Citation
Effects of language on anti-transgender attitudes

Material Information

Title:
Effects of language on anti-transgender attitudes
Creator:
Wucherpfennig, Andrea
Place of Publication:
Denver, Colo.
Publisher:
Metropolitan State University of Denver
Publication Date:

Record Information

Source Institution:
Metropolitan State University of Denver
Holding Location:
Auraria Library
Rights Management:
Copyright [name of copyright holder or Creator or Publisher as appropriate]. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item has the following downloads:


Full Text
Effects of Language on Anti-Transgender Attitudes by Andrea Wucherpfennig
An undergraduate thesis submitted in partial completion of the Metropolitan State University of Denver Honors Program
December 2014
Annie Miller
Dr. Robert Schatz Dr. Megan Hughes-Zarzo
Primary Advisor
Second Reader
Honors Program Director


Running head: EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER ATTITUDES
1
Effects of Language on Anti-Transgender Attitudes Andrea J. Wucherpfennig Metropolitan State University of Denver


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
2
Abstract
Language is known to influence perceptions of individuals and social groups, yet no research has investigated how language affects negative attitudes of transgender people. As such, we conducted two studies that examined how the use of two linguistic forms, nouns and adjectives, affects anti-transgender attitudes. In addition, we examined how beliefs in essentialism are related to the use of these linguistic forms and to participants prejudicial attitudes. We used two different experimental procedures to manipulate linguistic form. In Study 1, we manipulated whether the items on an anti-transgender attitude measure referred to transgender individuals using nouns or adjectives. In Study 2 we manipulated whether nouns or adjectives were used to refer to a transgender person in a scenario. Results showed that higher anti-transgender prejudice was found for participants given nouns compared to adjectives, and for high essentialist participants compared to low essentialist participants. Additionally, we found some support for the hypothesis that essentialism moderates the effect of linguistic form on anti-transgender attitudes. High essentialist participants given nouns reported higher anti-transgender attitudes than high essentialist participants given adjectives, but there was no difference in antitransgender attitudes between low essentialist participants given nouns compared to adjectives. These findings are particularly relevant to the transgender communitys current effort to have people refer to transgender individuals with adjectives rather than nouns because they illustrate how prejudicial attitudes can be affected by these two linguistic forms.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
3
Effects of Language on Anti-Transgender Prejudice
The acronym LGBT is well known; however, it is misleading because it groups together gay, lesbian, bisexual, and transgender communities. Gay, lesbian, and bisexual people are sexual minorities because their sexual orientation is other than heterosexual. Despite what the acronym implies, transgender people are not sexual minorities. This is because a persons status as transgender has nothing to do with his or her sexual attraction; instead, it refers to the persons gender identity. Transgender people are gender minorities rather than sexual minorities. Gender minorities are those whose gender identity (identity as a man, woman, or other) and assigned sex (designation at birth as male or female) do not match. Although gender and sexual minorities both experience discrimination and prejudice because of their minority status, they are distinct communities. The acronym LGBT fails to show this difference, which often leads to confusion and misunderstanding.
Public awareness of LGBT issues often focus on sexual minorities issues. The current fight for marriage equality is a prime example of this. Almost all people know about the fight for marriage equality and the majority of Americans have come to support it (Gallop, 2014). Meanwhile, important issues for gender minorities, such as the high costs of hormone therapy and surgeries, are very rarely acknowledged. The exclusion of the transgender community is seen in research as well. For example, of the 49 items on Holland, Matthews, and Schotts (2013) measure of attitudes towards the LGBT community, only three items mention transgender people despite the researchers claims of examining attitudes towards all LGBT people. Additionally, the researchers did not conduct separate analyses on these items to test for differences in attitudes between gender and sexual minorities, but assumed that attitudes towards transgender people would be the same as attitudes towards lesbian, gay, and bisexual people.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
4
Three scales have been developed that assess reactions toward transgender people in particular: The Genderism and Transphobia Scale (GTS; Hill and Willoughby 2005), the Transphobia Scale (Nagoshi et al., 2008), and the Attitudes Toward Transgendered Individuals Scale (ATTI; Walch, Ngamake, Francisco, Stitt & Schingler, 2012). However, the GTS focuses primarily on gender variant behavior (e.g., I cant understand why a woman would act masculine) and the Transphobia Scale focuses primarily on the acceptance of the gender binary (e.g., Apersons genitalia define what gender they are; Apenis defines a person as being a man, a vagina defines a person as being a woman). Only the ATTI measures attitudes toward transgender people specifically (e.g., I would feel uncomfortable if a close family member became romantically involved with a transgendered individual).
The present research examines the impact of language use on anti-transgender prejudice. Language is known to influence perceptions of individuals and social groups, yet no research has investigated how language affects prejudice against transgender people. We conducted two studies that examine how the use of two linguistic forms, nouns and adjectives, affects antitransgender attitudes. This research is especially relevant to the transgender community because there is a current effort within the community to encourage use the term transgender as an adjective (e.g., a transgender person) rather than a noun (e.g., a transgender). In addition, we also examine how beliefs in essentialism are related to the use of these linguistic forms and to participants prejudicial attitudes.
Literature Review: Language, Essentialism, and Prejudice
Previous research suggests that linguistic forms used to describe people can influence perceptions of their characteristics and behavior. An important linguistic theory, the Linguistic Category Model (LCM; Semin & Fiedler, 1988) outlines how four different linguistic forms


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
5
(description action verbs, interpretative actions verbs, state verbs, and adjectives) range on a continuum from most concrete to most abstract. According to the original LCM, adjectives are the most abstract of these forms because they convey qualities of a person rather than specific activities or psychological states. However, research built on the LCM has shown that nouns are even more abstract than adjectives (Carnaghi et al., 2008). Nouns identify the general category to which a person belongs. Whereas, adjectives denote what someone is like, nouns denote the kind of person someone is.
As nouns are the most abstract linguistic form, they are especially likely to elicit essentialist perceptions. Essentialism is a belief that a certain social category has an underlying, identity-determining essence that is permanent and common to all members of that category (Graf, Bilewicz, Finell, & Geschke, 2012, p. 63). Research by Haslam, Rothschild and Ernst (2002) has identified three main aspects of essentialist thinking. One is the belief that there are different types, or categories, of people that exist. As distinct boundaries exist between categories, membership in these categories is considered immutable. Another aspect is the belief that each category has an inherent, biological basis. Last, information can be gleaned simply from knowing the category a person belongs to, because all group members are thought to share uniform character traits.
Research has shown a link between nouns and essentialism. Carnaghi et al. (2008) showed that participants with high essentialist thinking were more likely to choose nouns as descriptors than adjectives. More importantly, the researchers also found that the reverse was true: People given nouns reported more essentialist thinking than people who were given adjectives. Participants were given a short description about a target that used either a noun or an adjective (e.g., Mark is a traditionalist or Mark is a traditional person) and a behavior (e.g.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
6
Mark likes to send postcards home for Christmas). Then, participants were asked to rate the behavior in three areas: strength (e.g., How strong is Marks preference for this activity?), stability (e.g., How likely is it that Marks preference for this activity will remain the same in the next two years?), and resilience (e.g., How likely is it that Marks preference for this activity would remain the same if he was surrounded by friends who did not enjoy the activity in question?). Participants given nouns reported higher strength, stability, and resilience for the behavior than participants given adjectives; that is, they rated the behavior as more essentialist.
Because nouns cause essentialist thinking, and essentialist thinking is the belief that people in certain categories share uniform traits, nouns should also cause stereotyping. Research by Carnaghi et al. (2008) has shown that nouns cause stereotyping in two ways: Nouns facilitate stereotypic-congruent descriptors and inhibit stereotypic-incongruent descriptors. In one of their studies, participants were given two descriptors about a person, one noun and one adjective (e.g., A homosexual, Catholic or A Catholic, homosexual). Then, participants were asked to rate the frequency of behaviors that are stereotypic to each descriptor (e.g., How often does he attend church in a year and How often does he have one-night stands in a year?). Participants estimated higher frequencies of stereotypical behavior for the noun descriptor than for the adjective descriptor. For example, participants given the word Catholic presented as a noun (A Catholic, homosexual) reported a higher frequency of church attendance than participants given Catholic as an adjective (A homosexual, Catholic) because attending church is congruent with the stereotype of Catholics. Additionally, results showed that participants expected the target to display stereotypic-incongruent behaviors less frequently for the noun than for the adjective. For example, participants given Catholic as a noun (A Catholic, homosexual) estimated lower frequencies for one-night stands than participants given


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
7
Catholic as an adjective (A homosexual, Catholic) because one-night stands are incongruent with the stereotype of Catholics.
Because of this link between nouns and stereotyping, it follows that nouns can cause negative attitudes, provided the stereotype is also negative. While there has been no previous research examining this hypothesis, there has been support for a similar theory that nouns elicit more in-group bias than adjectives. Graf et al. (2012) found that participants given nouns to describe a target persons nationality (e.g., A German, a painter) exhibited greater preferences for their own national group than participants who were given adjectives (e.g., A German painter). However, it is important to note that relatively favorable evaluation of an in-group need not involve negative evaluation of an out-group. While in-group bias can provide a foundation for prejudice, it does not necessarily lead to prejudice (Brewer, 1999). Therefore, research comparing nouns and adjectives on prejudicial attitudes is necessary.
Present Research
We tested the impact of linguistic form and essentialist beliefs on anti-transgender prejudice in two studies. In Study 1, we manipulated linguistic form in attitude items that referred to transgender people using either nouns or adjectives (e.g., a transgender, a transgender person). In Study 2, we compared the effects of using nouns and adjectives to describe a transgender person in a scenario. In both studies we predicted that participants given nouns would report higher anti-transgender prejudice than participants given adjectives. We also expected that participants beliefs in essentialism would be associated with more negative attitudes towards transgender people, and that essentialist beliefs would enhance this relationship.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
8
Study 1
In Study 1, participants completed a survey that contained items assessing attitudes toward transgender individuals. Linguistic form was experimentally manipulated by whether the items referred to transgender individuals using nouns (e.g., My friends and I have joked about transgenders) or adjectives (e.g., My friends and I have joked about transgender people). Average level of agreement with these statements was used to measure anti-transgender prejudice. We predicted higher scores for participants in the noun condition than those in the adjective condition (HI).
In order to assess individual differences in essentialism, we included the Essentialist Beliefs scale (Bastian & Haslam, 2006). The measure includes three subscales: Informativeness (e.g., When getting to know a person it is possible to get a picture of the kind of person they are very quickly), Biological Basis (e.g., Whether someone is one kind of person or another is determined by their biological make-up), and Discreteness (e.g., The kind of person someone is, is clearly defined; they either are a certain kind of person or they are not). Based on previous research linking essentialism with prejudicial attitudes towards other stigmatized group members (Howell, Weikum & Dyck, 2010; Haslam, Rothschild & Ernst, 2002), we predicted that high scores on the overall essentialism measure would predict more anti-transgender prejudice (H2). However, we also suspected that the strength of this relationship might vary between subscales. Specifically, we expected Informativeness scores would predict the highest anti-transgender prejudice of the three essentialism subscales (H3). This is because a person high in informative essentialism is likely to draw on the negative stereotype of transgender people to make assumptions about other qualities and characteristics the transgender person possesses. By drawing upon these negative qualities, the participant is then more likely to have a negative


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
9
evaluation of a transgender person. On the other hand, belief in a biological basis for being transgender might not cause negative attitudes towards that person because it may excuse that person for being transgender. This line of thinking has been shown in research on anti-gay attitudes, where belief in biological basis sometimes predicts lower prejudice (Haslam & Levy, 2006). As such, we did not expect a strong relationship between Biological Basis and antitransgender prejudice. Additionally, because a transgender persons outward appearance changes during the process of transitioning, being transgender might not be considered as a clearly defined, distinct category. Therefore we did not expect a strong relationship between Discreteness and anti-transgender prejudice either.
Additionally, we examined whether essentialist beliefs would moderate the relationship between linguistic form and anti-transgender attitudes. Previous research findings linking nouns to essentialism and stereotyping suggests that the predicted effect of linguistic form (i.e., stronger anti-transgender prejudice for nouns than for adjectives) might be especially pronounced for participants with high essentialism scores. If so, there should be a significant interaction between linguistic form and strength of essentialist beliefs (H4). As Informativeness is expected to have an especially strong relationship, we also expected a similar interaction between linguistic form and strength of informative essentialist beliefs (H5).
Finally, we compared the effect of group label transgender or transsexual on attitudes towards transgender people. The term transsexual has a more clinical connotation because it was included in the Diagnostic and Statistical Manual (DSM) in DSM-III (3rd ed; DSM-III; American Psychiatric Association, 1980) and DSM-TTTR (3rd ed., rev; DSM-IIIR; American Psychiatric Association, 1987) as a subtype of Gender Identity Disorder (Gender Identity Disorder/Children Transsexualism). As previous research has linked essentialism with negative


