Counter-Stereotypes in Entertainment Media

Master Thesis

Counter-Stereotypes in Entertainment Media

Daniella Janis Student #: 11374594 University of Amsterdam

Graduate School of Communication Entertainment Communication Supervisor: Dr. Jessica Piotrowski 26th June, 2018

Acknowledgement

I would like to thank all the individuals who helped me through the process of writing this thesis. First of all, I would like to thank Jessica Piotrowski my inspiring supervisor for her guidance throughout the process. I would also like to thank my incredible partner Tim for supporting me when times were tuff. My parents for teaching me how to be the best self I can be in this process. Furthermore, I could not have done it without all those who participated in the experimental survey. Finally, I would like to thank my friends for their never-ending encouragement and support.

Abstract

Counter-stereotypes are the conjunction of two non-overlapping (i.e. conflicting) categories. In past research, counter-stereotypes have been found to be an effective way to create positive effects among minorities such as Latino/a’s and black Americans and yet, the potential opportunities for supporting attributes of wellness among the LGBTQ community has been scarcely evaluated. To address this, an experiment was conducted to examined the effects of counter-stereotypical printed media messages on lesbian and gay individual’s state self-esteem (N=68). The potential mediating role of identification on state self-esteem, the potential moderating role of self-confidence on state self-esteem, and the combination of identification and self-confidence (moderated mediation) was also investigated. Participants were randomly assigned to either the control condition or the treatment condition. The control condition contained stereotypical media messages about a straight female make-up artist and a straight male professional skier. The treatment condition contained counter-stereotypical media messages about a lesbian female make-up artist and a gay male professional skier. As hypothesized, lesbian and gay individuals who viewed the counter-stereotypical content had higher levels of state self-esteem than individuals viewing the control stimuli. Individuals low in self-confidence when viewing the counter-stereotypical stimuli lead to low self-esteem. Identification did not mediate this overall effect. However, individuals, when viewing the counter-stereotypical stimuli, experienced lower self-esteem – a finding which is explained, in part, via identification. Implications are discussed.

Keywords: counter-stereotypes, self-categorization theory, social identity theory, stereotype

Literature Review Introduction

The high levels of discrimination that lesbian and gay individuals suffer within society is a leading cause of mental health problems among these groups. Lesbian and gay individuals are also more likely to inflict self-harm, attempt suicide, and experience suicidal thoughts than heterosexual individuals (Craig, McInroy, McCready, Alaggia, 2015). Disparaging stereotypes of gay and lesbian in the media reinforce discriminatory attitudes and behavior and can create a deep sense of alienation and marginalization within the community. At the same time, throughout the last 20 years’, lesbian and gay individuals are becoming more prominent in everyday media. Often portrayed as young, affluent, well-educated with interests limited to fashion and theater, these stereotypical images reflect little to capture the true lives of lesbian and gay individuals (McLnroy & Craig 2017). And problematically, the repetitive exposure to lesbian and gay stereotypes in the media can cause media consumers to believe these messages are true when in fact they are in not (Gerbner, 1998). Media reinforce attitudes and behaviors of cultural appropriateness within the community as well as how to act towards the community (McLnroy & Craig 2017; Berry, 2000).

Importantly, however, counter-stereotypical portrayals in the media are a powerful form of persuasion. They can supplant stereotypes and create positive effects within communities that have been discriminated against. A counter-stereotype is when an individual or group is classified by two somewhat unordinary classifications (Prati, Crisp & Rubini, 2015). Past studies have found that counter-stereotypical media messages contain a positive relationship between a viewer and those that are stigmatized with negative stereotypes (Morin-Messabe, Ferrire, Martinez, Devif, & Reeb, 2015; Pedulla, 2014). Previous research has also found that counter-stereotypical media can bring significant changes in behavior, attitudes, and state self-esteem in discriminated against minorities (Mckinley, Mastro &

Warber, 2014; Mastro & Tukachinsky, 2013; Simon & Hoyt, 2013; Ramasubramanian, 2011; Jennings-Walstedt, Geis & Brown 1980). Yet, the majority of this work has defined minority via racial groups – not sexual orientation. Moreover, the majority focuses on in-group views of out-group members, not how out-group members view themselves. To fill these gaps, this study is amongst the first to investigate the effects of counter-stereotypical printed media on lesbian and gay individuals’ state self-esteem.

