Размещено на http://www.allbest.ru/

ПРАВИТЕЛЬСТВО РОССИЙСКОЙ ФЕДЕРАЦИИ

ФЕДЕРАЛЬНОЕ ГОСУДАРСТВЕННОЕ АВТОНОМНОЕ ОБРАЗОВАТЕЛЬНОЕ УЧРЕЖДЕНИЕ ВЫСШЕГО ОБРАЗОВАНИЯ

«НАЦИОНАЛЬНЫЙ ИССЛЕДОВАТЕЛЬСКИЙ УНИВЕРСИТЕТ “ВЫСШАЯ ШКОЛА ЭКОНОМИКИ”»

Факультет социальных наук

Образовательная программа “Психология”

Выпускная квалификационная работа

ПРЕДИКТОРЫ ДЕВАЛИДИЗАЦИИ ЖЕНСКОЙ БИСЕКСУАЛЬНОСТИ

Балезина Марьяна Евгеньевна

Москва 2019



Table of contents

Introduction

Chapter 1. Previous research on the devalidization of female bisexuality

1.1 Strategies of devalidization of female bisexual identity

1.2 Predictors of devalidization of female bisexuality

1.2.1 Demographic predictors

1.2.2 Psychological predictors

1.3 The present study

Chapter 2. Devalidization of bisexual identity: predictors and strategy usage

2.1 Method

2.1.1 A prior power analysis

2.1.2 Participants

2.1.3 Procedure

2.2 Measures

2.3 Data analysis

2.4 Results

2.4.1 Predictors of the female bisexuality devalidization strategies

2.4.2 Latent profile analysis

2.5 Discussion

2.6 Limitations

Conclusion

References



Introduction

While being a part of the LBGTQ+ community, bisexual individuals and bisexuality itself are rarely studied across sciences: queer theory ignores bisexuality because it somewhat breaks the concept of monosexual identities that is prevalent in queer theory (Mitchell & Erickson-Schroth, 2009), whereas in psychology and other social sciences bisexuals are rarely analyzed as an independent from lesbian and gay sample (Monro, Hines, & Osborne, 2017). The invisibility of bisexual identity exists not only in the science but also in real life, as it is rarely present in the media and the mainstream culture (Eisner, 2013). To describe the tendency to deny the existence of bisexual identity in this work the term `devalidization' The more well-established term invisibilization (Alarie & Gaudet, 2013) was not used since the work focuses on conscious strategies of denying the existence of bisexual identity and not on ignoring bisexual identity. is used as opposed to the concept of `external validation' that means emphasizing and affirming one's identity in the eyes of other people (Eliason & Schope, 2003). Previous studies have shown that practices of exclusion and devalidization of the bisexual identity are positively associated with the spread of biphobia among heterosexuals (Popova, 2018), homosexuals (McLean, 2008) and internalized biphobia and denial of identity among bisexuals (Hoang, Holloway, & Mendoza, 2011; Roberts, Horne, & Hoyt, 2015).

Despite the fact that devalidization of identity can have negative impact on bisexuals' psychological and social well-being (Brewster & Moradi, 2010; Dyar, Feinstein, & London, 2014; Hoang et al., 2011), a little is known about the ways, in which individuals devalidate bisexual identity. This study seeks to fill this gap and analyze the various strategies of devalidization of bisexual identity and their predictors.

There are some differences in the perception of bisexual men and bisexual women. Bisexual women are perceived heterosexual even while performing as bisexual (Fahs, 2009), whereas bisexual men are often thought to be just closeted gays (Welzer-Lang, 2008). This creates an interesting difference between experiences of bisexual men and women, where men are usually more excepted in the LBGTQ+ communities (Friedman et al., 2014), and women called “traitors” who “sleep with the enemy”, especially by lesbians (Hutsell, 2012; Lingel, 2009). Therefore, the devalidization of male and female bisexuality may differ and should be analyzed in separate ways. The focus of this study is female bisexuality in particular due to its prevalence in the population: previous studies showed that there are more women who identify as bisexual in the population, then men, at least in the US (Diamond, 1993; Ward, Dahlhamer, Galinsky, & Joestl, 2013). Additionally, women tend to perform bisexual behavior more often than men (Fahs, 2009; National Health Interview Survey, 2015). Therefore, the chance of women facing the devalidization of their sexual identity is higher.

Following hypothesis were formulated based on the theoretical review and the distinguished predictors:

H1: People with monosexual and/or binary identities would devalidate bisexual women to a higher extent, than people with non-monosexual and/or non-binary identities.

H2: Age of participant would be a valid predictor for devalidization of female bisexual identity, with people of lower age showing less devalidization.

H3: Perception of bisexual women as low on entitativity predicts higher devalidization.

H4: The higher is ambivalent sexism, the greater is devalidization.

In the first chapter of this work, previous research on the topic of bisexuality is discussed and the hypotheses are established and explained. The second chapter presents used methods and received results, followed by discussion and limitations. This work is forty-pages long, including three tables and one figure. References section lists seventy-eight cited sources.



Chapter 1. Previous research on the devalidization of female bisexuality

1.1 Strategies of devalidization of female bisexual identity

As mentioned above, previous scholars distinguished several ways in which one can devalidate female bisexual identity. Based on the studies where devalidization was described at the theoretical level (Barker & Langdridge, 2008; Diamond, 2005; Eisner, 2013; Israel & Mohr, 2004) or as the part of the biphobia scales (Alarie & Gaudet, 2013; Brewster & Moradi, 2010; Eliason, 1997; Fahs, 2009; Friedman et al., 2014; Lannutti & Denes, 2012; MacDonald, 1981; Mohr & Rochlen, 1999; Savin-Williams & Diamond, 2000) four different strategies of devalidization of female bisexual identity were highlighted.