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
10
attitudes toward people with mental illnesses (Howell et al., 2010), this label might be expected to cause higher anti-transgender attitudes than the label transgender. However, transsexual was removed from subsequent versions of the DSM, lessening the clinical connotation of the term. Additionally, transgender currently is used more often, potentially causing the general population to associate more stigma with transgender than transsexual. Given the conflicting arguments, we did not make a formal prediction concerning which term would be associated with more negative attitudes.
Method
Participants. Two hundred thirty-nine participants (147 female, 92 male) were recruited from undergraduate introductory psychology courses at an urban university. The mean age was 23.3 years; 90% of the participants identified as straight, 1.7% identified as lesbian, 2.9% as gay, 3.3% as bisexual, 1.7% as queer, and 0% as questioning/unsure. Eight participants identified as a gender minority (e.g., transgender, transsexual). Fifty-seven percent of the participants identified as White/European-American, 20.4% as Latino(a)/Hispanic-American, 6.3% as Black/African-American, 5.4% as Asian-American, and 10.5% as Other.
Procedure. Small groups of participants completed the survey individually. After giving consent, each participant was given a survey titled Social and Interpersonal Attitudes. Participants were randomly assigned to condition based on which version of the survey they received; as the survey cover sheets were identical, the researcher was blind to condition. After completing the survey, participants were orally debriefed and given course credit as compensation for their time.
Materials and Design. As a distracter, participants first responded to the 18-item Beliefs About Groups scale (BAG; Karau & Elsaid, 2009; e.g., People tend to work especially hard on a


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
11
group task) using a 5-point Likert scale (1- Strongly Disagree, 5- Strongly Agree). The second questionnaire measured participants attitudes towards the transgender community and included the experimental manipulations (explained in the following paragraph). The measure consisted of 21 items (a = .95) adapted from previously established scales including the Transphobia Scale (Nagoshi et al., 2008), the Genderism and Transphobia Scale (GTS; Hill & Willoughby, 2005), and the Attitudes Towards Transgendered Individuals Scale (ATTI; Walch, et al., 2012). Items were chosen that focused on participants attitudes towards transgender individuals, rather than anti-transgender behavior. Participants responded using a 1-Strongly Disagree to 7-Strongly Agree scale. The average of the scores on these items constituted the dependent measure of antitransgender attitude, with a higher score equaling more prejudice.
Participants were randomly assigned to condition in a 2 (Linguistic Form: noun or adjective) x 2 (Group Label: transgender or transsexual) between-participants factorial design based on which version of the anti-transgender attitude measure they received; as the survey cover sheets were identical, the researcher was blind to condition. Participants completed one of four versions of the questionnaire: Noun/Transgender, Noun/Transsexual,
Adjective/Transgender, or Adjective/Transsexual (see Figure 1 for a sample item; see Appendix A for the full measure).
Noun Adjective
Transgender My friends and I have joked about transgenders My friends and I have joked about transgender people
Transsexual My friends and I have joked about transsexuals My friends and I have joked about transsexual people
Figure 1. A sample item from each of the four versions of the anti-transgender attitudes measure.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
12
Next was the Essentialist Beliefs scale (Bastian & Haslam, 2006; a = .78) which included three subscales: Biological Basis (8 items; a = .71; e.g., Whether someone is one kind of person or another is determined by their biological make-up), Discreteness (8 items; a = .65; e.g., The kind of person someone is, is clearly defined; they either are a certain kind of person or they are not), and Informativeness (7 items; a = .63; e.g., When getting to know a person it is possible to get a picture of the kind of person they are very quickly). Participants responded using a 6-point Likert scale anchored from 1-Strongly Disagree to 6- Strongly Agree. Lastly, participants provided basic demographic information including sexual orientation, ethnic-racial background, and identification as a gender minority (e.g., transgender, transsexual).
Results
We first performed a median split on participants essentialism scores to create high essentialism and low essentialism groups. A 4-way Linguistic Form (noun or adjective) x Group Label (transgender or transsexual) x Essentialism (high or low) x Sex (male or female) between-participants Analysis of Variance (ANOVA) was conducted on the anti-transgender attitude scores. Results indicated significant1 main effects for linguistic form, F(l, 239) = 8.112, p = 0.005, rjp2= 0.035, essentialism group, F(l, 239) = 12.876, p < 0.001, rjp2 = 0.055, and sex, F(l,
239) = 25.447, p < 0.001, rjp2 = 0.102. As predicted, significantly higher anti-transgender
prejudice was found for nouns than for adjectives (HI; See Figure 2; Ms = 3.249 and 2.809) and for high essentialist participants compared to low essentialist participants (H2; See Figure 3; Ms = 3.306 and 2.752). In addition, males reported significantly higher anti-transgender prejudice than females (Ms = 3.419 and 2.639). No main effect was found for group label (p = 0.46).
1 In our study, we considered significant effects to havep < .05 and marginal effects to have p < 10, two-tailed


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
13
However, there was a marginally significant Linguistic Form x Group Label x Essentialism interaction, U(l,239) = 3.350,p = .069, rfp2= .015.
Effect of Linguistic Form on Anti-Transgender Attitudes
4
O
£ 3.5
Nouns Adjectives
Figure 2. Results indicated a main effect for Linguistic Form: Participants given nouns reported more negative anti-transgender attitudes than participants given adjectives.
Effect of Essentialism on Anti-Transgender Attitudes
4
WJ 3.5
"3
<
-3
-5
CD
3
H 2.5
2
High E ssentialist Low E ssentialist
Figure 3. Results indicated a main effect for Essentialism: High essentialist participants reported more negative anti-transgender attitudes than low essentialist participants.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
14
As the 4-way ANOVA produced no significant interactions with sex, we reanalyzed the anti-transgender attitudes scores in a 3-way Linguistic Form (noun or adjective) x Group Label (transgender or transsexual) x Essentialism (high or low) ANOVA. The linguistic form and essentialism main effects remained significant in this analysis (ps < .001). In addition, the 3-way interaction reached statistical significance, F( 1, 239) = 3.876,p = .05, Tjp2 = .017. To explore the
3-way interaction, 2-way Linguistic Form x Essentialism ANOVAs were performed separately for participants in each group label condition. A marginally significant interaction was found for participants exposed to the label transsexual (H4), F (1, 239) =3.217,/) = .075, rjp2= .027.
This interaction is displayed in Figure 4. Simple effect tests showed that anti-transgender prejudice scores were significantly higher in the noun condition than in the adjective condition for participants who scored high in essentialism, F (1, 239) = 6.899, p = .01, rjp2= .055 (M =
3.617 and 2.813). However, there was no difference in anti-transgender prejudice between the noun and adjective conditions for participants who scored low in essentialism, F (1, 239) = .007,
p = .932.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
15
Anti-Transgender Attitudes of Participants in the "Transsexual" Group Label Condition 4 n
M 3.5
0
si 5 M 2.5
j
<
1 2
r"
S3
| 1-5
s
S 1
^ 0.5 0
Nouns Adjectives
Figure 4. High essentialist participants given nouns reported significantly higher antitransgender attitudes than high essentialist participants given adjectives; however, there was no difference in attitudes for low essentialist participants given nouns compared to adjectives.
We then conducted separate Linguistic Form (noun or adjective) x Group Label (transgender or transsexual) x Essentialism (high or low) ANOVAs that compared antitransgender attitudes for high and low scorers on each essentialism subscale. We performed median splits on each subscale to create high-essentialist and low-essentialist groups. The main effects for essentialism and linguistic form were significant or marginally significant in each analysis (ps < .052): Nouns and higher essentialism scores predicted more negative antitransgender attitudes. Contrary to our prediction, the relationship between essentialism and antitransgender prejudice was weakest for the Informativeness subscale (H3), F (1, 239) = 4.918, p = .064, r\p2 = .015, rather than Biological Basis, A (1, 239) = 3.824,p = .052, r\p2 = .018, or
Discreteness, A (2, 239) = 14.498,/? < .001, r\p2 = .065.
High Essentialism Low Essentialism


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
16
However, in support of prediction H5, the Linguistic Form x Group Label x Essentialism interaction was significant only for informativeness, A (1, 239) = 5.187,/? = .024, Tjp2 = .022, not
for the biological basis, A (1, 239) = 2.352,/? = .127, Tjp2 = .011 or discreteness A (1, 239) = .808,/? = .37, Tjp2 = .004. Analyses of anti-transgender attitudes in the transsexual label condition that compared high and low scorers on informativeness yielded the same pattern of results found for scores on the overall essentialism measure: a significant Linguistic Form x Informativeness interaction, F (1, 239) = 7.573,/? = .007, Tjp2 = .060, and significantly higher prejudice for nouns
relative to adjectives for participants high in informativeness {Ms = 3.509, 2.732) but not for participants low in informativeness {Ms = 2.976, 2.684). This interaction is displayed in Figure 5.
4 1 M 3.5
O
L_
0
O Q
m j
3 2.5 -
-4->
>
<
1 2
C"
S3
SM-5 -
c3
g i -
^ 0.5 -0 -
Anti-Transgender Attitudes
High Informative Essentialism
Low Informative Essentialism
Nouns
------------------1
Adjectives
Figure 5. Informative essentialism follows the same pattern as overall essentialism: High essentialist participants given nouns reported significantly higher anti-transgender attitudes than high essentialist participants given adjectives, but there was no difference in attitudes for low essentialist participants given nouns compared to adjectives.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
17
Discussion
The results of Study 1 supported the hypothesis that the use of nouns to refer to members of the transgender community causes more negative attitudes toward these individuals than does the use of adjectives (HI). We also found support for the hypothesis that essentialism moderates this effect (H4). For participants exposed to the group label transsexual, high essentialist participants reported significantly more negative attitudes in the noun condition than in the adjective condition; however this difference was not significant for low essentialist participants.
It is not clear why this interaction was found only in the transsexual condition. One possible explanation may involve the links between linguistic form, essentialism, and attitudes toward mental illness. As mentioned before, previous research has shown that the use of nouns predicts more essentialist thinking than the use of adjectives (Carnaghi et al., 2008), and that essentialist thinking predicts prejudicial attitudes towards people with mental illnesses (Howell et al., 2010). Given the inclusion of transsexual in the DSM, it may be that nouns elicited more antitransgender attitudes among high essentialist scorers only when this clinical term was used as a group label.
When we examined the subscales of essentialism separately, a significant interaction with linguistic form was found only in the ANOVA that compared high and low scorers on informativeness (H5). Informative essentialism claims that a persons characteristics and qualities can be inferred from knowledge of that persons group membership. This ideology is the foundation for stereotyping. Since the stereotype of transgender people is negative, it should generate prejudice. Because nouns are essentialist, they would have increased stereotyping among participants high in informativeness and caused significantly higher anti-transgender attitudes compared to adjectives.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
18
We conducted an additional study to further explore the effects of linguistic form and essentialism on anti-transgender prejudice. For Study 2, we examined these relationships with regard to a transgender person depicted in a scenario. We were interested to see if the effect of linguistic form obtained in Study 1 would generalize to this situation. Additionally, we designed Study 2 to test whether beliefs in essentialism mediate the effect of linguistic form on antitransgender attitudes.
Study 2
In Study 2, participants read a scenario that depicted the events surrounding a transgender persons decision to transition to another gender. In order to test the effect of linguistic form on anti-transgender attitudes, the target was described with either nouns or adjectives. After reading the scenario, participants responded to modified versions of the essentialism and anti-transgender attitude measures used in Study 1 that referred specifically to the target person. We predicted that nouns would induce more negative anti-transgender attitudes than adjectives (HI), and that essentialist beliefs about the target (H2), particularly informativeness (H3), would mediate this difference.
In addition to linguistic mode, we also examined whether transition type (male-to-female or female-to-male) affected participants anti-transgender attitudes towards the target. No previous research has examined attitudes towards transition types; however, there has been research on the experiences of transgender people. A survey from the National Center for Transgender Equality and the National Gay and Lesbian Task Force (Grant et al., 2011) found that transgender men (those who transition from female-to-male) reported higher frequencies of harassment and bullying in their K-12 schools years. However, transgender women (those who transition from male-to-female) reported more discrimination after age 18, including higher rates