Counter-Stereotypes

According to Prati, Crisp & Rubini (2015) “counter-stereotypes are characterized by the conjunction of two non-overlapping (i.e. conflicting) categories.” This means that in order to form an impression of someone, an unexpected fusion of social categories should be used to inhibit the use of the contradictory stereotypes generally associated with the inherent categories (Prati, Crisp & Rubini, 2015; Hastie et al., 1990; Kunda, 1990). In a study conducted by Hutter and Crisp (2005) participants exposed to counter-stereotypical combinations (male nurse), were more surprised and this created less stereotypical qualities to define the target in comparison to the stereotypical combination (female nurse). Morin-Messabel et al. (2015) and Pedulla (2014) further note that counter-stereotypical media messages create a positive association that decreases the connection between prejudicial attitudes and the negative stereotype.

Despite the theoretical argument in favor of counter-stereotypical messages, only a few studies thus far have demonstrated the effects of this specific process – and all have been within the realm of ethnic minorities and women. Some studies have found the positive effect of counter-stereotypes on women (Jennings et al., 1980; Fogliati & Bussey 2013; Simon & Hoyt, 2013). For example, Simon and Hoyt (2013) conducted an experiment in which they found that women who were exposed to mediated images of women in counter-stereotypical roles had greater leadership aspirations and less negative self-perceptions than women

exposed to women in stereotypical roles. Another study conducted by Marx, Ko, and Friendman (2009), it was found that after exposure to President Obama’s stereotype-defying speeches, African Americans overcame the so-called stereotype threat and instead where able to excel in post-viewing examinations. Moreover, Prati, Crisp, and Rubini (2015) found that counter-stereotype combinations can lead to less dehumanization and prejudices between intergroup members which can, in turn, support self-esteem (Blanton, Crocker & Miller, 2000).

Experimental research exploring the effects of stereotypical content on lesbian and gay individuals is even more limited. The existing work has explored the positive effects of counter-stereotypical content on heterosexual’s beliefs on lesbian and gay individuals. Further, a study conducted by Gillig and Murphy (2016) found that a narrative containing two young gay characters coming to understand their identity created positive emotions of hope, pathway goals, and an elevated sense of mental energy of lesbian and gay individuals. People take pride of their in-group by association when members of their in-group have an uplifting experience. This suggests that counter-stereotypical printed media can also positively influence lesbian and gay state self-esteem.

Theoretical Processes

Social Identity Theory and Self-Categorization Theory

To understand the development of self-esteem as it relates to others, it is helpful to understand social identify and self-categorization theory. Social identity theory is a core concept of self-categorization theory. It is defined by Tajfel (1972) as “the individual's knowledge that he belongs to certain social groups together with some emotional and value significance to him of this group membership". Social categorization is when an individual inherently defines their own place in society (Hogg & Terry, 2000; Tajfel, 1972). Social identity creates intergroup social comparisons that motivate individuals to establish and

confirm to an in-group by favorable judgments that are distinct between the in-group and outgroup, which is motivated by the fundamental desire for self-esteem (Hogg & Terry, 2000; Turner, 1975).