According to the first strategy `Bisexuality as a phase' bisexual identity does not exist because bisexuality is considered to be a temporary phase, after which one would come back or transition to their true identity, which can be either heterosexual or homosexual (Alarie & Gaudet, 2013; Barker & Langdridge, 2008; Brewster & Moradi, 2010; Diamond, 2005; Friedman et al., 2014). In this case, devalidization represents the idea that female bisexual experience doesn't really matter, and women don't really need to form an identity around a temporary label.

The second strategy `Bisexuality as promiscuousness' includes the conviction that bisexuality is a synonym for a high sexual activity or an unfaithfulness to one's partner (Alarie & Gaudet, 2013; Spalding & Peplau, 2006). MacDonald (1981) stated that the most popular belief about bisexuals is that they are extremely promiscuous. The beliefs that bisexual individuals have an obsession with sex (Brewster & Moradi, 2010), that they are interested in threesomes (Alarie & Gaudet, 2013), and would have sex with anyone (Armstrong & Reissing, 2014) are also related to this strategy. In other words, in this case, bisexual women are perceived not as somebody with a unique social identity, but only as a sexually fluffed heterosexuals or lesbians.

The third strategy can be called `Bisexuality as an internalized homophobia'. According to this type of devalidization, bisexual female are just lesbians, who are too afraid to admit it (Eliason, 1997; Fox, 1996; MacDonald, 1981; Mohr & Rochlen, 1999). In lay thinking the perceptions of bisexuals include the idea that homosexuals `pretend' to be bisexuals because they want to pass as heterosexuals and thus avoid extremely negative attitudes towards themselves (Hoang et al., 2011; Nichols, 1988).

According to the fourth strategy `Bisexuality as a way to get (men's) attention', bisexuality perceived as an act that heterosexual women perform in order to satisfy men's sexual desires. This strategy is supported through practices of eroticization (Yost & Thomas, 2012), male gaze (Eisner, 2013) and an idea of compulsory bisexuality (Alarie & Gaudet, 2013; Fahs, 2009) that suggests that heterosexual women engage in homoerotic behavior with other women usually in front of men and in social settings such as student parties, bars, and clubs (Laumann, Gagnon, Michael, & Michaels, 1994). According to Fash (2009), compulsory bisexuality is a result of pressure that women feel when it comes to satisfying straight men. Bisexual behavior, in this case, does not seem significant in the context of sexual orientation, rather looked upon as a practice in which women are still read as heterosexual.

Together, female bisexuality can be devaluated based on various grounds, but are these strategies different when it comes to their predictors?

1.2 Predictors of devalidization of female bisexuality

1.2.1 Demographic predictors

Previous studies have demonstrated that the fluid nature of bisexuality as non-monosexual orientation and its transgression of gender binarism provokes stress, fear, unease, and confusion among monosexual individuals (hetero- and homosexuals) (Bucholski, 2015; Worthen, 2013) due to people's desire to simplify information and keep a clear distinction between social categories (Burke et al., 2017; Mohr & Rochlen, 1999). As a result, monosexual individuals express binegativity and try to deny one's bisexual identity (Bucholski, 2015; Rodriguez Rust. 2003). Several studies showed that heterosexuals and homosexuals held more negative attitudes towards bisexuality than other-identified individuals (e.g., asexuals), thus supporting the existing theoretical deductions (De Bruin & Arndt, 2010; Dodge et al., 2016; Mulick & Wright, 2003). At the same time, heterosexual men rate bisexual women more positively than bisexual men (G.M., 2002; Yost & Thomas, 2012) and homosexual women perceive bisexual women as `traitors' and, as a result, rate them more negatively than individuals with another sexual orientation (Hutsell, 2012). Overall, individuals' gender and sexual orientation seemed to be related to the attitude towards bisexual women, in particular, to whether bisexual identity is devaluated or not. In general, the assumption was that individuals' gender and sexual orientation would predict the devalidization of female bisexuality. In particular, those who believe in gender binarism and have a monosexual identity (hetero- or homosexuals) would devalidate female bisexual identity to a higher extent, than individuals with non-monosexual identity (Hypothesis 1).

Previous studies have shown that lower age is associated with more liberal attitudes when it comes to gender and sex roles (Morgan & Walker, 1983), as well as more positive attitudes towards sexual minorities and homosexuality (Dejowski, 1992; Hicks & Lee, 2006). Younger people tend to be more open to new experience, and more politically and religiously liberal (Shackelford & Besser, 2007). Bisexuality as a fluid and ambiguous sexual orientation might be looked upon as a protests to the current monosexual system, and therefore is supported by the liberal youth (Eisner, 2013). As a result, it was assumed that respondents' age would predict the devalidization of female bisexuality. In particular, the lower age would be connected with less devalidization (Hypothesis 2).