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
19
of discrimination from their medical providers, more relationships ending due to transitioning, more access denied to shelters, and almost twice as many experiences being physically assaulted compared to transgender men. The fact that transgender women face so much more discrimination in their lives than transgender men likely reflects a societal prejudicial attitude towards them. Therefore, we predicted higher anti-transgender attitudes for participants who evaluated a person transitioning from male-to-female (transgender woman) than for participants who evaluated a person transitioning from female-to-male (transgender man; H4).
Method
Participants. Two hundred and sixty-nine participants (179 female, 90 male) were recruited from undergraduate introductory psychology courses at an urban university. The mean age was 22.23 years; 89.5% of the participants identified as straight, 2% identified as lesbian, 2% as gay, 5.4% as bisexual, 0% as queer, and 1% as questioning/unsure. One participant identified as a gender minority (e.g., transgender, transsexual). Fifty-four percent of participants identified as White/European-American, 27.8% as Latino(a)/Hispanic-American, 5.6% as Black/African-American, 4.4% as Asian-American, and 8.1% as Other. We excluded 6 participants from our analyses because they clearly did not take the survey seriously.
Procedures. Small groups of participants completed the surveys individually2. After giving consent, each participant was given a survey titled Group and Interpersonal Attitudes. Participants were randomly assigned to condition based on which version of the survey they received; as the survey cover sheets were identical, the researcher was blind to condition. After completing the survey, participants were orally debriefed and given course credit as compensation for their time.
2 Participants were recruited from the same source as Study 1. However, restrictions were put in place so that students were unable to participate in Study 2 if they had participated in Study 1.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
20
Materials and Design. As with Study 1, participants first responded to the 18-item distractor questionnaire, Beliefs About Groups (BAG; Karau & Elsaid, 2009). Following the distractor, participants read a scenario of approximately 250 words that described a transgender person Jessie. Jessie is described as identifying as another gender but hiding his or her gender from childhood through high school graduation. In college, Jessie experiments with gender presentation when alone, but is afraid to tell anyone else. Finally, the scenario ends with Jessie coming out to friends and changing the sex markers on official IDs.
Linguistic form and transition type were manipulated in a 2 (noun or adjective) x 2 (female-to-male or male-to-female) between-participants factorial design. Participants received one of four versions of the scenario (see Figure 6 for a sample of each scenario; see Appendix B for the entire scenario): Noun/Female-to-male, Noun/Male-to-Female, Adjective/Female-to-Male, or Adjective/Male-to-Female.
Noun Adjective
Female-to- Male As Jessie grew up, she often felt that she was a boy. However, she remembered how negatively her parents reacted to transgenders. As Jessie grew up, she often felt that she was a boy. However, she remembered how negatively her parents reacted to transgender people.
Male-to- Female As Jessie grew up, he often felt that he was a girl. However, he remembered how negatively his parents reacted to transgenders. As Jessie grew up, he often felt that he was a girl. However, he remembered how negatively his parents reacted to transgender people.
Figure 6. A sample from each of the four versions of the scenarios.
After reading the scenario, participants responded to a modified version of the Essentialist Beliefs scale (Bastian & Haslam, 2006; a = .78) that contained three subscales used in Study 1: Biological basis (a = .87), Discreteness (a = .74), and Informativeness (a = .65). Based on reliability analyses on the Discreteness subscale, 3 of 8 items were removed. All


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
21
essentialist items were modified to apply to the target Jessie or to people like Jessie (e.g., Jessies basic character is never easily defined; People are either like Jessie or they are not). Participants responded using a 6-point Likert scale (1- Strongly Disagree, 6- Strongly Agree). Then, participants responded to a modified version of the anti-transgender attitudes measure from Study 1 (a = .92). As with the Essentialist Beliefs scale, anti-transgender attitude items were modified to apply to the target Jessie or to people like Jessie (e.g., Jessie should not be allowed to work with children; People like Jessie endanger the institution of the family). The scale was anchored from 1- Strongly Disagree to 7- Strongly Agree. The dependent variable was the average scores on this measure, with higher scores equaling more prejudice. Last, participants provided the same basic demographic information asked in the first study.
Results
We first ran a Linguistic Form (noun or adjective) x Transition Type (male-to-female or female-to-male) x Sex (male or female) ANOVA on the anti-transgender attitude scores.
Contrary to the prediction that anti-transgender attitudes would be more negative for the male-to-female transition target than for the female-to-male transition target (H4), there were no significant effects for transition type. Because transition type did not have any significant effects, we reanalyzed the data in a Linguistic Form (noun or adjective) x Sex (male or Female)
ANOVA. There was a significant effect for sex, F( 1, 269) = 30.922, p < .001, t]p2= .104 that
showed males reported higher anti-transgender prejudice than females (Ms =2.980, 2.336). The effect for linguistic form was marginally significant (see Figure 7), F(1, 269) = 3.067,p= .081, rjp2= .011. As predicted (HI), participants given nouns reported higher anti-transgender attitudes than participants given adjectives (Ms = 2.753, 2.563). In addition, there was a significant Linguistic Form x Sex interaction, F( 1, 269) = 3.983,/) = .047, rjp2 = .015. Further


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
22
analysis of this interaction showed that males given nouns reported higher anti-transgender prejudice than males given adjectives {Ms = 3.293 and 2.807); however, no significant difference was shown for females given nouns compared to adjectives {Ms = 2.313 and 2.345).
Effect of Linguistic Form on Anti-Transgender Attitudes
3
o
u
5*2
U
-a
2 2.5
<
o
-a
(u
20
c 2
1.5
Nouns Adjectives
Figure 7. Results indicated a marginally significant main effect for linguistic form: Participants given nouns reported more negative anti-transgender attitudes than participants given adjectives.
To examine the possibility that essentialist beliefs about the target mediated the relationship between linguistic form and anti-transgender attitudes, we first tested whether linguistic form predicted scores on the overall essentialism measure. A one-way ANOVA comparing the scores of participants in the noun and adjective conditions was not significant, F{ 1, 268) = .938,p= .334. A Linguistic Form (noun or adjective) x Sex (male or female) ANOVA also found no effect for linguistic form, F{ 1, 265) = 1.896, p = .170. Since linguistic form was not shown to predict essentialism scores, essentialism could not mediate the effect of linguistic form on anti-transgender prejudice (H2, H3).


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
23
With no evidence for mediation, we decided to test if essentialism moderated the effect of linguistic form on anti-transgender attitudes as it had in Study 1. In order to test moderation, we ran a Linguistic Form (noun or adjective) x Sex (male or female) x Essentialism (high or low) ANOVA comparing anti-transgender attitudes for high and low scorers on the overall essentialism measure. Results showed significant effects for sexU(l, 269) = 26.318,/) < .001, rjp2= .092 and essentialism F(l, 269) = 13.464,p < .001, Tjp2 = .049. Higher anti-transgender
attitudes were found for males than females (Ms = 3.082, 2.33) and for high essentialist participants compared to low essentialist participants (see Figure 8; Ms = 2.895, 2.426).
Although the predicted Linguistic Form x Essentialism interaction was not significant, F( 1, 269), = 1.991,/) = .159, rjp2= .008, the mean scores were in the predicted directions (See Figure 9).
Simple effect tests showed high essentialist participants given nouns reported higher antitransgender prejudice than high essentialist participants given adjectives, F( 1, 261) = 3.846,/) = .051 (Ms = 3.059, 2.766); however, there was no difference in anti-transgender attitudes for low essentialist participants given nouns versus adjectives, F( 1, 261) = .008,/) = .930 (Ms = 2.324, 2.373).


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
24
Effect of Essentialism on Anti-Transgender Attitudes
QJ
"3
5 2.5
<
QJ
"3
CD
S 2
H
1.5
High Essentialism Low Essentialism
Figure 8. Results indicated a main effect for Essentialism: High essentialist participants reported more negative anti-transgender attitudes than low essentialist participants.
Anti-Transgender Attitudes
3.5 -|
*§ 0.5 -
0
Nouns
------------------1
Adjectives
High Essentialism Low Essentialism
Figure 9. Although the interaction is not significant, the mean score trends mimic Study 1: High essentialist participants given nouns reported significantly higher anti-transgender attitudes than high essentialist participants given adjectives; however, no significant difference was found for low essentialist participants given nouns compared to adjectives.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
25
We then conducted separate Linguistic Form (noun or adjective) x Sex (male or female) x Essentialism (high or low) ANOVAs that compared anti-transgender attitudes for high and low scorers on each essentialism subscale. Main effects for sex were significant for all subscales (p < .001) and main effects for essentialism were significant for both informativeness and discreteness (p < .001). In addition, there was a Linguistic Form x Sex interaction that was significant for informativeness, F( 1, 269) = 3.903,p = .049, rjp2= .015, and marginally
significant for biological basis, F( 1, 269) = 3.192,/) = .075, rjp2= .004. This interaction showed
that males given nouns reported higher anti-transgender prejudice than males given adjectives (Informativeness: Ms = 3.202, 2.757; Biological Basis: Ms = 3.189, 2.747); however, there was no difference for females given nouns and adjectives (Informativeness: Ms = 2.304, 2.369; Biological Basis: Ms = 2.309, 2.372).
The Linguistic Form x Essentialism interaction was not significant for any subscale: Informativeness, F(1, 269) = 2.201 ,p= .139, rjp2= .008, Biological Basis, F( 1, 269) = .210,p =
.647, rjp2= .000, or Discreteness, F( 1, 269) = 1.023,p = .313, rjp2= .004. However, simple effect
tests showed high scorers on informativeness given nouns reported significantly higher antitransgender attitudes than high essentialist participants given adjectives, F( 1, 269) = 4.37,p = .038, rjp2= .016 (Ms = 3.16, 2.795) but there was no difference in anti-transgender attitudes for low essentialist participants given nouns and adjectives, F(1, 269) = 0.00,p = .984 (Ms = 2.389, 2.392). This interaction can be seen in Figure 4. Anti-transgender attitudes also tended to be higher in the noun condition than in the adjective condition for high scorers on the other two essentialism subscales, although the difference was not significant for biological basis (p = .158) and was marginally significant for discreteness (p = .061). There was no significant difference in


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
26
anti-transgender attitudes between participants given nouns and adjectives for low scorers on either sub scale (ps > .48).
3.5 -|
Ifl p 3
o
CA 2.5 -
3
w > < 2 -
a3
r- 5 CD iri 1.5
S3 1
H 1
w
1 0.5
0 -
Anti-Transgender Attitudes
High Informative Essen tialism
Low Informative Essen tialism
Nouns
------------------1
Adjectives
Figure 10. Informative essentialism again follows the same pattern as overall essentialism: High essentialist participants given nouns reported significantly higher anti-transgender attitudes than high essentialist participants given adjectives; however, no significant difference was found for low essentialist participants given nouns compared to adjectives.
Discussion
We found some support for the hypothesis that nouns generate more anti-transgender prejudice than adjectives (HI), although the difference was marginally significant across all participants and statistically significant only for males. The results did not support our hypothesis that essentialism would mediate the effect of linguistic form on anti-transgender attitudes (H2). Interactions that tested whether essentialist beliefs moderated the effect of linguistic form on anti-transgender attitudes were not significant, however the trends were in the predicted direction and the difference between scores in the noun and adjective condition was significant for


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
27
informativeness. As mentioned in Study 1, participants high in informative essentialism infer negative characteristics from the negative stereotype of transgender people which causes more negative attitudes towards the target. Since nouns are essentialist, they increased stereotyping among these individual to cause significantly higher anti-transgender attitudes compared to adjectives.
Additionally, results from Study 2 yielded significant effects for sex. The main effect for sex was found in multiple ANOVAs, showing that males reported higher anti-transgender attitudes than females. Not only was this main effect present, but it had the largest effect3. This difference is consistent with previous prejudice literature that shows males are more prejudiced than females (Hill & Willoughby, 2005; Willoughby, Hill & Gonzalez, 2010; Nagoshi, et al. 2008; Lippa & Arad, 1999; Jewell & Morrison, 2012). Additionally, there was a significant interaction that showed males given nouns reported significantly higher anti-transgender attitudes than males given adjectives, but no differences in anti-transgender attitudes for females given nouns compared to adjectives. We did not predict this finding and we are not aware of any previous research that has examined the interactive effect of linguistic form and sex on expressions of prejudice. We suspect that the difference might be due to the fact that males in our study scored higher on essentialism than females. As nouns predicted more prejudice than adjectives for participants who scored high on essentialism, males exposed to nouns were more prejudiced than males exposed to adjectives. Future research should further examine this relationship.
General Discussion
Our hypothesis that participants given nouns would report higher anti-transgender
3 This effect size was considered medium. All other effects were considered small according to Gray & Kinnear (2012).