Self-Categorization theory assumes that individuals instinctively view a social situation that concerns two different groups. There is the in-group that is seen as similar to the self and there is the out-group that is viewed differently from the self. This categorization inherently causes individuals of the in-group to create common stereotypes by making inferences about the out-group’s behavior. When an individual categorizes themselves, it can create a group identity of how they view themselves when they are within this in-group. (Hewner & Knobelsdorf, 2008). As in-group identification grows, the incentive to protect the interests and status of that group also prospers (Mastro, 2009; Verkuyten & Brug, 2004). Group prototypes also allows individuals to evaluate themselves and others leading individuals to adjust their beliefs and behaviors which also affects their esteem and self-confidence (Mastro & Kopacz, 2006).

In the media space, Mckinley, Mastro & Warber (2014) found that positive media depictions of Latinos created favorable evaluations of in-group. The study proposes that if media promotes positive media images of certain minorities it can improve in-group member’s attitudes. More specifically, past research suggests that due to self-categorization, media can influence in-group state self-esteem (Mckinley, Mastro & Warber, 2014; Mastro, 2009). In as much, it can be reasonable to expect that if lesbian and gay individuals are exposed to positive counter-stereotypes, it could affect their state self-esteem as they are already established within an in- group that has been negatively represented throughout the history of media (McInroy & Craig, 2017). Positive counter-stereotypical representation of this group could create positive attitudes about being part of the lesbian and gay community. This could allow more individuals to have higher self-esteem and feel proud of being part of

this community. As such, the current study hypothesizes that counter-stereotypical printed media can have a positive effect on lesbian and gay individuals’ state self-esteem.

H1: Lesbian and gay individuals that view counter-stereotypical content will have higher levels of state self-esteem than individuals viewing control stimuli.

Identification as Mediator

Identification is an imaginative process in which a viewer bonds or connects with another person or entity causing that individual to undertake the perspectives, goals, and identity of that person’s characteristics into their sense of self (Hoffner, 2005; Cohen, 2001). Identification is when an individual imagines them to be someone else, and during this process the individual forgets who they are and becomes temporarily another. In media effect studies, identification is perceived as an association between exposure and impact (Hoffner and Buchanan, 2005; Cohen, 2001). Identification with media exposure can lead an individual to temporarily adopt an outward perception of the world (Cohen, 2001). Hoffner and Buchanan (2005) stated that media viewer’s identification to media models is one of the many outcomes that comes with television consumption, and is assumed to mediate the responses of the audience. People identify with groups and/or an individual in order to achieve valued outcomes or rewards. Maximizing their own potential, forming interpersonal connection, and or enhancing their self-esteem is part of the reason why people identify with individuals or groups (Hoffner, 2005). Wilcox & Laird (2000) found that when women identify with a thin model in a picture it can lead to a brief moment of pleasure. If lesbian and gay individuals identify with positive counter-stereotypes, it could similarly lead to positive state self-esteem. As such:

H2: Identification will mediate the relationship between counter-stereotypical messages (CSM) and state self-esteem such that CSM leads to increased identification which subsequently predicts increased state self-esteem.

Self-Confidence as a Moderator

Self-confidence is an individual’s perception of their skills and competence, their known ability of how to handle numerous situations successfully (Shrauger & Schohn, 1995). It is also considered to be one of the most powerful regulators and motivators of behavior in individual’s everyday lives. As it is a judgment of one owns accomplishments and capabilities of a goal, it must also be considered a motivator that provides the goal with context (Dumitrescu, Zetu, & Teslaru, 2011). Self-confidence is associated with depression, interpersonal conflicts, anxiety, gender, happiness, stress, loneliness, and physical appearance (Dumitrescu et al., 2011; Shrauger & Schohn, 1995). A study found that women with lower self-confidence were affected stronger by tokenism (Cohen & Swim, 1995). Therefor it can be assumed that individuals with high self-confidence may not be as affected by CSM as those with low self-confidence. In as much, it be may the case that self-confidence influences the extent to which media messages influence state self-esteem. In particular, it may be the case that individuals with high self-confidence will not be as affected by counter-stereotypical media as peers with lower self-confidence.

