reported feeling more knowledgeable about transgender individuals endorsed more positive attitudes (Flores, 2015). Other findings are also consistent with this interpretation. For example, anti-transgender bias can be decreased by intergroup contact (which increases familiarity; e.g., Broockman & Kalla, 2016; Norton & Herek, 2013) and by interventions that provide cisgender participants with opportunities to learn about transgender people’s experiences (Case & Stewart, 2013; McDermott et al., 2018; Walch et al., 2012). Overall, these findings suggest that cisgender people who better understand how transgender people experience their gender express less anti-transgender bias.
The Present Studies
The primary goal of the present work was to integrate and extend prior research on how cisgender people define “transgender” and how these definitions relate to their attitudes toward transgender people. In Study 1, we tested whether reading (vs. not reading) a definition of transgender that underscored the relevance of gender identity would relate to cisgender participants’ attitudes toward transgender people. Study 1 was therefore a conceptual replication of Flores et al. (2018). Study 2 focused on cisgender participants’ spontaneously generated definitions of “transgender” and tested whether the content of these definitions was associated with attitudes toward transgender people. As a secondary goal for both studies, we extended prior work by exploring the role of key individual differences in the relationship between definitions of transgender and attitudes toward transgender people. Our overall aim was to provide evidence that cisgender people’s definitions of “transgender” are associated with attitudes toward transgender people over and above established individual-difference predictors of intergroup attitudes.
Study 1
In Study 1, cisgender participants were randomly assigned to read or not read a definition of “transgender” (taken from the National Center for Transgender Equality, 2016) and then completed dependent measures. We hypothesized that cisgender participants who read (vs. did not read) the definition would report less transphobia, less support for public policies that restrict transgender rights, and more support for a policy aiming to provide equal rights to transgender people.
As a secondary goal, we examined whether the relationship between definitions of transgender and attitudes toward transgender people would emerge even when considering potentially relevant individual difference variables that might predict these attitudes. In Study 1, we specifically focused on Personal Need for Structure (PNS), which is a cognitive style that indexes a general preference for understanding the world in terms of simple structures (Neuberg & Newsom, 1993). People who score high on PNS seek to have routine, consistency, and order in their everyday lives, prefer familiar situations over novel ones, and find lack of structure uncomfortable (Neuberg & Newsom, 1993). As a result of their preference for understanding the world in terms of clearly defined categories, high-PNS individuals are prone to stereotyping (e.g., Newheiser & Dovidio, 2012). Moreover, PNS is associated with viewing female bisexuality as an unstable identity (Burke & LaFrance, 2016), supporting the notion that high-PNS individuals may experience discomfort when considering groups that do not appear to fit into clearly defined identity categories. Insofar as transgender people are perceived as violating conventional gender categories (e.g., Garelick et al., 2017), cisgender individuals who score high on PNS may therefore report anti-transgender attitudes. We also tested whether PNS moderated the potential impact of reading (vs. not reading) a definition of transgender on attitudes. Given that higher-PNS individuals are reluctant to change their existing cognitive structures (e.g., attitudes, stereotypes; Neuberg & Newsom, 1993), they may be less receptive to potentially new information about transgender people. Accordingly, higher PNS might reduce the impact of reading (vs. not reading) a definition of transgender on attitudes toward transgender people.
Method
Participants
Participants were recruited through Mechanical Turk and TurkPrime (now called CloudResearch; Litman et al., 2017) in return for $0.70. A total of 309 participants accessed the study. Ten participants did not complete any measures. Participants self-reported their gender (question wording: “What is your gender?”) by selecting one from the options of female, male, transgender, and other, please specify. Data from one participant who self-identified as transgender were excluded. We acknowledge that the wording of this item and response options is not ideal; indeed, the measurement of gender has advanced substantially since data for this study were collected. In retrospect, it would have been better to include separate items assessing participants’ gender and sex, which would have enabled us to differentiate between cisgender and transgender participants more reliably. Moreover, our response options may have been read to imply that transgender people cannot be female or male, and additional response options would have better accommodated participants with diverse gender identities. We attempted to address some of these issues in Study 2. We also excluded data from five participants who provided a nonsensical response (e.g., wrote the word “good” or “nice”) to an open-ended question asking participants to guess the purpose of the study. The final sample included 293 participants; demographic characteristics are reported in Table 1. Participants were included in all analyses for which they provided sufficient data, and variation in degrees of freedom reflects missing data. A sensitivity analysis indicated that N = 293 yielded 95% power to detect Cohen’s d = 0.42 (two-tailed, with α = 0.05 and cell ns = 146 and 147).
Measures
Attitudes toward transgender people were assessed with the 9-item Transphobia Scale (Nagoshi et al., 2008; e.g., “I would be upset if someone I’d known a long time revealed to me that they used to be another gender”; 1 = strongly disagree to 7 = strongly agree; α = 0.91; M = 3.60, SD = 1.46).
To assess policy support, participants were asked to read about policies related to restroom use (“A bathroom bill is the common name for legislation that defines access to public facilities—specifically restrooms—by transgender individuals. For example, bathroom bills may state that a person must use sex-segregated public facilities that match their sex as assigned at birth or their sex as listed on their birth certificate”), military service (“In August 2017, President Trump signed a directive that bans transgender individuals from serving in the U.S. military”), and healthcare provision (“Under federal and state laws, it is illegal for health providers, insurance companies, and other health programs to discriminate against transgender people. For example, insurance companies are not allowed to exclude gender transition-related care, refuse insurance enrollment, or limit coverage for any services based on transgender status”). These items were written for the purposes of the present study and as such results involving them should be interpreted with caution, given lack of evidence supporting their validity. Participants indicated the extent to which they supported or opposed each policy (1 = strongly oppose to 7 = strongly support; bathroom bill support M = 3.66, SD = 2.10; military ban support M = 2.82, SD = 2.07; healthcare rights support M = 5.21, SD = 2.00).
Personal Need for Structure (PNS) was assessed with an 11-item scale (Neuberg & Newsom, 1993; e.g., “I become uncomfortable when the rules in a situation are not clear”; 1 = strongly disagree to 7 = strongly agree; α = 0.89; M = 4.72, SD = 1.09).
The final survey item (completed after demographic questions) was intended to serve as a manipulation check. Participants were asked, “Which of the following would you say is the most accurate definition of what it means to be transgender?” and were given the response options “Transgender people are people whose gender identity is different from the gender they were thought to be at birth” (referencing gender identity), “Transgender people are people who change their gender to be the opposite of the gender they were born with” (referencing gender change) and “Transgender people are people who look different from the gender they were born with” (referencing gender expression). The response options were modeled after Buck (2016). We expected that participants who had (vs. had not) read a definition of transgender during the study would be more likely to select the definition referencing gender identity. We acknowledge that the response option that references gender identity does not directly contrast gender identity with assigned sex; however, by referencing a contrast between current gender identity and assumed gender at birth (which is typically assumed to coincide with assigned sex), this response option is the most comprehensive of the three.