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
28
attitudes than participants given adjectives was supported by two separate studies. These two studies also supported our hypothesis that essentialism predicts anti-transgender attitudes. When we examined how the linguistic form and essentialism interact, we found support for our prediction that essentialism would moderate the effect of linguistic form on anti-transgender prejudice (although the results were stronger in Study 1 than Study 2). In Study 1, for participants exposed to the group label transsexual, high essentialist participants given nouns reported higher anti-transgender prejudice than high essentialist participants given adjectives, but there was no significant difference for low essentialist participants. This interaction was significant for the overall essentialism measure and for informative essentialism. The same pattern of findings was obtained for both measures in Study 2 (higher prejudice for nouns than for adjectives among high essentialist participants but not low essentialist participants) although the interaction effects were not statistically significant.
Participants in our research were relatively young, primarily white undergraduates at a public university. Because of these characteristics and the urban setting of the campus, our participants probably were more liberal in their political beliefs and more tolerant of transgender people than the average United States citizen. Future research should be done with a wider demographic to assess the generalizability of our results. Additionally, as with any experiment, the artificial nature of the study can bring external validity into question. We attempted to combat this in Study 2 by creating a scenario about a transgender person, in order to create a more realistic situation. Future research should continue along this path to see if these effects continue to hold in more generalizable contexts.
Research Applications
The present research makes an important contribution to the literature on the effects of


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
29
language on prejudice. While previous research has established a link between linguistic form and essentialism, and has shown that linguistic form can affect stereotyping and in-group bias, our research is the first to demonstrate that linguistic form directly impacts prejudicial attitudes. By evaluating the effects of linguistic form on attitudes towards transgender people, our research also adds to the growing literature on anti-transgender prejudice.
Our research also has important practical implications outside of the academic community. The transgender community asks people to use the term transgender as an adjective instead of a noun and our research supports the merit of this aim. For example, our findings suggest that a news article that refers to a transgender person as a transgender (i.e., using a noun) may elicit more anti-transgender prejudice among readers than a comparable article that refers to a transgender person (i.e., using an adjective). Although the effects we found for linguistic form are small in size, repeated exposure to noun references in the media might reliably increase prejudice that hinders this already marginalized community in its fight for awareness and acceptance. As language both inside and outside of the transgender community continues to evolve, research should continue to examine its impact on prejudicial attitudes. In this way, we may lighten the load of marginalized groups and assist in their fights for equality.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
30
References
American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd ed.) Washington, DC: Author.
American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed., rev.) Washington, DC: Author.
Bastian, B. & Haslam, N. (2006). Psychological essentialism and stereotype endorsement. Journal of Experimental Social Psychology (42) 2. doi: 10.1016/j.jesp.2005.03.003
Brewer, M. (1999). The psychology of prejudice: Ingroup love and outgroup hate? Journal of Social Issues (55)3. doi: 10.1111/0022-4537.00126
Camaghi, A., Maass, A., Gresta, S., Bianchi, M., Cadinu, M. & Arcuri, L. (2008). Nomina sunt omina: On the inductive potential of nouns and adjectives in person perception. Journal of Personality and Social Psychology (94)5. doi: 10.1037/0022-3514.94.5.839
Gallup. (2014). Marriage: Do you think marriages between same-sex couples should or should not be recognized by the law as valid, with the same rights as traditional marriages?
May 8-11, 2014 [Survey Report], Retrieved from http://www.gallup.com/poll/117328/ma rriage.aspx
Graf, S., Bilewicz, M., Finella, E. & Geschke, D. (2013). Journal of Language and Social Psychology (32)62. doi: 10.1177/0261927X12463209
Grant, J., Mottet, L., Tanis, J., Harrison, J., Herman, J. & Keisling, M. (2011). Injustice at every turn: A report of the national transgender discrimination survey. Washington: National Center for Transgender Equality and National Gay and Lesbian Task Force.
Gray, C.D. & Kinnear, P.R. (2012). IBM SPSS Statistics 19 Made Simple. Hove and New York: Psychology Press.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
31
Haslam, N. & Levy, S. (2006). Essentialist beliefs about homosexuality: Structure and
implications for prejudice. Personality and Social Psychology Bulletin (32)4. doi:
10.1177/0146167205276516
Haslam, N., Rothschild, L. & Ernst, D. (2002). British Journal of Social Psychology (41)1. doi: 10.1348/014466602165072
Hill, D. & Willoughby, B. (2005). The development and validation of the genderism and transphobia scale. Sex Roles (53)1. doi: 10.1007/sl 1199-005-7140-x
Holland, Matthews, Schott (2013). Thats so gay!: Exploring college students attitudes toward the LGBT population. Journal of Homosexuality (60)4. doi: 10.1080/00918369.2013.760321
Howell, A., Weikum, B. & Dyck, H. (2011). Psychological essentialism and its association with stigmatization. Personality and Individual Differences (50) 1. doi: 10.1016/j.paid.2010.09.006
Jewell, L. & Morrison, M. (2012). Making sense of homonegativity: Heterosexual men and womens understanding of their own prejudice and discrimination toward gay men. Qualitative Research in Psychology (9)4. doi: 10.1080/14780887.2011.586098
Karau, S. & Elsaid, A. (2009). Individual differences in beliefs about groups. Group Dynamics: Theory, Research, and Practice (13)1. doi: 10.1037/a0013366
Lippa, R. & Arad, S. (1999). Gender, personality, and prejudice: The display of
authoritarianism and social dominance in interviews with college men and women. Journal of Research in Personality (33)4. doi: 10.1006/jrpe. 1999.2266


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
32
Nagoshi, J., Adams, K., Terrell, H., Hill, E., Brzuzy, S., & Nagoshi, C. (2008). Gender
differences in correlates of homophobia and transphobia. Sex Roles (59) 521-531. doi:
10.1007/sl 1199-008-94587-7
Semin, G. & Fiedler, K. (1991). European Review of Social Psychology (2)1. doi:
10.1080/14792779143000006
Walch, S., Ngamake, S., Francisco, J., Rashunda, S. & Shingler, K. (2012). Archives of Sexual Behavior (41)5. doi: 10.1007/sl0508-012-9995-6
Willoughby, B., Hill, D., & Gonzalez, C. (2010). Who hates gender outlaws? A multisite and
multinational evaluation of the genderism and transphobia scale. International Journal of Transgenderism (12)4. doi: 10.1080/15532739.2010.55082


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
33
Appendix A N oun/T ransgender
I would be upset if someone Id known for a long time revealed to me that they are a transgender.
I think there is something wrong with a person who says they are a transgender.
I dont like it when someone is flirting with me and I can tell they are a transgender.
I think there is something wrong with a person who says that they are a transgender.
If I found out my lover was a transgender, I would freak out.
I would avoid talking to someone if I knew they were transgenders.
I would go to a bar that was frequented by transgenders.
My friends and I have joked about transgenders.
If I found out that my best friend was a transgender, I would freak out.
Transgenders should not be allowed to work with children.
Transgenders are a viable part of our society.
Transgenders endanger the institution of the family.
Transgenders should be accepted completely into our society.
I avoid transgenders whenever possible.
I would enjoy attending social functions at which transgenders are present.
I would feel comfortable if I learned my neighbor was a transgender.
I would like to have friends who are transgenders.
I would feel uncomfortable if a close family member became romantically involved with a transgender.
Romantic partners of transgenders should seek psychological treatment.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
34
It would be beneficial to society to recognize transgenders as normal.
I would feel comfortable working closely with a transgender.
N oun/T rans sexual
I would be upset if someone Id known for a long time revealed to me that they are transsexual. I think there is something wrong with a person who says they are a transsexual.
I dont like it when someone is flirting with me and I can tell they are a transsexual.
I think there is something wrong with a person who says that they are a transsexual.
If I found out my lover was a transsexual, I would freak out.
I would avoid talking to someone if I knew they were a transsexual.
I would go to a bar that was frequented by transsexuals.
My friends and I have joked about transsexuals.
If I found out that my best friend was a transsexual, I would freak out.
Transsexuals should not be allowed to work with children.
Transsexuals are a viable part of our society.
Transsexuals endanger the institution of the family.
Transsexuals should be accepted completely into our society.
I avoid transsexuals whenever possible.
I would enjoy attending social functions at which transsexuals are present.
I would feel comfortable if I learned my neighbor was a transsexual.
I would like to have friends who are transsexuals.
I would feel uncomfortable if a close family member became romantically involved with a
transsexual.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
35
Romantic partners of transsexuals should seek psychological treatment.
It would be beneficial to society to recognize transsexuals as normal.
I would feel comfortable working closely with a transsexual.
Adj ective/T ransgender
I would be upset if someone Id known for a long time revealed to me that they are transgender. I think there is something wrong with a person who says they are transgender.
I dont like it when someone is flirting with me and I can tell they are transgender.
I think there is something wrong with a person who says that they are transgender.
If I found out my lover was transgender, I would freak out.
I would avoid talking to someone if I knew they were transgender.
I would go to a bar that was frequented by transgender people.
My friends and I have joked about transgender people.
If I found out that my best friend was transgender, I would freak out.
Transgender people should not be allowed to work with children.
Transgender people are a viable part of our society.
Transgender people endanger the institution of the family.
Transgender people should be accepted completely into our society.
I avoid transgender people whenever possible.
I would enjoy attending social functions at which transgender people are present.
I would feel comfortable if I learned my neighbor was transgender.
I would like to have friends who are transgender people.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
36
I would feel uncomfortable if a close family member became romantically involved with a transgender person.
Romantic partners of transgender people should seek psychological treatment.
It would be beneficial to society to recognize transgender people as normal.
I would feel comfortable working closely with a transgender person.
Adj ective/Transsexual
I would be upset if someone Id known for a long time revealed to me that they are transsexual. I think there is something wrong with a person who says they are transsexual.
I dont like it when someone is flirting with me and I can tell they are transsexual.
I think there is something wrong with a person who says that they are transsexual.
If I found out my lover was transsexual, I would freak out.
I would avoid talking to someone if I knew they were transsexual.
I would go to a bar that was frequented by transsexual people.
My friends and I have joked about transsexual people.
If I found out that my best friend was transsexual, I would freak out.
Transsexual people should not be allowed to work with children.
Transsexual people are a viable part of our society.
Transsexual people endanger the institution of the family.
Transsexual people should be accepted completely into our society.
I avoid transsexual people whenever possible.
I would enjoy attending social functions at which transsexual people are present.
I would feel comfortable if I learned my neighbor was transsexual.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
37
I would like to have friends who are transsexual people.
I would feel uncomfortable if a close family member became romantically involved with a transsexual person.
Romantic partners of transsexual people should seek psychological treatment.
It would be beneficial to society to recognize transsexual people as normal.
I would feel comfortable working closely with a transsexual person.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
38
Appendix B N oun/Male-to-F emale
As Jessie grew up, he often felt that he was a girl. However, he remembered how negatively his parents reacted to transgenders. Because of this, Jessie hid his feelings and made sure never to act like he was a transgender. He made sure to wear boy clothes and act like the other boys around him. Throughout middle school and high school he acted as masculine as he could, so others would not suspect him of being a transgender.
After graduating high school, Jessie moved out of state for college. He lived in the dorms with a roommate, but sometimes when his roommate was gone Jessie would acknowledge his feelings as a transgender. He would lock the door and dress up in womens clothing. He loved wearing dresses and jewelry but he was afraid to tell anyone. He didnt want to come out as a transgender because he was afraid that his friends would make fun of him or even beat him up.
In his last year in college, Jessie took a course on gender identification. After learning about transgenders, he began to wonder more about his gender identity. He presented more and more as a woman, sometimes even going out to bars dressed as a woman. The more he dressed as a woman, the more certain he became that he was a transgender. Jessie started dressing in womens clothing full-time, and decided to legally change the gender listed on official IDs and forms. Jessie also some told friends about being a transgender, and asked to be referred as she instead of he.
Noun/Female-to-Male
As Jessie grew up, she often felt that she was a boy. However, she remembered how negatively her parents reacted to transgenders. Because of this, Jessie hid her feelings and made


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
39
sure never to act like she was a transgender. She made sure to wear girl clothes and act like the other girls around her. Throughout middle school and high school she acted as feminine as she could, so others would not suspect her of being a transgender.
After graduating high school, Jessie moved out of state for college. She lived in the dorms with a roommate, but sometimes when her roommate was gone Jessie would acknowledge her feelings as a transgender. She would lock the door and dress up in mens clothing. She loved wearing suits and ties but she was afraid to tell anyone. She didnt want to come out as a transgender because she was afraid that her friends would make fun of her or even beat her up.
In her last year in college, Jessie took a course on gender identification. After learning about transgenders, she began to wonder more about her gender identity. She presented more and more as a man, sometimes even going out to bars dressed as a man. The more she dressed as a man, the more certain she became that she was a transgender. Jessie started dressing in mens clothing full-time, and decided to legally change the gender listed on official IDs and forms. Jessie also told some friends about being a transgender, and asked to be referred as he instead of she.
Adj ective/Male-to-F emale
As Jessie grew up, he often felt that he was a girl. However, he remembered how negatively his parents reacted to transgender people. Because of this, Jessie hid his feelings and made sure never to act like he was transgender. He made sure to wear boy clothes and act like the other boys around him. Throughout middle school and high school he acted as masculine as he could, so others would not suspect him of being transgender.


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
40
After graduating high school, Jessie moved out of state for college. He lived in the dorms with a roommate, but sometimes when his roommate was gone Jessie would acknowledge his feelings as transgender. He would lock the door and dress up in womens clothing. He loved wearing dresses and jewelry but he was afraid to tell anyone. He didnt want to come out as transgender because he was afraid that his friends would make fun of him or even beat him up.
In his last year in college, Jessie took a course on gender identification. After learning about transgender people, he began to wonder more about his gender identity. He presented more and more as a woman, sometimes even going out to bars dressed as a woman. The more he dressed as a woman, the more certain he became that he was transgender. Jessie started dressing in womens clothing full-time, and decided to legally change the gender listed on official IDs and forms. Jessie also told some friends about being transgender, and asked to be referred as she instead of he.
Adj ective/Female-to-Male
As Jessie grew up, she often felt that she was a boy. However, she remembered how negatively her parents reacted to transgender people. Because of this, Jessie hid her feelings and made sure never to act like she was transgender. She made sure to wear girl clothes and act like the other girls around her. Throughout middle school and high school she acted as feminine as she could, so others would not suspect her of being transgender.
After graduating high school, Jessie moved out of state for college. She lived in the dorms with a roommate, but sometimes when her roommate was gone Jessie would acknowledge her feelings as transgender. She would lock the door and dress up in mens clothing. She loved


EFFECTS OF LANGUAGE ON ANTI-TRANSGENDER PREJUDICE
41
wearing suits and ties but she was afraid to tell anyone. She didnt want to come out as transgender because she was afraid that her friends would make fun of her or even beat her up.
In her last year in college, Jessie took a course on gender identification. After learning about transgender people, she began to wonder more about her gender identity. She presented more and more as a man, sometimes even going out to bars dressed as a man. The more she dressed as a man, the more certain she became that she was transgender. Jessie started dressing in mens clothing full-time, and decided to legally change the gender listed on official IDs and forms. Jessie also told some friends about being transgender, and asked to be referred as he instead of she.