H3: Self-confidence will moderate the relationship between counter-stereotypical printed media messages and esteem such that individuals with lower self-confidence will experience particularly pronounced effects.

H4: The influence of self-confidence on self-esteem via CSM will be mediated by identification (i.e., moderated mediation; see conceptual figure appendix D.2).

Method – Pilot Study Design

Prior to conducting the main study, a pilot study was required to confirm that the stimulus material being used is counter-stereotypical. Ten lesbian and gay individuals took part in the pilot study to examine if the stimulus material being used represents lesbian and

gay counter-stereotypes. These participants were not used in the main study to prevent bias. These individuals were shown counter-stereotypical printed images followed by a small survey. The small survey addressed 1) whether the printed images produced positive or negative emotions and 2) whether the printed images embodied counter-stereotypical representation of lesbian and gay individuals and not negative stereotypes.

Stimulus Material

In the pilot study, four counter-stereotypical images of lesbian and gay communities were used. The images contain a lesbian Politian, a gay freestyler skier, a gay astronaut, and a lesbian make-up artist. These four categories represent counter-stereotypes in current society as they are the complete opposite of stereotypes such as a heterosexual female/male politician, a heterosexual man who participates in extreme action sports such as free skiing, heterosexual female/male astronaut and a gay male makeup artist or heterosexual female make-up artist. These categories represent two non-conflicting categories that are not the norm in everyday life (Prati, Crisp & Rubini (2013).

Measures

Counter-stereotypical Assessment. Participants were asked to indicate whether the

images were stereotypical or counter-stereotypical. Specifically, “Look at the image of Thomas Kennington/ Brookelle Jaffary/ Allison McKinley/Steven Goldberg above. Would you consider this image to be counter-stereotypical or stereotypical?” Respondents could answer in three different ways as follow: the image is clearly counter-stereotypical, the image is clearly stereotypical, the image has both counter-stereotypical and stereotypical components.

Procedure

Participants were sent the online survey via Qualtrics. Participants were asked to consent to the short pilot study. Thereafter, participants were shown the eight images. The

images consisted of four stereotypical printed images of a female politician, male astronaut, female make-up artist, and male skier while the other four images consisted of counter-stereotypical printed media message consisting of their counter-stereotypical counterpart (lesbian female politician, gay male astronaut, lesbian female make-up artist, and gay male skier). They were then asked questions regarding if these images were counter-stereotypical. Afterwards, participants were briefed on why the pilot study was conducted.

Method – Main Study Design

This study uses a between-subjects experimental design (control group vs treatment group) in order to compare the effects of counter-stereotypical printed media messages on lesbian and gay individuals’ state self-esteem. A web-based experiment was chosen because it can easily be sent to large groups of people via Facebook groups, messenger, and Whatsapp making it fast and easy. The experiment programmed and distributed via Qualtrics.

Sample

A convenience snowball was used for this study. Lesbian and gay participants were gathered via personal connection and social networking sites. Personal connections were then asked if they knew any individuals within this community that would want to help out with the survey. Participants were randomly assigned to either the control group or the treatment group (Bryan, 2012) – resulting in 30 respondents in the treatment condition and 38 in the control condition. Out of the lesbian and gay sample, 54.7% were gay (n = 41) while 45.3% (n = 34). On average, participants were 30 years old.

Stimulus Materials.

The stimulus material is inspired by a study done by Simon & Hoyt (20120). In their study, they used counter-stereotypical portrayals of women depicted in printed advertisements. For this study, as the participants are from the lesbian and gay community,

counter-stereotypical content was used in the treatment group and gender stereotypical content was used in the control. The images were identical in both groups with the manipulation occurring at the text level (Appendix B). In the treatment group, the featured image was described as homosexual whereas in the control group the image was described as heterosexual. All images were found online and edited via Photoshop.