1.2.2 Psychological predictors

First of all, devalidization of female bisexuality is essentially the denial of the fact that bisexual women could be considered to be a unique group with their own norms and values. The degree to which a collection of persons are perceived as a real group or as a coherent unit is called `group entitativity' (Lickel et al., 2000). Previous studies demonstrated that entitativity is a key construct for understanding group perception and is associated with intergroup relations (Gulevich, Osin, Isaenko, & Brainis, 2016), stereotyping (Grzesiak-Feldman & Suszek, 2008), ingroup bias and prejudice (Haslam, Rothschild, & Ernst, 2002; Hodson & Skorska, 2015). Moreover, it was shown that entitativity is associated with identification (Castano, Yzerbyt, & Bourguignon, 2003; Castano, Yzerbyt, Paladino, & Sacchi, 2002; Crawford & Salaman, 2012; Hogg, Sherman, Dierselhuis, Maitner, & Moffitt, 2007). In other words, the association between group membership and identity might exist in one's perception: if some people are a group, they have to identify themselves as a group member. Therefore, devalidization of female bisexuality should be related to whether bisexual women are perceived as a group or not, in particular, the less perceived entitativity of bisexual women, the higher the devalidization of female bisexual identity (Hypothesis 3).

Next, bisexual women don't comply with a traditional image of women§ that exists within the gender belief system, which is a `set of beliefs and opinions about males and females and about the purported qualities of masculinity and femininity' (Deaux & Kite, 1987, p. 97). According to this system, women must be feminine and sexually prefer men, and bisexual women break this norm with their queer identity. As a consequence, bisexual women often feel double oppression: while dealing with prejudice based on their sexuality (e.g., biphobia), they also experience prejudice based on their gender, e.g., sexism and objectivization (Aosved & Long, 2006). Glick and Fiske (1996, 2001) in their Ambivalent sexism theory have described two complementary yet different forms of sexist attitudes toward women: hostile sexism and benevolent sexism. Hostile sexism is the subjectively negative attitudes toward nontraditional female types who are perceived as insubordinate to men and as rivals in the struggle for power (e.g., feminists). Benevolent sexism is the subjectively positive attitudes towards traditional female types who fit gender stereotypes (e.g., housewife) (Gaunt, 2013; Glick, Wilkerson, & Cuffe, 2015). Taken together, both types of sexism reinforce gender expectations from women through rewarding those who meet the expectations and punishing those who do not. Therefore, individuals with higher levels of hostile and benevolent sexism may devalidate female bisexuality to a greater extent, perceiving it only as a deviation from traditional gender norms and `normal' heterosexual identity (Hypothesis 4).

1.3 The present study

This study is the first attempt to systematically view the devalidization of female bisexual identity through considering different strategies of devalidization. It aims to understand what demographic and psychological variables predict the different strategies of devalidization of female bisexuality and are there real differences between them.

This research differs from previous studies conducted in the field of bisexuality in two ways. First, the study involved participants with different gender identity (men, women, and non-binary persons), as well as people with different sexual orientation (heterosexuals, homosexuals, bisexuals, pansexuals, asexuals). Second, the study was conducted in Russia, which is the country with prevailing traditional values, where recognition of the diversity of sexuality is still not the norm for most people. Previous studies have shown that among Russians there are many people who share the negative attitudes towards any non-heterosexual identities (Agadullina, Lovakov, & Malysheva, 2018; Gulevich, Osin, Isaenko, & Brainis, 2016, 2018). Moreover, there are the federal laws that forbid LGBTQ+ activism (Wilkinson, 2014) and create a system that legally supports invisibility of LGBTQ+ community in media, real life and academia. This social context can additionally contribute to the dissemination of devalidization's strategies and the perception of bisexuality as an unreal identity.



Chapter 2. Devalidization of bisexual identity: predictors and strategy usage

2.1 Method

2.1.1 A prior power analysis

A power analysis was conducted to compute the required sample size using the G*Power 3.1 program (Faul, Erdfelder, Buchner, & Lang, 2009). Since there are no relevant findings for predictors of devalidization strategies, the Cohen's (1988) recommendation about small effect size (f2 = 0.02) was used to compute the required sample size with sufficient power (1-в > 0.90). The total sample size was estimated at 430 participants.

2.1.2 Participants

A total of 4131 Russians began the survey. Some 768 participants were excluded from the sample because they did not complete the survey. Fourteen respondents who identified themselves as trans*women and 64 who identified themselves as trans*men as well as 87 respondents who identified their sexual orientation as `other' were excluded from the sample due to the inability to draw any meaningful conclusions based on data obtained from such small groups. As a result, the final sample consisted of 3200 participants (Mage = 22.88. SD = 6.19), among which 372 men, 2556 women, 272 non-binary persons; 664 respondents identified themselves as heterosexuals, 543 - homosexuals, 1443 - bisexuals, 444 - pansexuals, 106 - asexuals The dataset and full description of the final sample are freely available at Open Science Framework.

2.1.3 Procedure

The study was conducted in Russia. Due to the specific nature of the sample (part of the sample is LGBTQ+ participants), the link to the survey was posted in LGBTQ+ groups on social media (Vkontakte Russian social network (an equivalent of Facebook). The average daily audience of Vkontakte is over 80 million visitors, over 460 million users are registered (51 % women, the most represented age group of users is those aged from 18 to 34).) and in Telegram-channels Russian Telegram channels about queer culture, such as `Okay. Sobi', `Resurs' and so on that write about LGBTQ+ in Russia. Each participant willing to participate in the study read the short description of the questionnaire (introduction sheet). In the introduction participants were informed that the study is anonymous and does not carry any psychological or other threats to the individual. Participants agreed to participate in the study by going to the next page of the questionnaire. The first page contained the general demographics questions (gender, age, sexual orientation). The second page included questions related to the strategies of devalidization of female bisexual identity. On the third page, participants responded to the Ambivalent sexism scale and the question related to group entitativity, and on the last page, they were thanked.