Full Text

PAGE 1

Effects of Language on Anti Transgender Attitudes by Andrea Wucherpfennig An undergraduate thesis submitted in partial completion of the M etropolitan State University of D enver Honors Program December 2014 Annie Miller Dr. Robert Schatz Dr. Megan Hughes Zarzo Primary Advisor Second Reader Honors Program Director

PAGE 2

Running head: EFFECTS OF LANGUAGE ON ANTI TRANSGENDER ATTITUDES 1 Effects of Language on Anti Transgender Attitudes Andrea J. Wucherpfennig Metropolitan State University of Denver

PAGE 3

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 2 Abstract Language is known to influence perceptions of individuals and social groups, yet no research has investigated how language affects negative attitudes of transgender people. As such, w e conducted two studies that examined how the use of two linguistic forms, nouns and adjectives, affects anti transgender attitudes. In addition, we examined how beliefs in essentialism are related to the use of these linguistic forms and to participants' prejudicial attitudes. We used two different experimental procedures to manipulate linguistic form. In Study 1, we manipulated whether the items on an anti transgender attitude measure referred to transgender individuals using nouns or adjectives. I n Study 2 we manipulated whether nouns or a djectives were used to refer to a transgender person in a scenario Results showed that higher anti transgender prejudice was found for participants given nouns compared to adjectives and for high essentialist participants compared to low essentialist participants. Additionally, we found some support for the hypothesis that essentialism moderates the effect of linguistic form on anti transgender atti tudes H igh essentialist participants given nouns report ed higher anti transgender attitudes than high essentialist participants given adjectives, but there wa s no difference in anti transgender attitudes between low essentialist participants given nouns c ompared to adjectives. These findings are particularly relevant to the transgender community's current effort to have people refer to transgender individuals with adjectives rather than nouns because they illustrat e how prejudicial attitudes can be affecte d by these two linguistic forms.

PAGE 4

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 3 Effects of Language on Anti Transgender Prejudice The acronym LGBT is well known; however, it is misleading because it groups together gay, lesbian, bisexual, and transgender communities. Gay, lesbian, and bisexual peo ple are sexual minorities because their sexual orientation is other than heterosexual. Despite what the acronym implies, transgender people are not sexual minorities. This is because a person's status as transgender has nothing to do with his or her sexual attraction; instead, it refers to the person's gender identity. Transgender people are gender minorities rather than sexual minorities. Gender minorities are those whose gender identity (identity as a man, woman, or other) and assigned sex (designation at birth as "male" or "female") do not match. Although gender and sexual minorities both experience discrimination and prejudice because of their minority status, they are distinct communities. The acronym LGBT fails to show this difference, which often lead s to confusion and misunderstanding. Public awareness of LGBT issues often focus on sexual minorities' issues. The current fight for marriage equality is a prime example of this. Almost all people know about the fight for marriage equality and the majorit y of Americans have come to support it (Gallop, 2014). Meanwhile important issues for gender minorities, such as the high costs of hormone therapy and surgeries, are very rarely acknowledged. The exclusion of the transgender community is seen in research as well. For example, of the 49 items on Holland, Matthews, and Schott's (2013) measure of attitudes towards the LGBT community, only three items mention transgender people despite the researchers' claims of examining attitudes towards all LGBT people. Add itionally, the researchers did not conduct separate analyses on these items to test for differences in attitudes between gender and sexual minorities, but assumed that attitudes towards transgender people would be the same as attitudes towards lesbian, gay and bisexual people.

PAGE 5

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 4 Three scales have been developed that assess reactions toward transgender people in particular: The Genderism and Transphobia Scale (GTS; Hill and Willoughby 2005), the Transphobia Scale (Nagoshi et al., 2008), and the Attitudes Towa rd Transgendered Individuals Scale (ATTI; Walch, Ngamake, Francisco, Stitt & Schingler, 2012). However, the GTS focuses primarily on gender variant behavior (e.g., I can't understand why a woman would act masculine ) and the Transphobia Scale focuses primar ily on the acceptance of the gender binary (e.g., A person's genitalia define what gender they are; A penis defines a person as being a man, a vagina defines a person as being a woman ). Only the ATTI measures attitudes toward transgender people specificall y (e.g., I would feel uncomfortable if a close family member became romantically involved with a transgendered individual ). The present research examines the impact of language use on anti transgender prejudice. Language is known to influence perceptions of individuals and social groups, y et no research has investigated how language affects prejudice against transgender people. We conducted two studies that examine how the use of two linguistic forms, nouns and adjectives, affects anti transgender attitude s. This research is especially relevant to the transgender community because there is a current effort within the community to encourage use the term "transgender" as an adjective (e.g., a transgender person) rather than a noun (e.g., a transgender). In ad dition, we also examine how beliefs in essentialism are related to the use of these linguistic forms and to participants' prejudicial attitudes. Literature Review: Language, Essentialism, and Prejudice Previous research suggests that linguistic forms used to describe people can influence perceptions of their characteristics and behavior. An important linguistic theory, the Linguistic Category Model (LCM; Semin & Fiedler, 1988) outlines how four different linguistic forms

PAGE 6

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 5 (description action verbs, interpre tative actions verbs, state verbs, and adjectives) range on a continuum from most concrete to most abstract. According to the original LCM, adjectives are the most abstract of these forms because they convey qualities of a person rather than specific activ ities or psychological states. However, r esearch built on the LCM has shown that nouns are even more abstract than adjectives (Carnaghi et al., 2008). Nouns identify the general category to which a person belongs. W hereas adjectives denote what someone "i s like," nouns denote the kind of person someone "is." As nouns are the most abstract linguistic form, they are especially likely to elicit essentialist perceptions. Essentialism is "a belief that a certain social category has an underlying, identity det ermining essence that is permanent and common to all members of that category" (Graf, Bilewicz, Finell, & Geschke, 2012, p. 63). Research by Haslam, Rothschild and Ernst (200 2 ) has identified three main aspects of essentialist thinking. One is the belief t hat there are different types, or categories, of people that exist. As distinct boundaries exist between categories, m embership in these categories is considered immutable. Another aspect is the belief that each category has an inherent, biological basis. Last information can be gleaned simply from knowing the category a person belongs to, because all group members are thought to share uniform character traits. Research has shown a link between nouns and essentialism. Carnaghi et al. (2008) showed that p articipants with high essentialist thinking were more likely to choose nouns as descriptors than adjectives. More importantly, the researche rs also found that the reverse wa s true: People given nouns reported more essentialist thinking than people who were given adjectives. Participants were given a short description about a target that used either a noun or an adjective (e.g., Mark is a traditionalist or Mark is a traditional person) and a behavior (e.g.

PAGE 7

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 6 Mark likes to send postcards home for Christmas ). Th en, participants were asked to rate the behavior in three areas: strength (e.g., How strong is Mark's preference for this activity? ), stability (e.g., How likely is it that Mark's preference for this activity will remain the same in the next two years? ), a nd resilience (e.g., How likely is it that Mark's preference for this activity would remain the same if he was surrounded by friends who did not enjoy the activity in question? ). Participants given nouns reported higher strength, stability, and resilience for the behavior than participants given adjectives ; that is, they rated the behavior as more essentialist. Because nouns cause essentialist thinking, and essentialist thinking is the belief that people in certain categories share uniform traits, nouns sh ould also cause stereotyping. Research by Carnaghi et al. (2008) has shown that nouns cause stereotyping in two ways: Nouns facilitate stereo typic congruent descriptors and inhibit stereotypic incongruent descriptors. In one of their studies, participants were given two descriptors about a person, one noun and one adjective (e.g., A homosexual, Catholic or A Catholic, homosexual ). Then, participants were asked to rate the frequency of behaviors that are stereotypic to each descriptor (e.g., "How often does he attend church in a year" and "How often does he have one night stands in a year?"). Participants estimated higher frequencies of stereotypical behavior for the noun descriptor than for the adjective descriptor For example, pa rticipants given the wor d "Catholic" presented as a noun ("A Catholic, homosexual") reported a higher frequency of church attendance than participants given "Catholic" as an adjective ("A homosexual, Catholic") because attending church is congruent with the stereotype of Catholics. Additionally, results showed that participants expected the target to display stereotypic incongruent behaviors less frequently for th e noun than for the adjective. For example, p articipants given "Catholic" as a noun ("A Catholic, homosexual") estimated lower frequencies for one night stands than participants given

PAGE 8

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 7 "Catholic" as an adjective ("A homosexual, Catholic") because one night stands are incongruent with the stereotype of Catholics. Because of this link between nouns and stereotyping, it follows that nouns can cause negative attitudes provided the stereotype is also negative. While there has been no previous research examining this hypothesis, there has been support for a similar theory that nouns elicit more in group bias than adjectives. Graf et al. (2012) found that participants given nouns to describe a target person's nationality (e.g., A German, a painter ) exhibited greater preferences for their own national group than participants who were given adjectives (e.g., A German painter ). However i t is important to note that relatively favorable evaluation of an in group need not involve negative evaluation of an out group. W hile in group bias can provide a foundation for prejudice, it doe s not necessarily lead to prejudice (Brewer, 1999). Theref ore, research comparing nouns and adjectives on prejudicial attitudes is necessary. Present Research We tested the impact of linguistic form and essentialist beliefs on anti transgender prejudice in two studies. In Study 1, we manipulated linguistic form in attitude items that referred to transgender people using either nouns or adjectives (e.g., a transgender, a transgender person ) In Study 2, we compared the effects of using nouns and adjectives to describe a transgender person in a scenario. In both s tudies we predicted that participants given nouns would report higher anti transgender prejudice than participants given adjectives. We also expected that participants' beliefs in essentialism would be associated with more negative attitudes towards transg ender people, and that essentialist beliefs would enhance this relationship.

PAGE 9

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 8 Study 1 In Study 1, participants completed a survey that contained items assessing attitudes toward transgender individuals. Linguistic form was experimentally manipulated by w hether the items referred to transgender individuals using nouns (e.g., My friends and I have joked about transgenders ) or adjectives (e.g., My friends and I have joked about transgender people ). Average level of agreement with these statements was used to measu re anti transgender prejudice. W e predicted higher scores for participants in the noun condition than those in the adjective condition ( H1 ) In order to assess individual differences in essentialism, w e included the Essentialist Beliefs scale (Bast ian & Haslam, 2006). The measure includes three subscales: Informativeness (e.g., When getting to know a person it is possible to get a picture of the kind of person they are very quickly ), Biological Basis (e.g., Whether someone is one kind of person or a nother is determined by their biological make up ), and Discreteness (e.g., The kind of person someone is, is clearly defined; they either are a certain kind of person or they are not ). Based on previous research linking essentialism with prejudicial attitu des towards other stigmatized group members (Howell, Weikum & Dyck, 2010; Haslam, Rothschild & Ernst, 2002), w e predicted that high scores on the overall essentialism measure would predict more anti transgender prejudice ( H2 ) However, we also suspected th at the strength of this relationship might vary between subscales. Specifically, w e expected Informativeness scores would predict the highest anti transgender prejudice of the three essentialism subscales ( H3 ) This is because a person high in informative essentialism is likely to draw on the negative stereotype of transgender people to make assumptions about other qualities and characteristics the transgender person possesses. By drawing upon these negative qualities, the participant is then more likely to have a negative

PAGE 10

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 9 evaluation of a transgender person On the other hand, belief in a biological basis for being transgender might not cause negative attitudes towards that person because it may "excuse" that person for being transgender. This line of thinki ng has been shown in research on anti gay attitudes, where belief in biological basis sometimes predicts lower prejudice (Haslam & Levy, 2006). As such, we did not expect a strong relationship between Biological Basis and anti transgender prejudice Additi onally, because a transgender person's outward appearance changes during the process of transitioning, being transgender might not be considered as a clearly defined, distinct category. Therefore we did not expect a strong relationship between Discreteness and anti transgender prejudice either. Additionally, we examined whether essentialist beliefs would moderate the relationship between linguistic form and anti transgender attitudes. Previous research findings linking nouns to essentialism and stereotypin g suggests that the predicted effect of linguistic form (i.e., stronger anti transgender prejudice for nouns than for adjectives) might be especially pronounced for participants with high essentialism scores. If so, there should be a significant interactio n between linguistic form and strength of essentialist beliefs ( H4 ). As Informativeness is expected to have an especially strong relationship, we also expected a similar interaction between linguistic form and strength of informative essentialist beliefs ( H5 ). Finally we compared the effect of group label "transgender" or "transsexual" on attitudes towards transgender people. The term "transsexual" has a more clinical connotation because it was included in the Diagnostic and Statistical Manual (DSM) in DS M III (3 rd ed; DSM III ; American Psychiatric Association, 1980) and DSM IIIR (3 rd ed., rev; DSM IIIR ; American Psychiatric Association, 1987) as a subtype of "Gender Identity Disorder" ( Gender Identity Disorder/Children Transsexualism ). As previous researc h has linked essentialism with negative