In the pilot study, it was concluded that - for the main study - the best images would be the make-up artist and the skier. Pilot survey results are depicted in the table below. As the table shows, the make-up artist and skier were used in the main study as most individuals found the counter-stereotypical images counter-stereotypical while most individuals found the stereotypical images stereotypical.

Table 1.

Pilot Survey Results

Stimulus Stereotypical (percentages) Counter-Stereotypical (percentages) Both (percentages) Total Make-Up Artist Treatment 20% 50% 30% 100% Politician Treatment 0% 50% 50% 100% Skier Treatment 0 40% 60% 100% Astronaut Treatment 20% 40% 40% 100% Makeup Artist Control 100% 0 0 100% Skier Control 60% 30% 10% 100% Astronaut Control 30% 20% 50% 100% Politician Control 20% 30% 50% 100% Measures

Control Variables: Gender, Sexual Orientation, and Age. Gender was measured using the question “what is your gender?” in which respondents could select from male, female, transgender, non-binary, and or prefer not to respond. The sexual orientation scale was derived from Bidell’s (2005) sexual orientation counselor competency scale. The

question asked was “what is your sexual orientation?” where respondents could answer with the following answer options: heterosexual, lesbian, gay, bisexual, queer and prefer not to respond. Respondents whose sexual orientation was not either lesbian or gay were filtered out. Age was operationalized using an open-ended question “What is your birthdate?”. Respondents were then asked to type in their age in the month, day, year format.

Moderator Variable: Self-Confidence. Self-Confidence was operationalized using a 7 point Likert scale ranging from 1 strongly disagree to 7 strongly agree using questions such as “I have more confidence than most people I know”. This scale is derived from Shrauger and Schonn’s (1995) Personal Evaluation Inventory Scale. The original scale contains many components such as mood and athletic ability. However, for the current study, the main concept of interest was the participants’ general confidence. Therefore, only questions measuring general confidence were used. In practice, this meant that six items were administered (Appendix A.1). Reliability of the items was sufficient - Cronbach's alpha  = .78 (M = 3.74 SD = 1.01) (Appendix C.2).

Moderator Variable: Identification. There is not one particular scale that measures identification so the scale was created with inspiration from Brown and Basil’s Identification scale (1996). The identification scale measures how much a person likes another person, whether they have feelings for that person, if they think of that person as a friend, whether they could relate to them, if that person would be a role model, and how well they could work together (Brown & Basil, 1996). This was operationalized using a 7 point Likert scale ranging from 1(strongly disagree) to 7 (strongly agree). Some questions were altered as the questions from Brown and Basil’s identification scale relate directly to Magic Johnson, while the questions here pertain to each character in the stimulus material. An example question is “I can easily relate to Brookelle Jaffary/Thomas Kennington.” (Appendix A.4) Reliability of the analysis of the 8 identification variables for the Brookelle Jaffary scale was good, Cronbach's

alpha  = .90 (M = 32.31, SD = 12.78). Similarly, reliability of the analysis of the 8 identification variables for the Thomas Kennington scale was good, Cronbach's alpha  = .93 (M = 33.20, SD = 12.28). Reliability of the analysis of all 16 identification variables was also good, Cronbach's alpha  = .95 (M = 65.51, SD = 24.45). Both the Brookelle latent variable and Thomas latent variable correlate highly with one another (r = .90) and, as such, were combined to represent one identification score (Appendix C.3).

Dependent Variables: State Self-Esteem. State Self-Esteem was operationalized using a 7 point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). Using only one subscale from Heatherton and Polivy’s (1991) social state esteem scale, this subscale focused on more general attributes of self-esteem as opposed to context specific items. An example question is “I feel self-conscious”. The state self-esteem scale also used questions inspired by Rosenbergs self-esteem scale (1965). An example question being “Right now I wish I could have more respect for myself” (Appendix A.5). All items load together as the reliability of the scale is good, Cronbach's alpha  = .95 (M = 4.51, SD = 1.17) (Appendix C.1).