2.2 Measures

Demographic. Participants completed the questions about their gender (male, female, trans*man, trans*woman, and non-binary person), age, and sexual orientation (heterosexual, homosexual, bisexual, pansexual, asexual, and other).

Devalidization of female bisexual identity was measured by four items, which corresponded to the main devalidization strategies described above (`Female bisexuality is only a temporary phase', `Female bisexuality is just synonymous with sexual promiscuity', `Bisexual women are lesbians who afraid to admit it', `Female bisexuality is just another way to attract the men's attention'). Ratings were made on a 7-point scale from 1 (absolutely disagree) to 7 (absolutely agree). The mean and standard deviation for each strategy are presented in Table 1.

Sexism was measured by Ambivalent Sexism Inventory (Glick & Fiske, 1996) adapted to Russian by Agadullina (2018). The statements about hostile sexism (HS) (e.g., `women seek to gain power by getting control over men') and benevolent sexism (BS) (e.g., `women should be cherished and protected by men') were rated on scale from 0 (disagree strongly) to 5 (agree strongly) (a = .89 and .89 for HS and BS respectively).

To measure the perceptions of group entitativity the item `Bisexual women can be considered as a group' (Agadullina & Lovakov, 2017) was used. Ratings were made on a 7-point scale from 1 (absolutely disagree) to 7 (absolutely agree) (M = 3.07, SD = 1.31).

2.3 Data analysis

To test the hypotheses, a hierarchical linear regression analyses (an ordinary least-squares method) was used, where different devalidization strategies were used as the dependent variables and demographic variables (age and the interaction between gender and sexual orientation) and psychological variables (hostile sexism, benevolent sexism, and group entitativity) as the predictors. Categorical variables (gender and sexual orientation) were dummy-coded. The reference category for gender was `women', for sexual orientation - `bisexuals'. These subgroups were chosen since the group `bisexual women' is an in-group for women and bisexuals; therefore, the values of the devalidization of these subgroups may differ from those obtained from respondents with other gender and sexual orientation. Age and the interaction between gender and sexual orientation were entered at the step 1. The psychological variables (hostile sexism, benevolent sexism, and group entitativity) were entered on step 2.

2.4 Results

Table 1 shows the descriptive statistics and correlations among the variables studied. The results demonstrated that the different strategies of devalidization of female bisexual identity are positively and moderately interrelated (all coefficients <.50). These findings mean that participants can simultaneously use different strategies for devalidization, even though they may contradict each other (for example, the belief that bisexual women are lesbians and that female bisexuality is a way to attract male attention). Interestingly, different strategies of devalidization are weakly associated with the perception of bisexual women as a unique group: the more respondents believed that female bisexuality is a temporary phase or an example of female promiscuous behavior the less they believed that bisexual women are a separate social group. At the same time, the perception of female bisexuality as internalized homophobia is positively related to group entitativity. Additionally, it was found that there are moderate positive associations between the strategies of devalidization and hostile and benevolent sexism.

Table 1. Mean, standard deviation, internal consistency and correlations among variables

	M	SD	б	1	2	3	4	5	6	7	8
BI as phase	1.88	1.02		-
BI as promiscuous	2.27	1.18		.473**	-
BI as male gaze	1.49	0.79		.470**	.489**	-
BI as internalized homophobia	1.91	1.07		.305**	.317**	.440**	-
BS	1.99	0.92	.89	.398**	.397**	.424**	.261**	-
HS	2.15	0.96	.89	.407**	.459**	.449**	.290**	.650**	-
7.Entitativity	3.07	1.31		-.033	-.103**	-.012	.044*	-.045*	-.099**	-
8.Age	22.88	6.19		.115**	.092**	.125**	.043*	.062**	.028	-.049**	-

Note: BI - bisexual identity, BS - benevolent sexism, HS - hostile sexism; **p < .01, * p < .0



2.4.1 Predictors of the female bisexuality devalidization strategies

Regression results are shown in Table 2. On step 1, demographic characteristics significantly predicted the devalidization of female bisexuality (R2 from 0.04 to 0.12 depending on devalidization strategy). Respondents' age predicted the perceptions of female bisexuality as a temporary phase, promiscuous behavior and `male gaze'. In particular, the older respondents, the stronger they support these devalidization strategies. Heterosexual men and women as well as homosexual men and women and bisexual men in compare to bisexual women more devalidizate of female bisexuality based on all strategies of devalidization. At the same time, pansexual women and pansexual non-binary persons less than bisexual women believed that female bisexuality is a temporary phase and promiscuous behavior.

Table 2. The results of hierarchical regression analysis.