Procedure
Both studies reported herein were deemed exempt from full review by the authors’ Institutional Review Board. The first page of the study contained an informed consent form that asked respondents to confirm their willingness to participate in the study. After providing informed consent, participants were told the study concerned attitudes toward transgender people and were randomly assigned to one of two conditions. In the Definition condition (n = 146), participants read: “Transgender people are people whose gender identity is different from the gender they were thought to be at birth. ‘Trans’ is often used as shorthand for transgender. When we’re born, a doctor usually says that we’re male or female based on what our bodies look like. Most people who were labeled male at birth turn out to actually identify as men, and most people who were labeled female at birth grow up to be women. But some people’s gender identity—their innate knowledge of who they are—is different from what was initially expected when they were born. Most of these people describe themselves as transgender” (from National Center for Transgender Equality, 2016). Participants in the Control condition (n = 147) did not receive this information. The conditions were otherwise identical. Participants then completed dependent measures (presentation order: attitudes, policy support, Personal Need for Structure), provided demographic information, responded to the manipulation check, and were debriefed. We report all measures and all experimental conditions. Data, stimuli, and measures are available at https://researchbox.org/365.
Results
Manipulation Check
We examined whether participants in the Definition (vs. Control) condition were more likely to select the response referencing gender identity when asked to identify the most accurate definition of transgender. Contrary to expectations, there was no reliable association between condition and responses on the manipulation check, \(\chi^{2}\)(2) = 0.88, p = 0.664. On average, 71% of participants (73% in the Definition condition; 69% in the Control condition) selected the definition referencing gender identity, whereas 26% selected the definition referencing gender change and 3% selected the definition referencing gender expression.
Tests of Main Hypotheses
We next tested the hypothesis that attitudes would be more positive among participants who had (vs. had not) read a definition of transgender. We did not observe support for this prediction (see Table 2): Participants’ attitudes and levels of policy support appeared independent of whether they had read the definition in the beginning of the study, ps ≥ 0.089, Cohen’s ds ≤ 0.20.
We then explored whether the impact of condition on attitudes might have been masked by key individual differences. Specifically, we examined interactions with Personal Need for Structure (PNS) and participant gender. As noted above, we expected PNS to correlate with anti-transgender attitudes. We examined associations with participant gender based on the consistent prior finding that cisgender women (vs. men) endorse more positive attitudes toward transgender people (e.g., Glotfelter & Anderson, 2017; Norton & Herek, 2013). Correlations appear in Table 3. As expected, PNS was positively correlated with transphobia and support for bathroom bills and the military ban and negatively with support for transgender healthcare rights. Participant gender was not correlated with transphobia or policy support; however, women (vs. men) scored higher on PNS.
We next conducted four linear regression analyses, estimating a separate model for each dependent variable (see Table 4). Condition (Definition = 0.5; Control = − 0.5) was entered as the independent variable and participant gender (women = 0.5; men = − 0.5) and PNS (mean-centered) were entered as moderators. We modeled all two-way interactions and the three-way interaction. We did not observe evidence for moderation (interaction ps ≥ 0.126), and the models reached statistical significance only for transphobia and military ban support. The sole consistent finding was an association between higher PNS and more negative attitudes on all four dependent measures (\(\left| {bs} \right|\)≥ 0.26, ps ≤ 0.020). Furthermore, this association was reliably stronger among female than male participants in the case of military ban support (as indicated by the two-way interaction, b = 0.55, t[279] = 2.43, p = 0.016). In addition, women (vs. men) expressed less transphobia, b = − 0.46, t(281) = − 2.72, p = 0.007.
Exploratory Analyses Based on the Manipulation Check
We finally explored the possibility that participants’ personally endorsed definitions of transgender might be a more powerful predictor of attitudes, relative to the impact of reading (vs. not reading) a definition (for which we did not observe evidence). Accordingly, we examined differences between participants who endorsed (n = 209) and did not endorse (n = 84) the definition of transgender that referenced gender identity on the manipulation check. As detailed in Table 5, participants who reported believing that the most accurate definition of transgender involved the notion of gender identity (vs. participants who indicated the most accurate definition involved the notion of gender change or gender expression) reported substantially more positive attitudes on all dependent measures, ps < 0.001, Cohen’s ds ≥ 0.86.
We also conducted exploratory analyses examining whether participant gender and PNS moderated the relationship between participant-endorsed definition and attitudes (analogous to the analyses reported in Table 4). These analyses replicated the differences between participants who endorsed (vs. did not endorse) the definition referencing gender identity while statistically adjusting for all main and interactive effects of participant gender and PNS. In addition, men (vs. women) and participants scoring higher (vs. lower) on PNS expressed more transphobia. The sole interaction that reached statistical significance was a Definition × Gender × PNS interaction on bathroom bill support. Given that this interaction was not hypothesized, we do not discuss it further. Full results from these exploratory analyses are available in Supplemental Materials.
Discussion
Study 1 revealed that cisgender participants who scored higher on Personal Need for Structure (PNS) reported more negative attitudes toward transgender individuals. This finding is in line with our predictions and with prior work demonstrating a link between PNS and prejudice against other groups that are perceived as not fitting into clearly defined identity categories (Burke & LaFrance, 2016). However, we did not find support for our main hypothesis: Cisgender participants who read (vs. did not read) a definition of transgender that focused on gender identity did not report more positive attitudes toward transgender people. This null finding is inconsistent with Flores et al. (2018), who found that participants who read (vs. did not read) a definition of transgender reported less transphobia.
What might explain the inconsistency between the present study and Flores et al. (2018)? One possible explanation is that the definition of transgender we used differed from that used by Flores et al. (2018). The definition in Flores et al. explicitly contrasted gender identity with birth-assigned sex and also referred to gender expression and gender transition (i.e., hormone therapy and surgery), whereas the definition we used in the present study focused on gender identity. Accordingly, it is possible that exposure to a more comprehensive definition that refers to dimensions of gender beyond gender identity is needed to produce an observable impact on attitudes toward transgender people. Another possibility is that enhancing awareness of the definition of transgender may no longer be sufficient for reducing anti-transgender bias because the general public’s familiarity with transgender issues has increased in recent years (Jones et al., 2019). In fact, the results we observed involving responses on the manipulation check (in which participants selected what they believed to be the most accurate definition of transgender) may be informative here. Participants who selected the definition referencing gender identity reported more positive attitudes across all dependent measures, a finding that held even when we considered the roles of PNS and participant gender (as reported in Supplemental Materials). Thus, Study 1 revealed a strong association between more positive attitudes and a tendency to define transgender by referencing gender identity rather than gender expression or gender change (consistent with Buck, 2016; see also Anderson, 2022). The fact that the majority of our sample (71%) selected the definition referencing gender identity suggests that most participants across both conditions were familiar with the relevance of gender identity in this context, which may help explain our failure to find an impact of reading (vs. not reading) the definition.
In summary, Study 1 yielded the insight that the definition of transgender that cisgender individuals endorse may be a stronger predictor of attitudes toward transgender people, relative to exposing them to a definition. However, this finding was not predicted. We therefore conducted a second study focusing on cisgender participants’ own, spontaneously expressed definitions of the word transgender.