PAGE 11

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 10 attitudes toward peopl e with mental illnesses (Howell et al., 2010), this label might be expected to cause higher anti transgender attitudes than the label "transgender." However, "transsexual" was removed from subse quent versions of the DSM, lessening the clinical connotation of the term Additionally, "transgender" currently is used more often, potentially causing the general population to associate more stigma with "transgender" than "transsexual." Given the confli cting arguments, we did not make a formal prediction concerning which term would be associated with more negati ve attitudes. Method Participants Two hundred thirty nine participants (147 female, 92 male) were recruited from undergraduate introductory psyc hology courses at an urban university. The mean age was 23.3 years; 90% of the participants identified as straight, 1.7% identified as lesbian, 2.9% as gay, 3.3% as bisexual, 1.7% as queer, and 0% as questioning/unsure. Eight participants identified as a g ender minority (e.g., transgender, transsexual). Fifty seven percent of the participants identified as White/European American, 20.4% as Latino(a)/Hispanic American, 6.3% as Black/African American, 5.4% as Asian American, and 10.5% as Other. Procedure Sm all groups of participants completed the survey individually After giving consent, each participant was given a survey titled "Social and Interpersonal Attitudes Participants were randomly assigned to condition based on which version of the survey they received; a s the survey cover sheets were identical, the researcher was blind to condition. After completing the survey, participants were orally debriefed and given course credit as compensation for their time. Materials and Design As a distracter, parti cipants first responded to the 18 item Beliefs About Groups scale (BAG; Karau & Elsaid, 2009; e.g., People tend to work especially hard on a

PAGE 12

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 11 group task ) using a 5 point Likert scale (1 Strongl y Disagree, 5 Strongly Agree). The second questionnaire measur ed participants' attitudes towards the transgender community and included the exper imental manipulation s ( explained in the following paragraph ). The measur e consisted of 21 items (! = .9 5) adapted from previously established scales including the Transphobi a Scale (Nagoshi et al., 2008), the Genderism and Transphobia Scale (GTS; Hill & Willoughby, 2005), and the Attitudes Towards Transgendered Individuals Scale ( ATTI; Walch et al. 2012). Items were chosen that focused on participants' attitudes towards tra nsgender individuals, rather than anti transgender behavior. Participants responded using a 1 Strongly Disagree to 7 Strongly Agree scale. The average of the scores on these items constituted the dependent measure of anti transgender attitude, with a highe r score equaling more prejudice. Participants were randomly assigned to condition in a 2 (Linguistic Form: noun or adjective) x 2 (Group Label: transgender or transsexual) between participants factorial design based on which version of the anti transgender attitude measure they received; as the survey cover sheets were identical, the researcher was blind to condition. Participants completed one of four versions of t he questionnaire : Noun/Transgender Noun/Transsexual, Adjective/Transgender, or Adjective/Tra nssexual (see Figure 1 for a sample item; see Appendix A for the full measure) Noun Adjective Transgender My friends and I have joked about transgenders My friends and I have joked about transgender people Transsexual My friends and I have joked about transsexuals My friends and I have joked about transsexual people Figure 1. A sample item from each of the four versions of the anti transgender attitudes measure.

PAGE 13

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 12 Next was the Essentialist Beliefs scale ( Bastian & Haslam, 2006; = .78 ) which included three subscales: Bio logical Basis (8 items; = .71 ; e.g., Whether someone is one kind of person or another is determined by their biological make up ), Discreteness (8 items; = .65 ; e.g., The kind of person someone is, is clearly defined; they either are a certain kind of person or they are not ), and In formativeness (7 items; = .63 ; e.g., When getting to know a person it is possible to get a picture of the kind of person they are very quickly ). Participant s responded using a 6 point Likert scale anchored from 1 Strongly Disagree to 6 Strongly Agree. Lastly, participants provided basic demographic information including sexual orientation, ethnic racial background, and identification as a gender minority (e. g., transgender, transsexual). Results We first performed a median split on participants' essentialism scores to create high essentialism and low essentialism groups A 4 way Linguistic Form (noun or adjective) x Group Label (tra nsgender or transsexual) x Essentialism (high or low) x Sex (male or female) between participants Analysis of Variance (ANOVA) was conducted on the anti transgender attitude scores. Results indicated significant 1 main effects for linguistic form F (1, 239) = 8.112, p = 0.005 p 2 = 0.035, essentialism group F (1, 239) = 12.876, p < 0.001 p 2 = 0.055, and sex F (1, 239) = 25.447, p < 0.001 p 2 = 0.102. As predicted, s ignificantly higher anti transgender prejudice was found for nouns than for adjectives ( H1 ; See Figure 2; Ms = 3.249 and 2.809 ) and for high essentialist participants compared to low essentialist participants ( H2 ; See Figure 3; Ms = 3.306 and 2.752). In addition, males reported significantly higher anti transgender prejudice than females ( Ms = 3.419 and 2.639). No main effect was found for group label ( p = 0.46). 1 In our study, we considered significant effects to have p .05 and marginal effects to have p .10 two tailed

PAGE 14

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 13 However, ther e was a marginally significant Linguistic Form x Group L a bel x E ssentialism interaction F (1,239) = 3.350, p = .069, p 2 = .015. Effect of Linguistic Form on Anti Transgender Attitudes Figure 2. Results indicated a main effect for Linguistic Form: Participants given nouns reported more negative anti transgender attitudes than participants given adjectives. Effect of E ssentialism on Anti Transgender Attitudes Figure 3. Results indicated a main effect for Essentialism: High essentialist participants reported more negative anti transgender attitudes than low essentialist participants.

PAGE 15

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 14 As the 4 way ANOVA produced no sign ificant interactions with sex, we reanalyzed the anti transgender attitudes scores in a 3 way Linguistic Form (noun or adjective) x Group Label (transgender or transsexual) x Essentialism (high or low) ANOVA. The linguistic form and essentialism main effec ts remained significant in this analysis ( p s < .001). In addition, the 3 way interaction reach ed statistical significance F (1, 239 ) = 3.876, p = .05, p 2 = .017. To explor e the 3 way interaction, 2 way Linguistic Form x E ssentialism ANOVAs were performed separately for participants in each group label condition. A marginally significant interaction was found for participants exposed to the label "transsexual" ( H4 ), F (1, 239) = 3.217 p = .075, p 2 = 027 This in teraction is displayed in Figure 4 Simple e ffect tests showed that anti transgender prejudice scores were s ignificantly higher in the noun condition t han in the adjective condition for participants w ho scored high in essentialism F (1, 239) = 6.899, p = .01, p 2 = .055 ( M = 3.617 and 2.813). Howev er, t here was no difference in anti transgender prejudice between the noun and adjective conditions for par ticipants who scored low in essentialism F (1, 239) = .007 p = .932

PAGE 16

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 15 Anti Transgender A ttitudes of Participants in the "T ranssexual" Group L abel C o ndition Figure 4 High essentialist participants given nouns reported significantly higher anti transgender attitudes than high essentialist participants given adjectives; however, there was no difference in attitudes for low essentialist participants g iven nouns compared to adjectives. We then conducted separate Linguistic Form (noun or adjective) x Group Label (transgender or transsexual) x Essentialism (high or low) ANOVAs that compared anti transgender attitudes for high and low scorers on each esse ntialism subscale. We performed median splits on each subscale to create high essentialist and low essentialist groups. The main effects for essentialism and linguistic form were significant or marginally significant in each analysis ( ps .052): Nouns and higher essentialism scores predicted more negative anti transgender attitudes. Contrary to our prediction the relationship between essentialism and anti transgender prejudice was weakest for the Informativeness subscale ( H3 ) F (1, 239) = 4.918, p = .064, p 2 = .015, rather than Biological Basis F (1, 239) = 3.824, p = .052, p 2 = .018, or Discreteness F (2, 239) = 14.498, p < .001, p 2 = .065.

PAGE 17

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 16 However, in support of prediction H5 t he Linguistic Form x Group Label x E ssentialism inter action was significant only for informativeness F (1, 239) = 5.187, p = .024, p 2 = .022 not for the biological basis F (1, 239) = 2.352, p = .127, p 2 = .011 or discreteness F (1, 239) = .808, p = .37, p 2 = .004 A nalyses of anti transgender attitude s in the transsexual label condition that compared high and low scorers on informativeness yielded the same pattern of results found for scores on the overall essen tialism measure: a significant Linguistic Form x Informativeness interaction F (1, 239) = 7. 573, p = .007, p 2 = .060 and significantly higher prejudice for nouns relative to adjectives for participants high in informativeness ( Ms = 3.509, 2.732 ) but not for participants low in informativeness ( Ms = 2.976, 2.684). This inte raction is displayed in Figure 5 An ti Transgender A ttitudes Figure 5 Informative essentialism follows the same pattern as overall essentialism: H igh essentialist participants given nouns reported significantly higher anti transgender attitudes than high essentialist participants given a djectives, but there was no difference in attitudes for low essentialist participants given nouns compared to adjectives.

PAGE 18

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 17 Discussion The results of Study 1 supported the hypothesis that the use of nouns to refer to members of the transgender community cau se s more negative attitudes toward these individuals than does the use of adjectives ( H1 ) We also found support for the hypot hesis that essentialism moderat es this effect ( H4 ) For participants exposed to the group label "transsexual," high essentialist p articipants reported significantly more negative attitudes in the noun condition than in the adjective condition ; however this difference was not significant for low essentialist participants It is not clear why this interaction was found only in the "tra nssexual" condition One possible explanation may involve the links between linguistic form, essentialism, and attitudes toward mental illness. As mentioned before previous research has shown that the use of nouns predicts more essentialist thinking than the use of adjectives (Carnaghi et al., 2008), and that essentialist thinking predicts prejudicial attitudes towards people with mental illnesses (Howell et al. 2010). Given the inclusion of "transsexual" in the DSM, it may be that nouns elicited more ant i transgender attitudes among high essentialist scorers only when this clinical term was used as a group label. W hen we examined the subscales of essentialism separately, a significant interaction with linguistic form was found only in the ANOVA that compa red high and low score r s on informativeness ( H5 ) Informative essentialism claims that a person's characteristics and qualities can be inferred from knowledge of that person's group membership. This ideology is the foundation for stereotyping Since the st ereotype of transgender people is negative, it should generate prejudice. Because nouns are essentialist, they would have increased stereotyping among participants high in informativeness and caused significantly higher anti transgender attitudes compared to adjectives.

PAGE 19

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 18 We conducted an additional study to further explore the effects of linguistic form and essentialism on anti transgender prejudice. For Study 2, we examined these relationships with regard to a transgender person depicted in a scenario. We w ere interested to see if the effect of linguistic form obtained in Study 1 would generalize to this situation. Additionally, we designed Study 2 to test whether beliefs in essentialism mediate the effect of linguistic form on anti transgender attitudes. S tudy 2 In Study 2, participants read a scenario that depicted the events surrounding a transgender person's decision to transition to another gender. In order to test the effect of linguistic form on anti transgender attitudes, the target was described wi th either nouns or adjectives. After reading the scenario, participants responded to modified version s of the essentialism and anti transgender attitude measures used in Study 1 that referred specifically to the target person. We predicted that nouns would induce more negative anti transgender attitudes than adjectives ( H1 ) and that essentialist beliefs about the target ( H2 ) particularly informativeness ( H3 ) would mediate this difference. In addition to linguistic mode, we also examined whether transiti on type (male to female or female to mal e) affected participants' anti transgender attitudes towards the target No previous research has examined attitudes towards transition types; h owever, there has been research on the experiences of transgender people A survey from the National Center for Transgender Equality and the National Gay and Lesbia n Task Force (Grant et al., 2011 ) found that transgender men (those who transition from female to male) reported higher frequencies of harassment and bullying in th eir K 12 schools years. However, transgender women (those who transition from male to female) reported more discrimination after age 18, including higher rates

PAGE 20

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 19 of discrimination from their medical providers, more relationships ending due to transitioning, more access denied to shelters, and alm ost twice as many experiences being physically assaulted compared to transgender men. The fact that transgender women face so much more discrimination in their lives than transgender men likely reflects a societal pre judicial attitude towards them. Therefore, we predicted higher anti transgender attitudes for participants who evaluated a person transitioning from male to female (transgender woman) than for participants who evaluated a person transitioning from female t o male (transgender man ; H4 ). Method Participants Two hundred and sixty nine participants (179 female, 90 male) were recruited from undergraduate introductory psychology courses at an urban univers ity. The mean age was 22.23 years; 89.5 % of the participan ts identified as straight, 2 % identified as lesbian, 2% as gay, 5.4% as bisexual, 0 % as queer, and 1 % as questioning/unsure. One participant identified as a gender minority (e.g., transgender, transsexual). Fifty four percent of participants identifi ed as White/European American, 27.8 % as Latino(a)/Hispanic American, 5.6% as Black/African American, 4.4 % as Asian American, and 8.1 % as Other. We excluded 6 participants from our analyses because they clearly did not take the survey seriously. Procedures. Small groups of participants completed the surveys individually 2 After giving consent, each participant was given a survey titled "Group and Interpersonal Attitudes Participants were randomly assigned to condition based on which version of the survey they re ceived; as the survey cover sheets were identical, the researcher was blind to condition. After completing the survey, participants were orally debriefed and given course credit as compensation for their time. 2 Participants were recruited from the same source as Study 1. However, res trictions were put in place so that students were unable to participate in Study 2 if they had participated in Study 1.