Procedure

The current study was conducted online using Qualtrics. At the start of the survey, participants were asked to confirm consent which emphasizes the nature of this study and that their participation is completely anonymous and voluntary. Demographic questions regarding nationality, gender, and age were asked as well. Participants were then asked questions regarding self-confidence, which was then followed by the main study that begins by showing the participants either the control condition stimulus material or the treatment condition stimulus material based on random assignment. In each condition, participants then completed the identification scale followed by questions about state self-esteem.

Descriptive Statistics

A randomization check was conducted to check if participants were randomly assigned to each of the conditions (control group vs treatment group) using a chi-square test. The condition variable was run through the chi-square test. A chi-square test of goodness-of-fit was performed to determine whether lesbian and gay individuals were equally distributed among the condition. Lesbian and gay individuals were equally distributed in the population,

X2 (1, N = 75) = 1.70, p = .192.

This concludes that participants were randomly assigned to each condition. In total, there were 75 participants (n = 75) however only 90.6% (n = 68) of participants made it till the end of the survey (as described in the table below). Calculated mean responses for each measure and the descriptive statistics are shown in the table below.

Table 2.

Descriptive statistics for the two stimulus conditions

Control Condition (0) (n = 38) Treatment Condition (1) (n = 30) Gays (n = 23) Lesbians (n=15) Gay (n=16) Lesbians (n = 14) M (SD) M (SD) M (SD) M (SD) State self-esteem 3.70 (.93) 4.04 (1.37) 4.13 (1.40) 4.98 (1.33) Identification 4.35 (.79) 4.03 (1.05) 4.85 (.77) 4.73 (.98) Self-confidence 3.76 (.65) 3.86 (1.28) 3.71 (1.10) 3.79 (1.41)

Hypothesis 1 predicted that lesbian and gay individuals that view counter-stereotypical content will have higher levels of state self-esteem than individuals viewing control stimuli. To check if participants’ state self-esteem was comparable over the conditions, a One-way ANOVA was conducted. This ANOVA had the condition variable

messages) as the independent variable, and state self-esteem as the dependent variable. Supporting hypothesis 1, state self-esteem in the counter-stereotypical group (M =4.52, SD = 1.41) was significantly higher than the stereotypical group (M = 3.83, SD = 1.12), F (5, 1) = 66, p < .05.

Hypothesis 2 predicted that identification will mediate the relationship between the condition and state self-esteem. This hypothesis was tested using PROCESS MODEL (Preacher & Hayes, 2008 PROCESS Model 4). There was no statistically significant evidence of a mediation effect, rejecting Hypothesis 2. While the model as a whole is significant, F (7.55, 1) = 66.00, p < .05, R2 = .32, there is a non-significant indirect effect of condition on state self-esteem through identification (b* = .20. t =1.20, p = .23, 95% CI [-.13, .53]).

The third hypothesis predicted that self-confidence will moderate the relationship between the condition and state self-esteem such that individuals with lower self-confidence viewing counter-stereotypical content will experience particularly pronounced effects on state self-esteem. This was tested using the PROCESS MODEL (Preacher & Hayes, 2008 PROCESS Model 1). The model as a whole was significant, F (11.3, 3) = 64.00, p < .001, R2 = .58. Both condition (b* = 4.77 t =- 4.9, p < .001, 95% CI [-6.71, -2.87]) and self-confidence (b* = 2.04, t = 4.9, p <.001, 95% CI [6.71, 2.87] have a positive effect on state self-esteem. The results showed the interaction of self-confidence with condition, b* = -1.07 t = 4.3, p < .001, 95% CI [.57, 1.57]. There was a negative effect of self-confidence on state self-esteem such that individuals with lower self-confidence (b* = 1.39 t = 4.56, p < .001, 95% CI [.78, 2.00]) seem to experience low self-esteem after viewing CSM (Conceptual model Appendix D.1)