Study 2
In Study 2, cisgender participants defined “transgender” in their own words before or after reporting their attitudes toward transgender people. We explored whether the content of participants’ definitions was associated with their attitudes toward transgender people. Whereas “transgender” has been defined in a multitude of ways (e.g., APA, n.d.; GLAAD, n.d.; National Center for Transgender Equality, 2016), we opted to focus on three dimensions of gender that are critical to the definition of transgender (Hyde et al., 2019; Tate et al., 2014): birth-assigned sex category, current gender identity, and gender presentation or expression. To do so, we adapted the procedure used by Buck (2016), who also examined the association between cisgender participants’ definitions of “transgender” and attitudes toward transgender people. Buck (2016) coded participants’ definitions for the presence of three themes: gender identity, gender expression, and gender change. Definitions that mentioned gender identity evoked internal psychological experiences of one’s gender; those that mentioned gender expression evoked gendered behavior or external appearance; and those that mentioned gender change defined “transgender” as denoting a change or switch of one’s birth-assigned sex (Buck, 2016). These three themes thus tap into the three dimensions of gender on which we sought to focus in this study. Accordingly, we used the coding categories developed by Buck (2016) and content-coded participants’ definitions for mentions of gender identity, gender expression, and gender change (with the third category tapping responses that referred to birth-assigned sex).
Buck (2016) found that cisgender participants who mentioned (vs. did not mention) gender identity in their definitions of “transgender” reported lower anti-transgender prejudice, whereas participants who mentioned (vs. did not mention) gender/sex change reported greater anti-transgender prejudice; finally, definitions that mentioned gender expression but omitted a mention of gender identity were correlated with greater prejudice. Based on the pattern observed in Study 1 and in Buck (2016; see also Anderson, 2022), we anticipated that cisgender participants who spontaneously referred to gender identity (either on its own or in combination with mentions of gender expression or change) when defining transgender would report more positive attitudes toward transgender people. We did not have a priori predictions regarding definitions that referred to gender expression or change. However, it is plausible that definitions that refer to gender change but do not refer to gender identity may be associated with more negative attitudes toward transgender people, insofar as such definitions suggest the belief that birth-assigned sex is more primary than current identity in determining a person’s gender (Buck, 2016; Hyde et al., 2019).
We additionally tested the preregistered hypothesis (see https://aspredicted.org/ga98z.pdf) that writing down one’s definition before (vs. after) reporting one’s attitudes toward transgender people might lead participants to express more positive attitudes. We reasoned that beginning the study by writing about how one would define transgender might encourage participants to more fully elaborate on their understanding of what it means to be transgender. Greater elaboration is associated with stronger attitudes that have more influence over behavior and cognition (for a review, see Petty & Cacioppo, 1986). Elaborating on one’s definition of transgender may also afford one a sense of being familiar with and knowledgeable about what it means to be transgender, which has been found to be associated with more positive attitudes (Broockman & Kalla, 2016; Flores, 2015; Flores et al., 2018). Although we preregistered this prediction, we acknowledge that merely writing down one’s definition of transgender might instead provide participants with a false sense of familiarity or may inform them about the goals of the study (creating demand characteristics). As such, testing this prediction was secondary to our main goal of examining the content of participants’ definitions of transgender.
Lastly, we explored several potentially relevant individual differences. As in Study 1, we measured Personal Need for Structure and participant gender. In addition, we assessed identification with one’s gender. Higher gender identification has been found to predict more negative attitudes toward transgender people among cisgender heterosexual and bisexual men but not among cisgender women or gay men (Anderson, 2018; Glotfelter & Anderson, 2017). Thus, transgender people may threaten the distinctiveness of gender categories that are central to the self-concepts of cisgender people (perhaps particularly cisgender men) who identify strongly with their gender. As such, we expected that stronger gender identification would predict more negative attitudes. Furthermore, we asked participants whether they were personally acquainted with a transgender person to assess the role of intergroup contact (which has been shown to reduce anti-transgender attitudes; Broockman & Kalla, 2016).
In addition to testing the main effects of Personal Need for Structure, participant gender, gender identification, and intergroup contact on attitudes toward transgender people, we tested whether these variables moderated any potential effects of writing down one’s own definition of transgender before (vs. after) reporting one’s attitudes. Although we did not have specific a priori predictions regarding potential moderation, we reasoned that individual differences associated with more negative attitudes toward transgender people might make participants less receptive to any intervention aiming to improve attitudes. Thus, writing down one’s definition of transgender before (vs. after) reporting one’s attitudes might be less likely to have a detectable impact among cisgender participants scoring higher on PNS, identifying as men, being more strongly identified with their gender, and not personally acquainted with a transgender person.
Method
Participants
Our preregistered goal (see https://aspredicted.org/ga98z.pdf) was to recruit 150 cisgender participants per condition for a total N = 300. Given anticipated data loss due to a priori exclusion criteria, the study was made available to 400 participants recruited via Mechanical Turk in return for $0.70. A total of 407 participants accessed the study.
Participants self-reported their gender (question wording: “What is your gender?”) by selecting one from the following response options: cis (non-transgender) woman, cis (non-transgender) man, trans woman, trans man, or other, please specify. Although this measure addressed many of the issues identified with the measure of gender we used in Study 1, we acknowledge that additional response options would still be needed to comprehensively assess the diversity of genders with which people may identify. As preregistered, we excluded data from participants who did not identify as cisgender or as exclusively or mostly heterosexual, or did not report their gender identity or sexual orientation. We made the a priori decision to exclude data from both non-cisgender and non-straight participants because non-heterosexual (vs. heterosexual) people have been found to report more positive attitudes toward transgender people (e.g., Anderson, 2022; Warriner et al., 2013), but the number of non-heterosexual participants was not expected to be sufficiently large to enable a formal test of sexual orientation as a moderator. This preregistered exclusion criterion was used only in Study 2 (Study 1 did not employ preregistered exclusion criteria).
As also preregistered, we additionally excluded data from Study 2 participants who had duplicate IP addresses or duplicate geolocation coordinates, did not provide their definition of transgender (i.e., did not follow instructions), or took less than two minutes to complete the study (reflecting potential inattention). The final sample included 295 participants; demographic characteristics are reported in Table 6. Participants were included in all analyses for which they provided sufficient data, and variation in degrees of freedom reflects missing data. Sensitivity analyses indicated that N = 295 yielded 95% power (two-tailed, with α = 0.05) to detect Cohen’s d = 0.42 (with cell ns = 147 and 148) and r = 0.21.
Measures
Attitudes toward transgender people were assessed with the Context-Dependent Transprejudice Scale (Buck & Obzud, 2018; 1 = strongly disagree to 7 = strongly agree), which indexes attitudes toward transgender individuals in gender-segregated settings (6 items; e.g., “I believe transgender individuals should be able to use whichever public restroom makes them feel most comfortable”; α = 0.93; M = 4.40, SD = 1.79) and gender-integrated settings (3 items; e.g., “I believe transgender job candidates should be given equal opportunity for employment and should not be subject to discrimination based on their gender identity”; α = 0.83; M = 6.13, SD = 1.13).