PAGE 21

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 20 Materials and Design As with Study 1, p artici pants first responded to the 18 item distractor questionnaire, Beliefs About Groups (BAG; Karau & Elsaid, 2009). Following the distractor, participants read a scenario of approximately 250 words that described a transgender person "Jessie." Jessie is descr ibed as identifying as another gender but hiding his or her gender from childhood through high school graduation. In college, Jessie experiments with gender presentation when alone, but is afraid to tell anyone else. Finally, the scenario ends with Jessie coming out to friends and changing the sex markers on official IDs. Linguistic form and transition type were manipulated in a 2 (noun or adjective) x 2 (female to male or male to female) between participants factorial design. Participants received one of four vers ions of the scenario ( see Figure 6 for a sample of each scenario; see Appendix B for the entire scenario ): Noun/Female to male, Noun/Male to Female, Adjective/Female to Male, or Adjective/Male to Female. Noun Adjective Female to Male As Jessie grew up, she often felt that she was a boy. However, she remembered how negatively her parents reacted to transgenders. As Jessie grew up, she often felt that she was a boy. However, she remembered how negatively her parents reacted to transgender people. Male to Female As Jessie grew up, he often felt that he was a girl. However, he remembered how negatively his parents reacted to transgenders. As Jessie grew up, he often felt that he was a girl. However, he remembered how negatively his parents reacted to transgender people. Figure 6 A sample from each of the four versions of the scenarios. After reading the scenario, participants responded to a modified version of the Essentialist Beliefs scale (Bastian & Haslam, 2006 ; = .78 ) that contained three subscales used in Study 1 : Biological basis (! = .87), Discreteness (! = .74 ), and Informativeness (! = .65 ). Based on reliability analyses on the Discreteness subscale, 3 of 8 items were removed. All

PAGE 22

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 21 essentialist i tems were modified to apply to the targe t Jessie or to people like Jessie (e.g., Jessie's basic character is never easily defined; People are either like Jessie or they are not ). Participants responded using a 6 point Liker t scale (1 Strongly Disagree, 6 Strongly Agree). Then participants res ponded to a modified version of the anti transgender attitudes measure from Study 1 ( = .92 ). As with the Essentialist Beliefs scale, anti transgender attitude items were modified to apply to the target Jessie or to people like Jessie (e.g., Jessie should not be allowed to work with children ; People like Jessie endanger the institution of the family ) The scale was anchored from 1 Strongly Disagree to 7 Strongly Agree. The dependent variable was the average scores on this measure, with higher scores equaling more prejudice. Last participants provided the same basic demographic inform ati on asked in the first study. Results We first ran a Linguistic Form (noun or adjective) x Transition Type (male to female or female to male) x Sex (male or female) ANOVA on the anti transg ender attitude scores. Contrary to the prediction that anti transge nder attitudes would be more negative for the male to female transition target than for the female to male transition target ( H4 ) there were no significant effects for transition type. Because t ransition t ype did not have any significant effects, we reana lyzed the data in a Linguistic Form (noun or adjective) x Sex (male or Female) ANOVA. There was a significant effect for sex, F (1, 269) = 30.922, p < .001, p 2 = .104 that showed males reported higher anti transgender prejudice than females ( Ms =2.980, 2.336). The effect for linguistic form was marginally significant (see Figure 7) F (1, 269) = 3.067, p = .081, p 2 = .011 As predicted ( H1 ) participants give n nouns reported higher anti transgender attitudes than participants given adjectives ( Ms = 2.753, 2. 563 ). In addition, there was a significant Linguistic Form x Sex interaction, F (1, 269) = 3.983, p = .047, p 2 = .015. Further

PAGE 23

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 22 analysis of this interaction showed that males given nouns reported higher anti transgender prejudice than males given adjectives ( Ms = 3.293 and 2.807); however, no significant difference was shown for females given nouns compared to adjectives ( Ms = 2.313 and 2.345). Effect of Ling uistic Form on Anti Transgender Attitudes Figure 7 Results indicated a marginally significant main effect for linguistic form: Participants given nouns reported more negative anti transgender attitudes than participants given adjectives. To examine t he possibility that essentialist beliefs about the target mediated the relationship between linguistic form and anti transgender attitudes, we first tested whether linguistic form predicted scores on the overall essentialism measure. A one way ANOVA compar ing the scores of participants in the noun and adjective conditions was not significant F (1, 268) = .938, p = .334. A Linguistic Form (noun or adjective) x Sex (male or female) ANOVA also found no effect for linguistic form F (1, 265) = 1.896, p = .170 S ince linguistic form was not shown to predict essentialism scores, essentialism could not mediate the effect of linguistic form on anti transgender prejudice ( H2, H3 )

PAGE 24

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 23 With no evidence for mediation, we decided to test if essentialism moderated the effect of linguistic form on anti transgender attitudes as it had in Study 1. In order to test moderation, we ran a Linguistic Form (noun or adjective) x Sex (male or female) x Essentialism (high or low) ANOVA comparing anti transgender attitudes for high and lo w scorers on the overall essentialism measure. Results showed significant effects for sex F (1, 269) = 26.318, p < .001 p 2 = .092 and essentialism F (1, 269) = 13.464, p < .001 p 2 = .049. Higher anti transgender attitudes were found for males than female s ( Ms = 3.082, 2.33) and for high essen tialist participants compared to low essentialist participants ( see Figure 8; Ms = 2.895, 2.426). Although the predicted Linguistic Form x Essentialism interaction was not significant F (1, 269), = 1.991, p = .159, p 2 = .008, the mean scores were in the pr edicted directions (See Figure 9 ). Simple effect tests showed high essentialist participants given nouns reported higher anti transgender prejudice than high essentialist participants given adjectives F (1, 261) = 3. 846, p = .051 ( Ms = 3.059, 2.766 ); however, there was no difference in anti transgender attitudes for low essentialist participants given nouns versus adjectives F (1, 261) = .008, p = .930 ( Ms = 2.324, 2.373)

PAGE 25

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 24 Effect of Essentialism on Anti Transg ender Attitudes Figure 8 Results indicated a main effect for Essentialism: High essentialist participants reported more negative anti transgender attitudes than low essentialist participants. Anti Transgender A ttitudes Figure 9 Although the interact ion is not significant, the mean score trends mimic Study 1: H igh essentialist participants given nouns reported significantly higher anti transgender attitudes than high essentialist participants given adjectives; however, no significant difference was f ound for low essentialist participants given nouns compared to adjectives.

PAGE 26

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 25 We then conducted separate Linguistic Form (noun or adjective) x Sex (male or female) x Essentialism (high or low) ANOVAs that compared anti transgender attitudes for high and low scorers on each essentialism subscale. Main effects for sex were significant for all subscales ( p < .001) and main effects for essentialism were significant for both informativeness and discreteness ( p < .001). In addition, there was a Linguistic Form x Se x interaction that was significant for informativeness, F (1, 269) = 3.903, p = .049, p 2 = .015, and marginally significant for biological basis, F (1, 269) = 3.192, p = .075, p 2 = .004. This interaction showed that males given nouns reported higher anti t ransgender prejudice than males given adjectives (Informativeness: Ms = 3.202, 2.757; Biological Basis: Ms = 3.189, 2.747 ); however, there was no difference for females given nouns and adjectives (Informativeness: Ms = 2.304, 2.369; Biological Basis: Ms = 2.309, 2.372). The Linguistic Form x Essentialism interaction was not significant for any subscale: Informativeness, F (1, 269) = 2.201, p = .139, p 2 = 008, Biological Basis, F (1, 269) = .210, p = .647, p 2 = .000 or Discreteness, F (1, 269) = 1.023, p = .313, p 2 = .004 However, simple effect tests showed high scorers on informativeness given nouns reported significantly higher anti transgender a ttitudes than high essentialist participants given adjectives, F (1, 269) = 4.37, p = .038, p 2 = .016 ( Ms = 3.16, 2.795) but there was no difference in anti transgender attitudes for low essentialist participants given nouns and adjectives F (1, 269) = 0.0 0, p = .984 ( Ms = 2.389, 2.392). This interaction can be seen in Figure 4. Anti transgender attitudes also tended to be higher in the noun condition than in the adjective condition for high scorers on the other two essentialism subscales, although the diff erence was not significant for biological basis ( p = .158) and was marginally significant for discreteness ( p = .061) There was no significant difference in

PAGE 27

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 26 anti transgender attitudes between participants given nouns and adjectives for low scorers on eith er subscale ( ps # .48). Ant i Transgender A ttitudes Figure 10 Informative essentialism again follows the same pattern as overall essentialism: H igh essentialist participants given nouns reported significantly higher anti transgender attitudes than high essentialist participants given adjectives; however, no significant difference was found for low essentialist participants given nouns compared to adjectives. Discussion We found some support for the hypothesis that nouns generate more anti transgender p rejudice than adjectives ( H1 ) although the difference was marginally significant across all participants and statistically significant only for males The results did not support our hypothesis that essentialism would mediate the effect of linguistic form on anti transgender attitudes ( H2 ) I nteractions that tested whether essentialist beliefs moderated the effect of linguistic form on anti transgender attitudes were not significant, however the trends were in the predicted direction and the difference bet ween scores in the noun and adjective condition was significant for

PAGE 28

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 27 informativeness. As mentioned in Study 1, participants high in informative essentialism infer negative characteristics from the negative stereotype of transgender people which causes more negative attitudes towards the target. Since nouns are essentialist, they increased stereotyping among these individual to cause significantly higher anti transgender attitudes compared to adjectives. Additionally, results from Study 2 yielded significan t effects for sex. The main effect for sex was found in multiple ANOVAs, showing that males reported higher anti transgender attitudes than females. Not only was this main effect present, but it had the largest effect 3 This difference is consistent with p revious prejudice literature that shows males are more prejudiced than females ( Hill & Willoughby, 2005; Willoughby, Hill & Gonzalez, 2010; Nagoshi, et al. 2008; Lippa & Arad, 1999 ; Jewell & Morrison, 2012 ). Additionally, there was a significant interactio n that showed males given nouns reported significantly higher anti transgender attitudes than males given adjectives, but no differences in anti transgender attitudes for females given nouns compared to adjectives. We did not predict this finding and we ar e not aware of any previous research that has examined the interactive effect of linguistic form and sex on expressions of prejudice. W e suspect that the difference might be due to the fact that males in our study scored higher on essen tialism than females As nouns predicted more prejudice than adjectives for participants who scored high on essentialism, males exposed to nouns were more prejudiced than males exposed to adjectives. F uture research should further examine this relationship General Discussion Our hypothesis that participants given nouns would report higher anti transgender 3 Thi s effect size was considered medium. All other effects were considered small according to Gray & Kinnear (2012).

PAGE 29

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 28 attitudes than participants given adjectives was supported by two separate studies These two studies also supported our hypothesis that essentialism predicts anti transge nder attitudes. When we examined how the linguistic form and essentialism interact, we found support for our prediction that essentialism would moderate the effect of linguistic form on anti transgender prejudice (although the results were stronger in Stud y 1 than Study 2) In Study 1, for participants exposed to the group label "transsexual," high essentialist participants given nouns reported higher anti transgender prejudice than high essentialist participants given adjectives, but there was no significa nt difference for low essentialist participants. This interaction was significant for the overall essentialism measure and for informative essentialism. The same pattern of findings was obtained for both measures in Study 2 (higher prejudice for nouns than for adjectives among high essentialist participants but not low essentialist participants) although the interaction effects were not statistically significant. Participants in our research were relatively young, primarily white undergraduates at a publi c university. Because of these characteristics and the urban setting of the campus, our participants probably were more liberal in the i r political beliefs and more tolerant of transgender people than the average United States citizen. Future research shou ld be done with a wider demographic to assess the generalizability of our results. Additionally, as with any experiment, the artificial nature of the study can bring external validity into question. We attempted to combat this in Study 2 by creating a scen ario about a transgender person in order to create a more realistic situation. Future research should continue along this path to see if these effects continue to hold in more generalizable contexts. Research Applications The present research makes an im portant contribution to the literature on the effects of

PAGE 30

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 29 language on prejudice. While previous research has established a link between linguistic form and essentialism, and has shown that linguistic form can affect stereotyping and in group bias, our rese arch is the first to demonstrate that linguistic form directly impacts prejudicial attitudes. By evaluating the effects of linguistic form on attitudes towards transgender people, our research also adds to the growing literature on anti transgender prejudi ce. Our research also has important practical implications outside of the academic community. The transgender community asks people to use the term "transgender" as an adjective instead of a noun and o ur research supports the merit of this aim. For exampl e, our findings suggest that a news article that refers to a transgender person as "a transgender" (i.e., using a noun) may elicit more anti transgender prejudice among readers than a comparable article that refers to "a transgender person" (i.e., using an adjective) Although the effect s we found for linguistic form are small in size, repeated exposure to noun references in the media might reliably increase prejudice that hinders this already marginalized community in its fight for awareness and acceptance As language both inside and outside of the transgender community continues to evolve, research should continue to examine its impact on prejudicial attitudes. In this way, we may lighten the load of marginalized groups and assist in their fights for equa lity.