Variable	BI as phase	BI as promiscuous	BI as male gaze	BI as internalized homophobia
	Coeff.	95 % CI	p-value	Coeff.	95 % CI	p-value	Coeff.	95 % CI	p-value	Coeff.	95 % CI	p-value
Step 1
Constant	1.373	[1.242;1.504]	<.001	1.807	[1.652;1.961]	<.001	1.063	[0.961;1.165]	<.001	1.69	[1.547;1.834]	<.001
Age	0.014	[0.008;0.019]	<.001	0.012	[0.006;0.019]	<.001	0.012	[0.007;0.016]	<.001	0.003	[-0.003;0.009]	.266
Inter: Heterosexual men (n = 183)	0.969	[0.821;1.118]	<.001	1.076	[0.901;1.251]	<.001	0.903	[0.788;1.018]	<.001	0.466	[0.303;0.628]	<.001
Inter: Homosexual men (n = 102)	0.364	[0.17;0.557]	<.001	0.457	[0.229;0.685]	<.001	0.356	[0.206;0.506]	<.001	0.412	[0.202;0.623]	<.001
Inter: Bisexual men (n = 64)	0.334	[0.094;0.574]	.006	0.699	[0.416;0.982]	<.001	0.615	[0.429;0.802]	<.001	0.391	[0.128;0.653]	.004
Inter: Heterosexual women (n = 447)	0.723	[0.623;0.823]	<.001	0.631	[0.514;0.749]	<.001	0.436	[0.358;0.513]	<.001	0.394	[0.285;0.503]	<.001
Inter: Homosexual women (n = 405)	0.278	[0.171;0.385]	<.001	0.202	[0.076;0.327]	.002	0.23	[0.147;0.312]	<.001	0.427	[0.311;0.544]	<.001
Inter: Pansexual women (n = 283)	-0.163	[-0.286;-0.041]	.009	-0.186	[-0.331;-0.042]	.011	-0.067	[-0.162;0.029]	.170	-0.132	[-0.266;0.002]	.053
Inter: Asexual women (n = 77)	-0.087	[-0.307;0.134]	.440	-0.073	[-0.332;0.187]	.583	0.128	[-0.043;0.299]	.141	0.211	[-0.029;0.451]	.085
Inter: Bisexual non-binary (n = 65)	0.025	[-0.214;0.264]	.837	-0.035	[-0.316;0.246]	.806	-0.059	[-0.244;0.126]	.531	-0.185	[-0.446;0.075]	.163
Inter: Pansexual non-binary (n = 141)	-0.249	[-0.415;-0.083]	.003	-0.223	[-0.419;-0.027]	.026	-0.094	[-0.223;0.035]	.152	-0.143	[-0.325;0.038]	.121
Adjusted R2	.121		.093		.123		.043
F	45.234	<.001	32.772	<.001	44.881	<.001	15.254	<.001
Step 2
Constant	0.515	[0.347;0.682]	<.001	0.878	[0.685;1.07]	<.001	0.267	[0.14;0.395]	<.001	0.761	[0.573;0.949]	<.001
Age	0.013	[0.008;0.018]	<.001	0.011	[0.005;0.017]	<.001	0.011	[0.007;0.015]	<.001	0.003	[-0.002;0.009]	.256
Inter: Heterosexual men (n = 183)	0.408	[0.261;0.556]	<.001	0.358	[0.188;0.527]	<.001	0.409	[0.297;0.521]	<.001	-0.028	[-0.193;0.138]	.742
Inter: Homosexual men (n = 102)	0.209	[0.026;0.391]	.025	0.225	[0.015;0.434]	.036	0.212	[0.074;0.351]	.003	0.248	[0.043;0.452]	.018
Inter: Bisexual men (n = 64)	0.012	[-0.215;0.24]	.915	0.236	[-0.025;0.497]	.076	0.329	[0.157;0.501]	<.001	0.099	[-0.156;0.353]	.448
Inter: Heterosexual women (n = 447)	0.41	[0.313;0.507]	<.001	0.228	[0.116;0.339]	<.001	0.162	[0.088;0.235]	<.001	0.127	[0.018;0.236]	.022
Inter: Homosexual women (n = 405)	0.326	[0.226;0.426]	<.001	0.26	[0.146;0.375]	<.001	0.271	[0.196;0.347]	<.001	0.467	[0.355;0.579]	<.001
Inter: Pansexual women (n = 283)	-0.078	[-0.193;0.037]	.183	-0.087	[-0.218;0.045]	.198	0.008	[-0.079;0.095]	.862	-0.061	[-0.19;0.067]	.352
Inter: Asexual women (n = 77)	-0.073	[-0.278;0.133]	.487	-0.04	[-0.276;0.196]	.739	0.135	[-0.021;0.291]	.090	0.199	[-0.031;0.429]	.090
Inter: Bisexual non-binary (n = 65)	0.021	[-0.202;0.244]	.854	-0.041	[-0.297;0.215]	.753	-0.064	[-0.233;0.105]	.458	-0.194	[-0.443;0.056]	.128
Inter: Pansexual non-binary (n = 141)	-0.239	[-0.394;-0.084]	.003	-0.186	[-0.364;-0.007]	.041	-0.091	[-0.208;0.027]	.132	-0.158	[-0.332;0.016]	.075
BS	0.211	[0.165;0.257]	<.001	0.194	[0.142;0.247]	<.001	0.176	[0.141;0.211]	<.001	0.147	[0.096;0.198]	<.001
HS	0.245	[0.201;0.29]	<.001	0.394	[0.343;0.445]	<.001	0.223	[0.189;0.257]	<.001	0.242	[0.192;0.292]	<.001
Entitativity	0.004	[-0.02;0.028]	0.719	-0.054	[-0.082;-0.027]	<.001	0.017	[-0.001;0.035]	0.072	0.062	[0.035;0.088]	<.001
Adjusted R2	.236		.250		.270		.125
F	77.080	<.001	81.871	<.001	90.547	<.001	35.102	<.001
Note: BI - bisexual identity. BS - benevolent sexism. HS - hostile sexism. Inter - interaction between gender and sexual orientation (reference group - bisexual women (n = 1314)). The subgroup pansexual men (n = 20). asexual men (n = 3). heterosexual non-binary persons (n = 4). homosexual non-binary persons (n = 36). and asexual non-binary persons (n = 26) were not included in the analyses due to small size and inability to make a reliable comparison with the reference group.