Policy support was assessed with five items from Tee and Hegarty (2006) referring to rights concerning legal documentation (“Transgender individuals should have the right to have a new passport issued,” “Transgender individuals should have the right to have a new birth certificate issued,” and “Transgender individuals should have the right to have a new driving license issued”) and gender-segregated facilities (“Transgender individuals should have the right to be treated in a hospital appropriate to their ‘new’ gender” and “Transgender individuals should have the right to be detained in a prison appropriate to their ‘new’ gender”). Participants indicated their agreement with each statement (1 = strongly disagree to 7 = strongly agree), and scores were averaged into a single index (α = 0.90; M = 4.83, SD = 1.65).
Participants next completed measures of individual difference predictors. Personal Need for Structure was assessed as in Study 1 (α = 0.88; M = 4.86, SD = 1.02). We adapted Luhtanen and Crocker’s (1992) Collective Self-Esteem scale to the context of gender, and the four items comprising the Identity subscale were averaged into an index of gender identification (following Anderson, 2018; e.g., “The gender group I belong to is an important reflection of who I am”; 1 = strongly disagree to 7 = strongly agree; α = 0.82; M = 4.52, SD = 1.47). Finally, the last survey item (completed after demographic questions) measured contact with transgender people. Participants were asked, “Do you personally know anyone who is transgender?” (response options: yes, no). Those who selected “yes” (107 participants; 36.3% of the final sample) were asked to specify their relationship with the transgender person(s) they knew by selecting all that applied from the options of family member (selected by 13 participants), friend (selected by 45 participants), acquaintance (selected by 47 participants), coworker (selected by 27 participants), and “other, please specify” (selected by seven participants; e.g., “friend’s child”).
Procedure
The first page of the study contained an informed consent form that asked respondents to confirm their willingness to participate in the study. In the Definition First condition (n = 148), participants began the study by defining transgender in their own words. Participants were instructed as follows: “In this research, we are interested in learning about people’s attitudes toward and opinions about transgender people. First, we’re interested in what you think it means to be transgender. In your own words, how would you define the word transgender?” After typing in their definitions, participants completed measures of attitudes toward transgender people. Participants in the Definition Last condition (n = 147) began the study by completing the attitude measures, after which they were asked to define transgender in their own words. Thus, the sole difference between the conditions was the order in which these two tasks were completed. All participants then completed measures of gender identification and Personal Need for Structure (in counterbalanced order), provided demographic information, including a question assessing contact with transgender people, and were debriefed. We report all measures and all experimental conditions. Stimuli and measures are available at https://researchbox.org/365.
Measure (Content Coding)
Participants’ definitions of transgender were categorized by two independent coders. Coders were provided with definitions and example responses for three coding categories; the category definitions and examples were taken verbatim from Buck (2016). Gender identity was defined as “the participant mentioned internal psychological features, such as identifying with or feeling like a particular gender” with the example response of “identifying with a different gender (or sex) than the one assigned at birth.” Gender expression was defined as “the participant mentioned gendered behavior, clothing, appearance, or anatomy” with the example response of “a person who dresses in the opposite fashion of their birth gender.” Gender change was defined as “the participant described switching or changing gender” with the example response of “someone who changes their gender to the opposite of their natural-born gender.” Coders were told responses could be categorized into no, any, or all categories and were instructed to assign a code of 0 (category was not mentioned) or 1 (category was mentioned) for all three categories for each response. Thus, the criterion for determining whether a particular response fit a particular coding category was simply whether coders believed the response matched the description of each category. Coders were blind to condition and to participants’ responses on other measures included in the study.
We assessed interrater reliability with percent agreement and Cohen’s κ. Greater than 80% interrater agreement and κ > 0.6 is typically taken to indicate sufficient reliability (Cohen, 1960; McHugh, 2012). Interrater reliability was acceptable, albeit close to the lower limit of the acceptable range for κ for the gender expression category (gender expression: Cohen’s κ = 0.61, coder agreement = 91.9%; gender identity: κ = 0.73, coder agreement = 87.1%; gender change: κ = 0.74, coder agreement = 90.8%). Discrepancies were discussed in a meeting with the coders and the first and third authors and were reconciled based on consensus.
Results
Preregistered Tests of Differences Between Conditions
All data are available at https://researchbox.org/365. Following our preregistered analysis plan (https://aspredicted.org/ga98z.pdf), we conducted independent-samples t-tests to test whether attitudes toward transgender people were more positive among participants who wrote down their definitions of transgender before (vs. after) reporting attitudes (see Table 7). Contrary to predictions, we did not observe reliable differences between conditions, ps ≥ 0.826, Cohen’s ds ≤ 0.03.
Preregistered and Exploratory Moderation Analyses
We next examined whether condition (Definition First vs. Last) interacted with gender identification, PNS, contact with transgender people, and participant gender to predict attitudes toward transgender people. We first computed bivariate correlations across all variables (see Table 8). Gender identification was not correlated with attitudes. PNS was negatively correlated with attitudes toward transgender people in gender-segregated settings. Participants who reported personally knowing (vs. not knowing) a transgender person and women (vs. men) expressed more positive attitudes on all three dependent measures.
Our preregistered analysis plan specified regression models in which the impact of condition on attitudes was separately moderated by gender identification and PNS. We additionally conducted exploratory moderation analyses including participant gender as a second moderator in these models, paralleling the analyses reported in Study 1. Finally, we assessed intergroup contact as an additional exploratory moderator. Because we did not observe evidence for moderation across these analyses, for brevity they are detailed in Supplemental Materials. The sole interaction term that reached statistical significance suggested that higher gender identification predicted lower support for transgender rights policies among male (but not female) participants. Thus, none of the individual differences qualified the impact of condition on attitudes, and associations among these variables may be best understood in terms of the correlations reported in Table 8.
Content Analysis of Participants’ Definitions of “Transgender”
We finally examined participants’ mentions of gender identity, gender expression, and gender change in their definitions of “transgender” (these analyses were not preregistered). A total of 186 participants (63% of the sample) mentioned gender identity (example response: “Someone who was born one gender but identifies with another gender”); 29 participants (10%) mentioned gender expression (example response: “A person who dresses up to look like a member of the opposite sex and adopts that genders [sic] characteristics also”); and 67 participants (23%) mentioned gender change (example response: “A person who has changed his/her birth sex to the opposite sex”). Participants who mentioned (vs. did not mention) gender identity were less likely to also mention gender change, r(293) = − 0.36, p < 0.001. Mentions of gender expression were not reliably associated with mentions of gender identity, r(293) = − 0.08, p = 0.184, or gender change, r(293) = 0.01, p = 0.848.