PAGE 31

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 30 References American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3 rd ed.) Washington, DC: Author. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3 rd ed., rev.) Washington, DC: Author. Bastian, B. & Haslam, N. (2006). Psychological essentialism and stereotype endorsement. Journal of Experimental Social Psychology (42) 2. doi: 10.1016/j.jesp.2005.03.003 Brewer, M. (1999). The psychology of prejudice: In group love and outgroup hate? Journal of Social Issues (55) 3. doi: 10.1111/0022 4537.00126 Carnaghi, A., Maass, A., Gresta, S., Bianchi, M., Cadinu, M. & Arcuri, L. (2008). Nomina sunt omina: On the inductive potential of nouns and adjectives in person p erception. Journal of Personality and Social Psychology (94) 5. doi: 10.1037/0022 3514.94.5.839 Gallup. (2014). "Marriage: Do you think marriages between same sex couples should or should not be recognized by the law as valid, with the same rights as tradi tional marriages?" May 8 11, 2014 [Survey Report]. Retrieved from http://www.gallup.com/poll/117328/ma rriage.aspx Graf, S., Bilewicz, M., Finella, E. & Geschke, D. (2013). Journal of Language and Social Psychology (32) 62. doi: 10.1177/0261927X12463209 Gr ant, J., Mottet, L., Tanis, J., Harrison, J., Herman, J. & Keisling, M. (2011). Injustice at every turn: A report of the national transgender discrimination survey. Washington: National Center for Transgender Equality and National Gay and Lesbian Task Fo rce. Gray, C.D. & Kinnear, P.R. (2012). IBM SPSS Statistics 19 Made Simple. Hove and New York: Psychology Press.

PAGE 32

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 31 Haslam, N. & Levy, S. (2006). Essentialist beliefs about homosexuality: Structure and implications for prejudice. Personality and Social Psyc hology Bulletin (32) 4. doi: 10.1177/0146167205276516 Haslam, N., Rothschild, L. & Ernst, D. (2002). British Journal of Social Psychology (41) 1. doi: 10.1348/014466602165072 Hill, D. & Willoughby, B. (2005). The development and validation of the genderism and transphobia scale. Sex Roles (53) 7. doi: 10.1007/s11199 005 7140 x Holland, Matthews, Schott (2013). "That's so gay!": Exploring college students' attitudes toward the LGBT population. Journal of Homosexuality (60) 4. doi: 10.1080/00918369.2013.76032 1 Howell, A., Weikum, B. & Dyck, H. (2011). Psychological essentialism and its association with stigmatization. Personality and Individual Differences (50) 1. doi: 10.1016/j.paid.2010.09.006 Jewell, L. & Morrison, M. (2012). Making sense of homonegativity: Heterosexual men and women's understanding of their own prejudice and discrimination toward gay men. Qualitative Research in Psychology (9) 4. doi: 10.1080/14780887.2011.586098 Karau, S. & Elsaid, A. (2009). Individual differences in beliefs about groups. Group Dynamics: Theory, Research, and Practice (13) 1. doi: 10.1037/a0013366 Lippa, R. & Arad, S. (1999). Gender, personality, and prejudice: The display of authoritarianism and social dominance in interviews with college men and women. Journal of Resear ch in Personality (33) 4. doi: 10.1006/jrpe.1999.2266

PAGE 33

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 32 Nagoshi, J., Adams, K., Terrell, H., Hill, E., Brzuzy, S., & Nagoshi, C. (2008). Gender differences in correlates of homophobia and transphobia. Sex Roles (59) 521 531. doi: 10.1007/s11199 008 94587 7 Semin, G. & Fiedler, K. (1991). European Review of Social Psychology (2) 1. doi: 10.1080/14792779143000006 Walch, S., Ngamake, S., Francisco, J., Rashunda, S. & Shingler, K. (2012). Archives of Sexual Behavior (41) 5. doi: 10.1007/s10508 012 9995 6 Willou ghby, B., Hill, D., & Gonzalez, C. (2010). Who hates gender outlaws? A multisite and multinational evaluation of the genderism and transphobia scale. International Journal of Transgenderism (12) 4. doi: 10.1080/15532739.2010.55082

PAGE 34

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 33 Appendix A Noun/Transgender I would be upset if someone I'd known for a long time revealed to me that they are a transgender. I think there is something wrong with a person who says they are a transgender. I don't like it when someone is flirting with me and I can te ll they are a transgender. I think there is something wrong with a person who says that they are a transgender. If I found out my lover was a transgender, I would freak out. I would avoid talking to someone if I knew they were transgenders. I would go to a bar that was frequented by transgenders. My friends and I have joked about transgenders. If I found out that my best friend was a transgender, I would freak out. Transgenders should not be allowed to work with children. Transgenders are a viable part of o ur society. Transgenders endanger the institution of the family. Transgenders should be accepted completely into our society. I avoid transgenders whenever possible. I would enjoy attending social functions at which transgenders are present. I would feel c omfortable if I learned my neighbor was a transgender. I would like to have friends who are transgenders. I would feel uncomfortable if a close family member became romantically involved with a transgender. Romantic partners of transgenders should seek psy chological treatment.

PAGE 35

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 34 It would be beneficial to society to recognize transgenders as normal. I would feel comfortable working closely with a transgender. Noun/Transsexual I would be upset if someone I'd known for a long time revealed to me that they are transsexual. I think there is something wrong with a person who says they are a transsexual. I don't like it when someone is flirting with me and I can tell they are a transsexual. I think there is something wrong with a person who says that they are a tra nssexual. If I found out my lover was a transsexual, I would freak out. I would avoid talking to someone if I knew they were a transsexual. I would go to a bar that was frequented by transsexuals. My friends and I have joked about transsexuals. If I found out that my best friend was a transsexual, I would freak out. Transsexuals should not be allowed to work with children. Transsexuals are a viable part of our society. Transsexuals endanger the institution of the family. Transsexuals should be accepted comp letely into our society. I avoid transsexuals whenever possible. I would enjoy attending social functions at which transsexuals are present. I would feel comfortable if I learned my neighbor was a transsexual. I would like to have friends who are transsexu als. I would feel uncomfortable if a close family member became romantically involved with a transsexual.

PAGE 36

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 35 Romantic partners of transsexuals should seek psychological treatment. It would be beneficial to society to recognize transsexuals as normal. I would feel comfortable working closely with a transsexual. Adjective/Transgender I would be upset if someone I'd known for a long time revealed to me that they are transgender. I think there is something wrong with a person who says they are transgender. I don 't like it when someone is flirting with me and I can tell they are transgender. I think there is something wrong with a person who says that they are transgender. If I found out my lover was transgender, I would freak out. I would avoid talking to someone if I knew they were transgender. I would go to a bar that was frequented by transgender people. My friends and I have joked about transgender people. If I found out that my best friend was transgender, I would freak out. Transgender people should not be a llowed to work with children. Transgender people are a viable part of our society. Transgender people endanger the institution of the family. Transgender people should be accepted completely into our society. I avoid transgender people whenever possible. I would enjoy attending social functions at which transgender people are present. I would feel comfortable if I learned my neighbor was transgender. I would like to have friends who are transgender people.

PAGE 37

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 36 I would feel uncomfortable if a close family member became romantically involved with a transgender person. Romantic partners of transgender people should seek psychological treatment. It would be beneficial to society to recognize transgender people as normal. I would feel comfortable working closely wit h a transgender person. Adjective/Transsexual I would be upset if someone I'd known for a long time revealed to me that they are transsexual. I think there is something wrong with a person who says they are transsexual. I don't like it when someone is fli rting with me and I can tell they are transsexual. I think there is something wrong with a person who says that they are transsexual. If I found out my lover was transsexual, I would freak out. I would avoid talking to someone if I knew they were transsexu al. I would go to a bar that was frequented by transsexual people. My friends and I have joked about transsexual people. If I found out that my best friend was transsexual, I would freak out. Transsexual people should not be allowed to work with children. Transsexual people are a viable part of our society. Transsexual people endanger the institution of the family. Transsexual people should be accepted completely into our society. I avoid transsexual people whenever possible. I would enjoy attending social functions at which transsexual people are present. I would feel comfortable if I learned my neighbor was transsexual.

PAGE 38

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 37 I would like to have friends who are transsexual people. I would feel uncomfortable if a close family member became romantically involved with a transsexual person. Romantic partners of transsexual people should seek psychological treatment. It would be beneficial to society to recognize transsexual people as normal. I would feel comfortable working closely with a transsexual person.

PAGE 39

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 38 Appendix B Noun/Male to Female As Jessie grew up, he often felt that he was a girl. However, he remembered how negatively his parents reacted to transgenders. Because of this, Jessie hid his feelings and made sure never to act like he was a tran sgender. He made sure to wear boy clothes and act like the other boys around him. Throughout middle school and high school he acted as masculine as he could, so others would not suspect him of being a transgender. After graduating high school, Jessie moved out of state for college. He lived in the dorms with a roommate, but sometimes when his roommate was gone Jessie would acknowledge his feelings as a transgender. He would lock the door and dress up in women's clothing. He loved wearing dresses and jewelry but he was afraid to tell anyone. He didn't want to come out as a transgender because he was afraid that his friends would make fun of him or even beat him up. In his last year in college, Jessie took a course on gender identification. After learning ab out transgenders, he began to wonder more about his gender identity. He presented more and more as a woman, sometimes even going out to bars dressed as a woman. The more he dressed as a woman, the more certain he became that he was a transgender. Jessie started dressing in women's clothing full time, and decided to legally change the gender listed on official IDs and forms. Jessie also some told friends about being a transgender, and asked to be referred as "she" instead of "he." Noun/Female to Male A s Jessie grew up, she often felt that she was a boy. However, she remembered how negatively her parents reacted to transgenders. Because of this, Jessie hid her feelings and made

PAGE 40

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 39 sure never to act like she was a transgender. She made sure to wear girl clot hes and act like the other girls around her. Throughout middle school and high school she acted as feminine as she could, so others would not suspect her of being a transgender. After graduating high school, Jessie moved out of state for college. She lived in the dorms with a roommate, but sometimes when her roommate was gone Jessie would acknowledge her feelings as a transgender. She would lock the door and dress up in men's clothing. She loved wearing suits and ties but she was afraid to tell anyone. She didn't want to come out as a transgender because she was afraid that her friends would make fun of her or even beat her up. In her last year in college, Jessie took a course on gender identification. After learning about transgenders, she began to wonder more about her gender identity. She presented more and more as a man, sometimes even going out to bars dressed as a man. The more she dressed as a man, the more certain she became that she was a transgender. Jessie started dressing in men's clothing ful l time, and decided to legally change the gender listed on official IDs and forms. Jessie also told some friends about being a transgender, and asked to be referred as "he" instead of "she." Adjective/Male to Female As Jessie grew up, he often felt that he was a girl. However, he remembered how negatively his parents reacted to transgender people. Because of this, Jessie hid his feelings and made sure never to act like he was transgender. He made sure to wear boy clothes and act like the other boys aroun d him. Throughout middle school and high school he acted as masculine as he could, so others would not suspect him of being transgender.

PAGE 41

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 40 After graduating high school, Jessie moved out of state for college. He lived in the dorms with a roommate, but sometim es when his roommate was gone Jessie would acknowledge his feelings as transgender. He would lock the door and dress up in women's clothing. He loved wearing dresses and jewelry but he was afraid to tell anyone. He didn't want to come out as transgender b ecause he was afraid that his friends would make fun of him or even beat him up. In his last year in college, Jessie took a course on gender identification. After learning about transgender people, he began to wonder more about his gender identity. He pr esented more and more as a woman, sometimes even going out to bars dressed as a woman. The more he dressed as a woman, the more certain he became that he was transgender. Jessie started dressing in women's clothing full time, and decided to legally chang e the gender listed on official IDs and forms. Jessie also told some friends about being transgender, and asked to be referred as "she" instead of "he." Adjective/Female to Male As Jessie grew up, she often felt that she was a boy. However, she remember ed how negatively her parents reacted to transgender people. Because of this, Jessie hid her feelings and made sure never to act like she was transgender. She made sure to wear girl clothes and act like the other girls around her. Throughout middle school and high school she acted as feminine as she could, so others would not suspect her of being transgender. After graduating high school, Jessie moved out of state for college. She lived in the dorms with a roommate, but sometimes when her roommate was gone Jessie would acknowledge her feelings as transgender. She would lock the door and dress up in men's clothing. She loved

PAGE 42

EFFECTS OF LANGUAGE ON ANTI TRANSGENDER PREJUDICE 41 wearing suits and ties but she was afraid to tell anyone. She didn't want to come out as transgender because she was afraid that her fr iends would make fun of her or even beat her up. In her last year in college, Jessie took a course on gender identification. After learning about transgender people, she began to wonder more about her gender identity. She presented more and more as a man, sometimes even going out to bars dressed as a man. The more she dressed as a man, the more certain she became that she was transgender. Jessie started dressing in men's clothing full time, and decided to legally change the gender listed on official IDs and forms. Jessie also told some friends about being transgender, and asked to be referred as "he" instead of "she."