On step 2, the inclusion of psychological variables (hostile sexism, benevolent sexism, and group entitativity) in the regression equations in all cases significantly increased the quality of the models (ДR2 from 0.08 to 0.16 depending on devalidization strategy). Age was still a significant predictor for all devalidization strategies (except the strategy `Bisexual identity as internalized homophobia'), which, in general, confirms hypothesis 2.

The higher level of hostile and benevolent sexism independently predicted more approving all devalidization strategies and this result confirms the hypothesis 3. Individuals with monosexual identity (heterosexual men and women and homosexual men and women) stronger devalidizated the female bisexuality than bisexual women. Pansexual non-binary persons still less than bisexual women perceived bisexuality as a temporary phase. Taken together, these findings allow to accept hypothesis 1. Interestingly, after inclusion the hostile and benevolent sexism in the model, the differences in devalidization of female bisexual identity (excluding the perceptions of female bisexual identity as male gaze) between bisexual men and bisexual women disappeared.

Finally, the perceptions of bisexuals as a unique group positively predicted the perception of female bisexuality as internalized homophobia, however it negatively predicted the perception as a promiscuous sexual behavior. Since the group entitativity predicted only two of the four strategies for devalidization (in the opposite way), it is impossible to accept hypothesis 4.

2.4.2 Latent profile analysis

The correlation results support the notion that all of the strategies are connected, and the results of the regression analysis show that the strategies not only connected but are predicted by the same demographic and psychological characteristics. This leads to a question of strategies usage. A proposition can be made that the devalidization strategies are used altogether, rather than separately, despite their contradictory nature. In order to check this proposition, a latent profile analysis (LPA) was performed. LPA is based on the assumption that the correlations between the indicators may be explained by the existence of a categorical latent variable representing qualitatively and quantitatively distinct latent profiles of individuals within the sample (Morin et al., 2011). The LPA was conducted using the tidyLPA package for R (Rosenberg, Beymer, Anderson, & Schmidt, 2018). The tested models were estimated with 3000 random sets of start values, each of which was allowed 200 iterations, and the 200 best were retained for final stage optimization. The Bayesian information criterion (BIC; Schwarz, 1978) and entropy were used to identify the optimal number of the latent profiles. Those parameters indicate the precision with which the cases are classified into the various extracted latent profiles (Ramaswamy, Desarbo, Reibstein, & Robinson, 1993). A lower BIC value indicates a better fitting model and the larger and closer to 1 the value of entropy indicates the less classification errors in the model (Henson, Reise, & Kim, 2007; McLachlan & Peel, 2000).

LPA models, including one to five latent profiles, were consistently estimated on the whole data (N=3200). The fit indices for the LPA models are shown in the Table 3. A combination of BIC and entropy indices makes the model with three profiles the most suitable for the data.

Table 3. Fit indices for the LPA models.

Model	BIC	Entropy	Profile' size
1 Profile	36564.624	1	3200
2 Profiles	33754.718	0.959	2519/681
3 Profiles	31824.934	0.999	284/2053/863
4 Profiles	31465.792	0.967	864/1842/211/283
5 Profiles	31210.845	0.957	-/283/864/265/1788

The visual representation of the profiles in the best fitting model is illustrated in Figure 1. Profile 2 (N=2053) and 3 (N=863) represent balanced profiles. They don't deviate from the mean on more than one standard deviation, and therefore represent people who are only have a tendency to devalidate the bisexual identity to a greater or lesser extent. Profile 1, on the other hand, represent people who devalidate bisexuality to a greater extent, with a strategy of bisexuality as a way to get men's attention being the most prevalent, though not significantly. In terms of demographic characteristics, profile 1 had the highest percentage of men (30%) and mostly consisted of heterosexual participants (47%). Profile 3 mostly consisted of women (76%), bisexual and heterosexual participants made up almost 70% of the profile (38% and 30%, respectively). In profile 2, the prevalent majority were women (83%), and exactly half of people in the profile identified as bisexual. devalidation bisexual identity gender



Figure 1. Three profile model.

2.5 Discussion

The present study explored devalidization of female bisexuality and assessed its possible psychological and demographic predictors. The literature to date explores and describes different strategies of devalidization of bisexual identity, however, no systematic understanding of this process exists and very little is known about the relationship between these strategies and their predictors.

One of the most interesting finding of this research is the role of ambivalent sexism in predicting devalidization of female bisexual identity. The assumption was that since the traditional view on gender roles and expectations is fixed in sexist beliefs, bisexual women have to be perceived firstly as women, with bisexuality as their secondary characteristics due to instability and novelty of bisexual women as a group. It was found that benevolent and hostile sexism both lead to usage of all devalidization strategies, supporting the theoretical argument for ambivalent sexism as a uniform factor (Glick & Fiske, 1996). The lack of differences in the devalidization between men and women after controlling the sexism can be explained by the fact that men are more sexist than women (Glick et al., 2000).