Given our coding scheme, participants could be categorized as mentioning any combination of gender identity, expression, and change. We explored possible clustering of participants based on patterns of these themes being mentioned in their definitions. Specifically, we conducted a two-step cluster analysis on the binary definition codes using log-likelihood as the distance measure, automatic determination of clusters, and Schwarz’s BIC as the clustering criterion. The silhouette measure of cohesion and separation indicated good cluster quality (average silhouette coefficient = 0.9) and the analysis suggested the presence of five clusters.
The largest cluster consisted of the 154 participants (52.2% of the sample) who mentioned gender identity but did not mention gender expression or gender change in their definitions (labeled as the Identity Cluster). Another cluster consisted of the 43 participants (14.6% of the sample) who mentioned gender change but did not mention gender expression or identity (labeled as the Change Cluster). A third cluster contained the 17 participants (5.8% of the sample) who mentioned both gender identity and gender change but did not mention gender expression (labeled as the Identity and Change Cluster). A fourth cluster consisted of 29 participants (9.8% of the sample), all of whom mentioned gender expression; 15 of these 29 participants also mentioned gender identity, seven also mentioned gender change, and four mentioned all three themes. Given the consistent mentioning of gender expression among these participants (even though other themes were also mentioned), they were labeled as the Expression Cluster. Finally, the last cluster consisted of the 52 participants (17.6% of the sample) who did not mention any of gender identity, expression, or change (labeled as the “Other” Cluster).
We tested whether these five clusters of participants reported differential attitudes toward transgender people; means and standard deviations are presented in Table 9. Due to unequal cluster sizes and heterogeneity of variance, we used Welch’s ANOVA with Games–Howell post hoc tests (Frost, n.d.; Games & Howell, 1976). We observed reliable differences across the five clusters on attitudes toward transgender people in gender-segregated settings, F(4, 70.22) = 4.68, p = 0.002, \(\eta_{p}^{2}\) = 0.21, attitudes toward transgender people in gender-integrated settings, F(4, 67.26) = 9.51, p < 0.001, \(\eta_{p}^{2}\) = 0.36, and policy support, F(4, 69.65) = 9.01, p < 0.001, \(\eta_{p}^{2}\) = 0.34. Post hoc comparisons revealed that participants in the “Other” Cluster reported less positive attitudes in gender-segregated settings than did participants in the Identity Cluster, p = 0.003, d = 0.61, and participants in the Identity and Change Cluster, p = 0.037, d = 0.71. Participants in the “Other” Cluster also reported less positive attitudes in gender-integrated settings than did participants in the Identity Cluster, p < 0.001, d = 1.17, and participants in the Identity and Change Cluster, p = 0.001, d = 0.84; in addition, participants in the Change Cluster reported less positive attitudes in gender-integrated settings than did participants in the Identity Cluster, p = 0.008, d = 0.79. Finally, participants in the “Other” Cluster reported lower policy support than did participants in the Identity Cluster, p < 0.001, d = 0.97, and participants in the Identity and Change Cluster, p = 0.001, d = 0.98. All other pairwise comparisons were statistically nonsignificant (see Table 9).
Discussion
Study 2 revealed several correlational patterns that are consistent with prior research. Specifically, positive attitudes and policy support were higher among cisgender women (vs. cisgender men) and among participants who reported personally knowing (vs. not knowing) a transgender person. In addition, Personal Need for Structure was correlated with less positive attitudes on items concerning gender-segregated life domains (e.g., bathrooms). However, we did not find support for the prediction that writing down one’s own definition of transgender before (vs. after) reporting one’s attitudes toward transgender people would result in more positive attitudes. Although lack of statistical significance cannot be taken to directly indicate the absence of an effect, this pattern suggests that simply thinking about the definition of transgender is not sufficient to provide a sense of familiarity that might improve attitudes. Indeed, prior work suggests that interventions that promote sustained attitude change involve direct contact with transgender people and exposure to the social issues faced by this community (Broockman & Kalla, 2016; Walch et al., 2012).
The main contribution of the present study pertained to participants’ spontaneous definitions of transgender. About half of our participants were coded as defining “transgender” by referring to gender identity. Additionally, correlations across the definition codes revealed that participants who referred to gender identity in their definitions were less likely to also refer to gender change but were not any more or less likely to refer to gender expression. This pattern suggests that participants’ definitions of transgender were generally consistent with institutional definitions that can include both gender identity and gender expression (e.g., APA, 2015, APA, n.d.). Moreover, cluster analysis pointed to the potential existence of five types of definitions: exclusively referencing gender identity; exclusively referencing gender change; referencing both gender identity and gender change; primarily relying on the notion of gender expression; and defining transgender without reference to gender identity, expression, or change.
Importantly, we observed differences in attitudes toward transgender people across these five definition clusters. Some clusters were small, which reduced statistical power for detecting differences (potentially explaining the lack of statistically significant differences of the Expression cluster from all other clusters). We nonetheless observed that attitudes indexed by all three dependent measures were substantially more positive among participants who defined transgender by bringing up the notion of gender identity (either on its own or in combination with gender change), especially relative to participants whose definitions did not mention any of gender identity, expression, or change. Thus, we conceptually replicated the primary finding from Study 1, which suggested that endorsing a definition of transgender that relies on gender identity is correlated with more positive attitudes.
We explored definitions categorized into the “Other” cluster to better understand why participants in this cluster might have reported less positive attitudes. Definitions fell into this cluster if independent coders determined that a response did not completely match any of the gender identity, expression, or change categories; thus, some responses in the “Other” cluster shared aspects with definitions in the additional clusters but were not fully in line with how gender identity, expression, and change were defined for coders. For instance, some of these responses defined transgender as referring to people who “try” to be or “think” they are the “opposite” gender. Examples of such definitions written by participants in this cluster included “a male or female trying to be the opposite of the sex they were born” and “someone who thinks they can be the opposite sex.” Such definitions downplay or deny the importance of gender identity by suggesting that transgender people may want to or try to “change” their gender but ultimately cannot do so. These definitions therefore also differed from those expressed by participants in the Change cluster, which asserted a successful “change” in gender. These definitions also imply a reliance on the gender binary by referring to “opposite” genders. A second theme that emerged in the “Other” cluster involved definitions referring to mental illness (e.g., “Someone who has psychological problems, seeks attention and is in desperate need of help”; “To me, a transgender person is one who is mentally ill. It is not normal to think you are supposed to be the opposite gender you were born with”) or confusion (e.g., “A very confused individual”; “People who are confused about what gender they really are”). Despite these interesting themes, we urge some caution in interpreting these patterns given the heterogeneity of and lack of unity among definitions that fell into the “Other” cluster. Overall, we interpret the observed patterns as indicating that cisgender participants who defined “transgender” by explicitly and clearly referring to gender identity also expressed more positive attitudes toward transgender people.
General Discussion
The present studies revealed that cisgender respondents who referred to gender identity when defining “transgender” (by selecting a definition from a list of options in Study 1 or spontaneously in their own words in Study 2) reported more positive attitudes toward transgender people on a variety of measures. By contrast, referencing gender expression or gender change when defining transgender did not systematically predict attitudes. Thus, although we did not replicate previous findings that exposure to a definition of transgender can improve attitudes (Flores et al., 2018), we did find that how cisgender individuals define gender identity-relevant terms is meaningfully related to their attitudes toward transgender people (in line with Buck, 2016; see also Anderson, 2022). In summary, our findings suggest that cisgender people who do not prioritize people’s internal experience of their own gender when thinking about what “transgender” means might be particularly likely to express contemporary forms of transphobia.