The fourth hypothesis predicted that the influence of self-confidence on self-esteem via CSM will be mediated by identification. This was tested using the PROCESS MODEL (Preacher & Hayes, 2008 PROCESS Model 14). The results showed that there was a

significant positive direct effect of counter-stereotypical printed media messages on state self-esteem b* = .59, t = 2.14, p < .05, 95% CI [.04, 1.15]. The results also show that there is a significant positive relationship between counter-stereotypical messages and identification b* =.62, t = 2.74, p < .05, 95% CI [.17, 1.08]. Within the moderation-mediation model identification was found to have a positive significant effect on state self-esteem b* = 1.34, t = 4.33, p < .01, 95% CI [-.73, 1.97. Accepting hypothesis 4, the influence of self-confidence on selfesteem via CSM was found to be significant when mediated by identification (b* = -.36, t = -4.13, p< .01, 95% CI [-.53, -.18) There we no significant conditional effects of low, medium or high self-confidence.

Discussion

The purpose of this study was to examine the effects of counter-stereotypical printed media messages on lesbian and gay individuals’ state self-esteem. The effects of counter-stereotypes were also investigated through a potential mediating role of identification on state self-esteem, a potential moderating role of self-confidence on state self-esteem, and the combination of identification and self-confidence (moderated mediation). It was hypothesized that lesbian and gay individuals that view counter-stereotypical content will have higher levels of state self-esteem than individuals viewing control stimuli.

Results were somewhat mixed. Encouragingly, the study showed that lesbian and gay individuals who were exposed to counter-stereotypical printed media messages held higher levels of self-esteem after viewing than individuals who viewed stereotypically printed media messages (H1). There was no evidence to suggest that identification mediated this process and, contradicting hypothesis 3, self-confidence worked differently than anticipated. Specifically, we see a negative effect of self-confidence on state self-esteem such that individuals with lower self-confidence seem to experience depleted self-esteem after viewing counter-stereotypical messages. And, coinciding with this, this pattern was echoed in test for

hypothesis 4 whereby the mediating role of identification was significant when self-confidence moderated identification leading to lower levels of state self-esteem.

Studies have not previously explored the effects of counter-stereotypical printed media messages on lesbian and gay individuals’ state self-esteem. Therefore, this study, joining other more recent pieces (Mckinley, Mastro & Warber, 2014; Mastro & Tukachinsky, 2013; Simon & Hoyt, 2013; Ramasubramanian, 2011; Jennings-Walstedt, Geis & Brown 1980), highlights ways that that the media space can help – rather than harm – members of marginalized communities. In particular, the study offers empirical support that counter-stereotypical printed media messages have a (short-term) positive effect on lesbian and gay individuals’ state self-esteem. From the perspective of classic ‘drip drip’ theories a la cultivation theory and others (Gerbner, 1998), this suggests that – with time – a consistent ‘dose’ of counter-stereotypical content could go a long way towards producing positive well-being among the specific marginalized community of interest (i.e., here well-being lesbian and gay individuals). This makes intuitive sense – just as stereotypical representations can harm community members – the reverse may have the counter effect (Mckinley, Mastro & Warber, 2014; Mastro & Tukachinsky, 2013; Simon & Hoyt, 2013; Ramasubramanian, 2011; Jennings-Walstedt, Geis & Brown 1980).

That said, this study says nothing about audience members that are not members of the particularly marginalized community. So, for example, in this study, we do not know how these counter-stereotypical representations might affect heterosexual audience members. A past study found that after participants were shown gender counter-stereotypical media messages which caused the participants to have decreased perceptions associated with derogation and dehumanization of target groups as well as unrelated target groups, which was argued to be a result of a reduction in heuristic thinking (Prati, Crisp, & Rubini, 2015). The same study found that exposure to counter-stereotypes reduced discrimination and created

positive attitudes on intergroup emotions. Such findings are encouraging, and in the context of this study, suggest that future research would be wise to look at the effects of counter-stereotypical media content from an outgroup perspective as well.