The results provide additional evidence that participant's gender and sexual orientation could be significant predictors of devalidization of female bisexuality (Dodge et al., 2016; Rubinstein, Makov, & Sarel, 2013; Worthen, Lingiardi, & Caristo, 2017; Yost & Thomas, 2012). Individuals with monosexual identity maintain gender binarity and, therefore, negatively relate to anyone whose behavior cast doubt on the significance and necessity of existing gender system. Since monosexual individuals initially perceived others in the binary way, bisexual women are perceived primarily as members of traditional groups, who violate the group norms. For non-monosexual individuals, the binary gender system is not something essential; therefore, it is easier for them to perceive bisexual women as people with a unique social identity. It is important to note that some bisexual women, despite having a non-monosexual identity, still might accept the gender binary, therefore being more likely than pansexual and/or non-binary persons (who do not accept gender binary) to devalidate their own identity.

The results also confirmed that bisexual women are subject of double pressure, as their identity is not recognized both by heterosexuals and homosexuals. As previous research show, bisexual women are often thought to be heterosexual even while performing as bisexuals (Fahs, 2009). It might be argued that in the eyes of homosexuals (especially lesbian women), bisexual women perceived as a member of out-group, or as someone who violates the norms related to lesbians' community. This perception might explain the popular lesbians' stereotypes about bisexual women as those who `sleep with an enemy' (Rust, 1993) or as those who intermingle with lesbian politics (Eisner, 2013). For heterosexuals, bisexual women are those who violate a gender belief system, in particular, in the matter of sexual preferences and, therefore, also deserve a negative attitude.

Interestingly, for pansexual women, pansexual non-binary persons and bisexual non-binary persons, the interaction between gender and sexual orientation was not significant in both steps of the regression analysis. Despite a rich history of a debate between pansexual and bisexual community on the topic of biphobia (Eisner, 2013), resent discourse shifted towards `bisexual umbrella' as a description of a larger community for people with non-monosexual identities (Flanders, 2017). In this lay of thinking, pansexual participants might perceive bisexual individuals as members of the same community, who experience similar identity threats and discrimination. When it comes to non-binary persons, their experience as people who transgress the gender binary might affect their expectations of others in terms of traditional behavior (Callis, 2014): they don't comply with gender roles and do not expect others to do otherwise. As a result, it seems logical that individuals with non-monosexual and/or non-binary identities don't differ in the levels of devalidization of female bisexuality.

Surprisingly, devalidization was not connected with the perception of bisexual women as a group in the case of perception female bisexuality as a temporary phase and male gaze. However, entitativity positively predicted the perception of female bisexuality as internalized homophobia and negatively predicted perception of bisexuals as sexually promiscuous women. According to the strategy `Bisexuality as internalized homophobia', bisexual women are ranked among lesbians, consequently, their perceptions depend on the perceptions and attitudes towards lesbians. As a result, the more lesbians are perceived as a group with clear boundaries, which includes bisexual women, the more devalidization of bisexual identity based on this strategy. In other words, for individuals supporting the gender and sexuality binaries, devalidization can be connected to the entitativity of heterosexuals or homosexuals, but not to the perception of bisexual women as a separate group. The fact that age does not predict this strategy (unlike other strategies) may indicate that the belief that there are heterosexuals and homosexuals (as social groups) does not depend on age, in contrast to the attitude towards non-heterosexuals, which might become less liberal and benevolent with age.

Based on the additional analysis, conducted with the LPA, the strategies seemed to be working altogether when it comes to devalidization. Previous research has not looked into that question in particular, however, the results are in line with previous approach to devalidization as a part of overall biphobia, rather than an analysis of a distinct beliefs on bisexuality. Some previous work on the strategies suggest that they are somewhat situational and are based on the place bisexual behavior occur (Swan & Habibi, 2015). Therefore, all possible strategies and explanation tactics exist in one's perception, however, are used interchangeably, making it almost impossible for the belief to be staged by arguments. However, the results might have been the outcome of some limitations, that are discussed in the following section.

2.6 Limitations

The current study meets some limitations. First, the used sample was unbalanced and mostly consisted of women and bisexuals. Previous research proposed that women have more tolerant attitudes towards bisexual individuals than men (Dodge et al., 2016; Rubinstein et al., 2013; Yost & Thomas, 2012). Moreover, when it comes to the perception of female bisexuality, 35% of the sample answered the questions about their in-group, as a result, the in-group favoritism effect could have impacted the results obtained. In common, the sample of this study can be more tolerant than general population due to some specifics of the data collection. The link to the questionnaire was distributed through LGBTQ+ related sources, and it can be proposed that most of the participants were in some way familiar with bisexuality and issues surrounding it.

Second, the list of devalidization strategies of female bisexuality cannot be considered as final. Probably, individuals can use other strategies of devalidization, for example, Savin-Williams and Diamond (2000) have described the strategy that supports the idea of female bisexuality as a fluid by nature and something rather normal, but it was not included in the study due to the lack of researchers that support that explanation.

Lastly, since while completing the questionnaire participants saw all strategies at the same time, it might have led to a response bias, where participant did not choose their prevalent strategy but rather had a tendency to either agree or disagree with all devalidization strategies. Findings show that there are moderate correlations between all the strategies: it might be assumed that it was the result of a response bias or, as discussed above, devalidization might in reality be a one-factor belief, despite possible theoretical division. Participants in this study used the strategies altogether, despite a notional contradiction between them, and additional research is needed to confirm or deny the possible explanation presented.