Future research may benefit from examining how cisgender people arrive at their personal definitions of transgender, and from exploring factors that contribute to the formation of these definitions. Contact with transgender people may represent one such factor. Furthermore, it is important to consider how individuals’ social group memberships may influence how they define identity-relevant terms such as transgender. For example, Schudson et al. (2019) asked a gender-diverse sample to define the words “woman,” “man,” “female,” “male,” “feminine,” and “masculine” and found that gender-minority participants provided more complex definitions than did gender-nonminority participants. Although we limited our samples to cisgender participants, it is likely that people with diverse gender or sexual identities might articulate more nuanced definitions of transgender as well (Beischel et al., 2021a, 2021b; Schudson et al., 2019). Future studies might test whether exposing straight cisgender people to definitions of terms such as “transgender” or “gender identity” as articulated by non-cisgender and non-straight people would have a measurable impact on attitudes toward gender-diverse groups.
Whereas both of our studies demonstrated that the inclusion of gender identity in one’s definition of transgender was related to more positive attitudes toward transgender people, these findings are correlational. Neither of our experimental manipulations provided support for a hypothesized causal relation: Exposure (vs. lack thereof) to a definition of transgender did not affect attitudes in Study 1 and considering one’s own definition of transgender before (vs. after) completing dependent measures did not affect attitudes in Study 2. We thus cannot conclude whether believing that being transgender involves gender identity leads to more positive attitudes, or whether holding more positive attitudes leads cisgender people to endorse a definition of transgender that centers the notion of gender identity. It is also possible that some participants mentioned gender identity when defining transgender simply because they were aware that this is a socially desirable response. Future work would benefit from directly examining the extent to which cisgender people have a comprehensive grasp on what gender identity means to gender-diverse populations.
We additionally suggest that assessing how individuals define transgender may have implications for the measurement of attitudes toward transgender people. Several current measures of transphobia consist of strongly worded items (e.g., “Transgender men/women are unnatural” from Billard, 2018). As a result, respondents may be more concerned with appearing politically correct than with answering candidly. Researchers have suggested that implicit measures may hold promise in this context (Axt et al., 2021; Kanamori et al., 2020). Many implicit attitude measures, including the popular Implicit Association Test (IAT), often rely on visual stimuli. Moreover, even though some of these measures (including the IAT) can rely solely on text stimuli, generating text stimuli beyond the words “transgender” and “cisgender” for use in such measures seems conceptually and practically difficult. As a result, traditional implicit measures may not be easily adapted to assess attitudes toward transgender people. We propose that these attitudes could be indirectly measured by content-coding respondents’ spontaneous definitions of transgender for mentions of themes such as gender identity. This measurement approach would not rely on the development of large sets of visual or verbal stimuli to represent gender identity categories and is less overt than existing explicit measures of transphobia. Content-coding definitions of “transgender” for themes other than gender identity might also aid in the development of more modern forms of transprejudice, paralleling measures of modern sexism and modern heterosexism (Morrison & Morrison, 2003; Morrison et al., 1999). For example, cisgender people express reliably more negative attitudes toward transgender people on items that concern life domains that are typically gender-segregated (e.g., public restrooms), relative to items that concern gender-integrated domains (e.g., employment rights; Buck & Obzud, 2018). Modern transphobia may thus manifest itself particularly strongly in terms of how cisgender people think about gendered spaces and whether individuals with specific gender identities belong in “women-only” or “men-only” spaces. Asking respondents to define “transgender” in relation to gender-segregated spaces in particular might enable future studies to capture critical nuance in contemporary attitudes toward transgender people.
Despite the contributions of our work, we acknowledge key limitations. First, although prior studies have found that gaining information from, listening to, or seeing someone who is transgender can lower prejudice (Flores et al., 2018; Gillig et al., 2018; Tompkins et al., 2015), our Study 1 suggested that providing (vs. not providing) cisgender people with a definition of transgender was not associated with attitudes toward transgender people. The source of the information may be an important factor. In our Study 1, participants simply read a definition. In contrast, Gillig et al. (2018) and Tompkins et al. (2015) used documentary films and TV shows to educate participants. The inclusion of emotional and social elements may have made these interventions more impactful than our approach. It is also possible that the definition to which we exposed Study 1 participants may have differed from definitions participants might have learned about previously (e.g., ones placing greater emphasis on gender expression), and exposure to the definition during Study 1 may thus not have impacted attitudes due to a conflict with pre-existing knowledge.
Additionally, our samples consisted nearly exclusively of people residing in the US. We thus cannot speak to any possible cross-cultural generalization of the patterns we observed. There are established cultural differences in how people conceptualize gender, for example with some cultures (unlike the US) recognizing additional genders beyond women and men (e.g., India; Goel, 2016). Given that research using US samples has demonstrated that cisgender people’s attitudes toward transgender people are associated with a belief in binary gender (e.g., Norton & Herek, 2013), it is plausible that the correlates of attitudes toward transgender people vary cross-culturally as well (Elischberger et al., 2018; Worthen et al., 2017). Future work would greatly benefit from examining whether cisgender people’s definitions of “transgender” differ across cultures, particularly in terms of whether the centrality of the notion of gender identity in these definitions (as well as its connection to attitudes toward transgender people) is observed in cultures other than the US. Moreover, existing work on cisgender people’s definitions of transgender conducted with US-based respondents has tended to rely on convenience samples that overrepresent individuals self-identifying as White, middle-class, and native speakers of English; the present studies are no exception to this trend. Generalization to more nationally representative samples is therefore another productive direction for future studies.
In conclusion, with increasingly many people—especially young people (e.g., Kidd et al., 2021)—identifying with genders that go beyond the traditional binary of women and men, how people think about gender identity and the relationship between gender and sex continues to evolve. Our findings suggest that cisgender people who center the notion of gender identity when thinking about what the term transgender means also endorse more positive attitudes toward transgender individuals. Efforts to reduce transphobia may therefore benefit from expanding people’s awareness of the multidimensional nature of gender (Hyde et al., 2019; Tate et al., 2014) and how gender identity, gender expression, gender roles, and birth-assigned sex jointly inform gendered life experiences.