Conclusion

The findings of the current study have significant implications for understanding attitudes towards bisexual women in the context of perceptions of female bisexuality. Despite being conducted on a sample, which mostly consists of members of the LGBTQ+ community, the data shows that overall people tend to devaluate bisexuality in women, which supports the existing discourse of the so-called `invisibilization' of the bisexual identity and its cross-cultural nature. This study is the first to implement psychological predictors, such as ambivalent sexism, to the field of bisexual studies, which proven to be significant for understanding the complexity behind biphobia. Another important theoretical conclusion, that was somewhat unexpected, is the combination of strategies, that tend to prevail in people's perception of bisexual women. Future research should look more closely into the strategy usage, their situational ground and possible semantic differences.



References

1. Agadullina, E. R. (2018). Sexism towards Women: Adaptation of the Ambivalent Sexism Scale (P. Glick and S. Fiske) on a Russian Sample. Psychology. Journal of Higher School of Economics, 15(3), 447-463.

2. Agadullina, E. R., Lovakov, A. V., & Malysheva, N. G. (2018). Essentialist beliefs and social distance towards gay men and lesbian women: a latent profile analysis. Psychology and Sexuality, 9(4), 288-304.

3. Agadullina, E. R., & Lovakov, A. V. (2017). Understanding entitativity: are there real differences between approaches? Psychology. Journal of Higher School of Economics, 14(3), 536-554.

4. Alarie, M., & Gaudet, S. (2013). “I Don't Know If She Is Bisexual or If She Just Wants to Get Attention”: Analyzing the Various Mechanisms Through Which Emerging Adults Invisibilize Bisexuality. Journal of Bisexuality, Vol. 13, pp. 191-214. https://doi.org/10.1080/15299716.2013.780004

5. Aosved, A. C., & Long, P. J. (2006). Co-occurrence of rape myth acceptance, sexism, Racism, Homophobia, Ageism, classism, and religious intolerance. Sex Roles, 55(7-8), 481-492.

6. Armstrong, H. L., & Reissing, E. D. (2014). Attitudes Toward Casual Sex, Dating, and Committed Relationships With Bisexual Partners. Journal of Bisexuality, Vol. 14, pp. 236-264.

7. Barker, M., & Langdridge, D. (2008). II. Bisexuality: Working with a Silenced Sexuality. Feminism & Psychology, 18(3), 389-394.

8. Brewster, M. E., & Moradi, B. (2010). Perceived Experiences of Anti-Bisexual Prejudice: Instrument Development and Evaluation. Journal of Counseling Psychology, Vol. 57, pp. 451-468.

9. Bucholski, M. (2015). Intersectionally Bi: The how and why of Bi-negativity.

10. Burke, S. E., Dovidio, J. F., LaFrance, M., Przedworski, J. M., Perry, S. P., Phelan, S. M., … van Ryn, M. (2017). Beyond generalized sexual prejudice: Need for closure predicts negative attitudes toward bisexual people relative to gay/lesbian people. Journal of Experimental Social Psychology, 71, 145-150.

11. Callis, A. S. (2014). Bisexual, pansexual, queer: Non-binary identities and the sexual borderlands. Sexualities, 17(1-2), 63-80.

12. Castano, E., Yzerbyt, V. Y., & Bourguignon, D. (2003). We are one and I like it: The impact of ingroup entitativity on ingroup identification. European Journal of Social Psychology, 33(6), 735-754.

13. Castano, E., Yzerbyt, V. Y., Paladino, M.-P., & Sacchi, S. (2002). I Belong, therefore, I Exist: Ingroup Identification, Ingroup Entitativity, and Ingroup Bias. Personality and Social Psychology Bulletin, 28(2), 135-143.

14. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

15. Crawford, M. T., & Salaman, L. (2012). Entitativity, identity, and the fulfilment of psychological needs. Journal of Experimental Social Psychology, 48(3), 726-730.

16. De Bruin, K., & Arndt, M. (2010). Attitudes toward bisexual men and women in a university context: Relations with race, gender, knowing a bisexual man or woman and sexual orientation. Journal of Bisexuality, Vol. 10, pp. 233-252.

17. Deaux, K., & Kite, M. E. (1987). Thinking about gender. In B. B. Hess & M. M. Ferree (Eds.), Analyzing gender: A handbook of social science research (pp. 92-117). Thousand Oaks, CA, US: Sage Publications, Inc.

18. Dejowski, E. F. (1992). Public Endorsement of Restrictions on Three Aspects of Free Expression by Homosexuals: Socio-Demographic and Trends Analysis 1973-1988. Journal of Homosexuality, 23(4), 1-18.

19. Diamond, L. M. (1993). Homosexuality and bisexuality in different populations. Archives of Sexual Behavior, 22(4), 291-310. Retrieved from

20. Diamond, L. M. (2005). `I'm Straight, but I Kissed a Girl': The Trouble with American Media Representations of Female-Female Sexuality. Feminism & Psychology, 15(1), 104-110.

21. Dodge, B., Herbenick, D., Friedman, M. R., Schick, V., Fu, T. C., Bostwick, W., … Sandfort, T. G. M. (2016). Attitudes toward bisexual men and women among a nationally representative probability sample of adults in the United States. PLoS ONE, 11(10), 1-18.

22. Dyar, C., Feinstein, B. A., & London, B. (2014). Dimensions of sexual identity and minority stress among bisexual women: The role of partner gender. Psychology of Sexual Orientation and Gender Diversity, 1(4), 441-451.

23. Eisner, S. (2013). Bi: Notes for a Bise...