References
Adams, K. A., Nagoshi, C. T., Filip-Crawford, G., Terrell, H. K., & Nagoshi, J. L. (2016). Components of gender-nonconformity prejudice. International Journal of Transgenderism, 17(3–4), 185–198. https://doi.org/10.1080/15532739.2016.1200509
Anderson, V. N. (2018). Cisgender men and trans prejudice: Relationships with sexual orientation and gender self-esteem. Psychology of Men and Masculinity, 19(3), 373–384. https://doi.org/10.1037/men0000125
Anderson, V. N. (2022). What does transgender mean to you? Transgender definitions and attitudes toward trans people. Psychology of Sexual Orientation and Gender Diversity. https://doi.org/10.1037/sgd0000566
APA. (2015). Guidelines for psychological practice with transgender and gender nonconforming people. American Psychologist, 70(9), 832–864. https://doi.org/10.1037/a0039906
APA. (n.d.). Transgender. Retrieved from https://dictionary.apa.org/transgender
Axt, J. R., Conway, M. A., Westgate, E. C., & Buttrick, N. R. (2021). Implicit transgender attitudes independently predict beliefs about gender and transgender people. Personality and Social Psychology Bulletin, 47(2), 257–274. https://doi.org/10.1177/0146167220921065
Beischel, W. J., Schudson, Z. C., & van Anders, S. M. (2021a). Visualizing gender/sex diversity via sexual configurations theory. Psychology of Sexual Orientation and Gender Diversity, 8(1), 1–13. https://doi.org/10.1037/sgd0000449
Beischel, W. J., Schudson, Z. C., & van Anders, S. M. (2021b). “This is mind expanding”: Reactions to an online survey using sexual configurations theory. Psychology of Sexual Orientation and Gender Diversity, 8(1), 14–24. https://doi.org/10.1037/sgd0000450
Billard, T. J. (2018). Attitudes toward transgender men and women: Development and validation of a new measure. Frontiers in Psychology, 9, 387. https://doi.org/10.3389/fpsyg.2018.00387
Broockman, D., & Kalla, J. (2016). Durably reducing transphobia: A field experiment on door-to-door canvassing. Science, 352(6282), 220–224. https://doi.org/10.1126/science.aad9713
Buck, D. M. (2016). Defining transgender: What do lay definitions say about prejudice? Psychology of Sexual Orientation and Gender Diversity, 3(4), 465–472. https://doi.org/10.1037/sgd0000191
Buck, D. M., & Obzud, T. (2018). Context-dependent transprejudice: Attitudes toward transgender individuals in gender-integrated and gender-segregated settings. Psychology of Sexual Orientation and Gender Diversity, 5(1), 117–121. https://doi.org/10.1037/sgd0000251
Burke, S. E., & LaFrance, M. (2016). Lay conceptions of sexual minority groups. Archives of Sexual Behavior, 45, 635–650. https://doi.org/10.1007/s10508-015-0655-5
Case, K. A., & Stewart, B. (2013). Intervention effectiveness in reducing prejudice against transsexuals. Journal of LGBT Youth, 10(1–2), 140–158. https://doi.org/10.1080/19361653.2012.718549
Ching, B. H. H., & Xu, J. T. (2018). The effects of gender neuroessentialism on transprejudice: An experimental study. Sex Roles, 78, 228–241. https://doi.org/10.1007/s11199-017-0786-3
Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20(1), 37–46. https://doi.org/10.1177/001316446002000104
Dar-Nimrod, I., & Heine, S. J. (2011). Genetic essentialism: On the deceptive determinism of DNA. Psychological Bulletin, 137(5), 800–818. https://doi.org/10.1037/a0021860
Elischberger, H. B., Glazier, J. J., Hill, E. D., & Verduzco-Baker, L. (2018). Attitudes toward and beliefs about transgender youth: A cross-cultural comparison between the United States and India. Sex Roles, 78, 142–160. https://doi.org/10.1007/s11199-017-0778-3
Flores, A. R. (2015). Attitudes toward transgender rights: Perceived knowledge and secondary interpersonal contact. Politics, Groups, and Identities, 3(3), 398–416. https://doi.org/10.1080/21565503.2015.1050414
Flores, A. R., Haider-Markel, D. P., Lewis, D. C., Miller, P. R., Tadlock, B. L., & Taylor, J. K. (2018). Challenged expectations: Mere exposure effects on attitudes about transgender people and rights. Political Psychology, 39(1), 197–216. https://doi.org/10.1111/pops.12402
Frost, J. (n.d.). Benefits of Welch’s ANOVA compared to the classic one-way ANOVA. Retrieved from https://statisticsbyjim.com/anova/welchs-anova-compared-to-classic-one-way-anova/
Games, P. A., & Howell, J. F. (1976). Pairwise multiple comparison procedures with unequal n’s and/or variances: A Monte Carlo study. Journal of Educational Statistics, 1(2), 113–125. https://doi.org/10.3102/10769986001002113
Garelick, A. S., Filip-Crawford, G., Varley, A. H., Nagoshi, C. T., Nagoshi, J. L., & Evans, R. (2017). Beyond the binary: Exploring the role of ambiguity in biphobia and transphobia. Journal of Bisexuality, 17(2), 172–189. https://doi.org/10.1080/15299716.2017.1319890
Gillig, T. K., Rosenthal, E. L., Murphy, S. T., & Folb, K. L. (2018). More than a media moment: The influence of televised storylines on viewers’ attitudes toward transgender people and policies. Sex Roles, 78, 515–527. https://doi.org/10.1007/s11199-017-0816-1
GLAAD. (n.d.). GLAAD media reference guide (11th edition)—Glossary of terms: Transgender. Retrieved from https://www.glaad.org/reference/trans-terms
Glotfelter, M. A., & Anderson, V. N. (2017). Relationships between gender self-esteem, sexual prejudice, and trans prejudice in cisgender heterosexual college students. International Journal of Transgenderism, 18(2), 182–198. https://doi.org/10.1080/15532739.2016.1274932
Goel, I. (2016). Hijra communities of Delhi. Sexualities, 19(5–6), 535–546. https://doi.org/10.1177/1363460715616946
Hill, D. B., & Willoughby, B. L. (2005). The development and validation of the genderism and transphobia scale. Sex Roles, 53, 531–544. https://doi.org/10.1007/s11199-005-7140-x
Hyde, J. S., Bigler, R. S., Joel, D., Tate, C. C., & van Anders, S. M. (2019). The future of sex and gender in psychology: Five challenges to the gender binary. American Psychologist, 74(2), 171–193. https://doi.org/10.1037/amp0000307
Jones, J. M. (2022). LGBT identification in U.S. ticks up to 7.1%. Gallup. https://news.gallup.com/poll/389792/lgbt-identification-ticks-up.aspx
Jones, R. P., Jackson, N., Najle, M., Bola, O., & Greenberg, D. (2019). America’s growing support for transgender rights. PRRI. https://www.prri.org/research/americas-growing-support-for-transgender-rights
Kanamori, Y., Harrell-Williams, L. M., Xu, Y. J., & Ovrebo, E. (2020). Transgender affect misattribution procedure (transgender AMP): Development and initial evaluation of performance of a measure of implicit prejudice. Psychology of Sexual Orientation and Gender Diversity, 7(1), 3–11. https://doi.org/10.1037/sgd0000343
Kidd, K. M., Sequeira, G. M., Douglas, C., Paglisotti, T., Inwards-Breland, D. J., Miller, E., & Coulter, R. W. (2021). Prevalence of gender-diverse youth in an urban school district. Pediatrics, 147(6), e2020049823. https://doi.org/10.1542/peds.2020-049823
Litman, L., Robinson, J., & Abberbock, T. (2017). TurkPrime.com: A versatile crowdsourcing data acquisition platform for the behavioral sciences. Behavior Research Methods, 49, 433–442. https://doi.org/10.3758/s13428-016-0727-z
Luhtanen, R., & Crocker, J. (1992). A collective self-esteem scale: Self-evaluation of one’s social identity. Personality and Social Psychology Bulletin, 18(3), 302–318. https://doi.org/10.1177/0146167292183006
McDermott, D. T., Brooks, A. S., Rohleder, P., Blair, K., Hoskin, R. A., & McDonagh, L. K. (2018). Ameliorating transnegativity: Assessing the immediate and extended efficacy of a pedagogic prejudice reduction intervention. Psychology & Sexuality, 9(1), 69–85. https://doi.org/10.1080/19419899.2018.1429487
McHugh, M. L. (2012). Interrater reliability: The kappa statistic. Biochemia Medica, 22(3), 276–282.
Morgenroth, T., & Ryan, M. K. (2021). The effects of gender trouble: An integrative theoretical framework of the perpetuation and disruption of the gender/sex binary. Perspectives on Psychological Science, 16(6), 1113–1142. https://doi.org/10.1177/1745691620902442
Morrison, M. A., & Morrison, T. G. (2003). Development and validation of a scale measuring modern prejudice toward gay men and lesbian women. Journal of Homosexuality, 43(2), 15–37. https://doi.org/10.1300/J082v43n02_02
Morrison, M. A., Morrison, T. G., Pope, G. A., & Zumbo, B. D. (1999). An investigation of measures of modern and old-fashioned sexism. Social Indicators Research, 48(1), 39–49. https://doi.org/10.1023/A:1006873203349
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. https://doi.org/10.1007/s11199-008-9458-7
National Center for Transgender Equality. (2016). Frequently asked questions about transgender people. Retrieved from https://transequality.org/issues/resources/frequently-asked-questions-about-transgender-people
Neuberg, S. L., & Newsom, J. T. (1993). Personal need for structure: Individual differences in the desire for simpler structure. Journal of Personality and Social Psychology, 65(1), 113–131. https://doi.org/10.1037/0022-3514.65.1.113
Newheiser, A. K., & Dovidio, J. F. (2012). Individual differences and intergroup bias: Divergent dynamics associated with prejudice and stereotyping. Personality and Individual Differences, 53(1), 70–74. https://doi.org/10.1016/j.paid.2012.02.024
Norton, A. T., & Herek, G. M. (2013). Heterosexuals’ attitudes toward transgender people: Findings from a national probability sample of US adults. Sex Roles, 68, 738–753. https://doi.org/10.1007/s11199-011-0110-6
Parker, K., Horowitz, J. M., & Brown, A. (2022). Americans’ complex views on gender identity and transgender issues. Pew Research Center. https://www.pewresearch.org/social-trends/2022/06/28/americans-complex-views-on-gender-identity-and-transgender-issues/
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. Advances in Experimental Social Psychology, 19, 123–205. https://doi.org/10.1016/S0065-2601(08)60214-2
Prusaczyk, E., & Hodson, G. (2020). The roles of political conservatism and binary gender beliefs in predicting prejudices toward gay men and people who are transgender. Sex Roles, 82, 438–446. https://doi.org/10.1007/s11199-019-01069-1
Rad, M. S., Shackleford, C., Lee, K. A., Jassin, K., & Ginges, J. (2019). Folk theories of gender and antitransgender attitudes: Gender differences and policy preferences. PLoS ONE, 14, e0226967. https://doi.org/10.1371/journal.pone.0226967
Schudson, Z. C., Beischel, W. J., & van Anders, S. M. (2019). Individual variation in gender/sex category definitions. Psychology of Sexual Orientation and Gender Diversity, 6(4), 448–460. https://doi.org/10.1037/sgd0000346
Tate, C. C., Youssef, C. P., & Bettergarcia, J. N. (2014). Integrating the study of transgender spectrum and cisgender experiences of self-categorization from a personality perspective. Review of General Psychology, 18(4), 302–312. https://doi.org/10.1037/gpr0000019
Tee, N., & Hegarty, P. (2006). Predicting opposition to the civil rights of trans persons in the United Kingdom. Journal of Community and Applied Social Psychology, 16(1), 70–80. https://doi.org/10.1002/casp.851
Tompkins, T. L., Shields, C. N., Hillman, K. M., & White, K. (2015). Reducing stigma toward the transgender community: An evaluation of a humanizing and perspective-taking intervention. Psychology of Sexual Orientation and Gender Diversity, 2(1), 34–42. https://doi.org/10.1037/sgd0000088
Walch, S. E., Sinkkanen, K. A., Swain, E. M., Francisco, J., Breaux, C. A., & Sjoberg, M. D. (2012). Using intergroup contact theory to reduce stigma against transgender individuals: Impact of a transgender speaker panel presentation. Journal of Applied Social Psychology, 42(10), 2583–2605. https://doi.org/10.1111/j.1559-1816.2012.00955.x
Warriner, K., Nagoshi, C. T., & Nagoshi, J. L. (2013). Correlates of homophobia, transphobia, and internalized homophobia in gay or lesbian and heterosexual samples. Journal of Homosexuality, 60(9), 1297–1314. https://doi.org/10.1080/00918369.2013.806177
Wilton, L. S., Bell, A. N., Carpinella, C. M., Young, D. M., Meyers, C., & Clapham, R. (2019). Lay theories of gender influence support for women and transgender people’s legal rights. Social Psychological and Personality Science, 10(7), 883–894. https://doi.org/10.1177/1948550618803608
Worthen, M. G. F., Lingiardi, V., & Caristo, C. (2017). The roles of politics, feminism, and religion in attitudes toward LGBT individuals: A cross-cultural study of college students in the USA, Italy, and Spain. Sexuality Research and Social Policy, 14, 241–258. https://doi.org/10.1007/s13178-016-0244-y
Funding
The authors did not receive support from any organization for the submitted work.
Author information
Authors and Affiliations
Contributions
All authors contributed to this research. Material preparation and data collection were performed by A-KR. Data analysis was performed by A-KR and JLM. The first draft of the manuscript was written by A-KR and TSO, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.
Data Availability
Data, stimuli, and measures for both studies are available at https://researchbox.org/365
Code Availability
Not applicable.
Consent to Participate
Participants provided their consent by reading and agreeing to an informed consent form (presented electronically online) before proceeding to the study.
Consent for Publication
All authors read and approved the final manuscript for submission.
Ethical Approval
Both studies were reviewed by the authors’ Institutional Review Board and were deemed to have exempt status.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Reiman, AK., Ocasio, T.S. & Mezzapelle, J.L. How Cisgender People Define “Transgender” Is Associated with Attitudes Toward Transgender People. Arch Sex Behav 52, 991–1007 (2023). https://doi.org/10.1007/s10508-022-02454-w
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10508-022-02454-w