Boundary Conditions of Effects

While the main effects uncovered in this study are perhaps the most encouraging, it is also important to recognize that an attempt was made to better understand the process of these effects via identification and self-confidence. In terms of identification, somewhat surprisingly there was no evidence for statistical mediation. Initially, it was expected that identification will mediate the relationship between counter-stereotypical message and state self-esteem – with CMS content increasing identification and subsequently increasing state self-esteem. But alas – this expectation was not supported. There are two reasons to explain this. On the one hand, it may simply be the case that identification is not a key variable in this process. On the other hand, it may be that identification towards an advertisement is not powerful enough to induce effects, particularly in a one-off study. Most studies of identification focus on television rather than printed messages and assume a lengthy time to develop a relationship with the characters (Bond & Drogos, 2014; Hoffner & Buchananm 2005). The short exposure duration combined with printed media format may simply have been insufficient to promote any identification.

Moreover, it is also important to recognize that the identification scale was derived from Brown and Basil’s identification scale which relates directly to Magic Johnson - a widely known public figure that was open about his issue with AIDS. The identification scale closely relates to a narrative as Magic Johnson story was covered heavily in magazines and newspapers. This identification scale was applied to an image rather than a narrative making it harder for individuals to relate to the characters. These media messages gave a small message but there was no meaningful narrative involved. Perhaps this made it difficult for

individuals to identify with the characters in the images. If any identification did result, the measure used may have been insufficient to capture this process.

Also, just as identification turned out to be a null predictor in the study, self-confidence seems to work in an unexpected direction. Recall that initially it was expected that individuals with low confidence with moderate the relationship condition and state self-esteem such that they will experience particularly robust benefits of counter-stereotypical stimuli. But, instead, here we see a negative effect of self-confidence on state self-esteem such that individuals with lower self-confidence seem to experience depleted self-esteem after viewing counter-stereotypical messages. Based on past research, a study has found that when women viewed images of fit models instead of actually inspiring them to conduct a healthy life style it had the opposite effect which led to negative moods as well as negative state appearance self-esteem (Tiggerman & Zaccardo, 2015). Perhaps this is what could have happened here. That perhaps the individuals that view counter-stereotypical media messages of successful lesbian and gay individuals who already have low self-confidence could lead them to feeling worse about themselves than better about themselves causing lower levels of state self-esteem. That being said, the images may have had the reverse effect instead of inspiring individuals it made them feel more upset with themselves. Future research should look into why it led to a decrease in self-esteem when self-confidence moderated the relationship. The mediation moderation results showed that when self-confidence moderated the mediation effect it was significant. However just like in hypothesis two, even when self-confidence moderate’s identification it decreases state self-esteem. It seems as though whenever self-confidence moderates a relationship whether direct or indirect it leads to individuals having less state self-esteem after viewing counter-stereotypical printed media messages. However, in hypothesis 4 it did not matter whether individuals had low or high self-confidence, viewing the counter-stereotypical media messages led these individuals to

have a depleted sense of state self-esteem even when self-confidence moderated the mediation effect of identification. The mediation effect seems only to be present for a certain group of people where identification mediates self-confidence. This means that there must be a different process at work for the other individuals than identification. Future research to look to find what the different process could potentially be.

Limitations and Future Research

Out of 75 participants only 68 individuals fully completed the study. In as much, the total sample contained a low number of respondents creating difficulties in generalizability. Future research should gather a larger generalizable sample to better support external validity.

The main study of this research contained stimuli that was not found by lesbian and gay individuals cohesively to be largely stereotypical or counter-stereotypical. The results varied from image to image making it difficult to choose the proper stimuli for the main study. Stimuli in future research should be found to be highly stereotypical and highly counter-stereotypical to ensure robust contrasts. Future research should also take into account sexual differences within the stimuli and how the effects could be different from those of the same sex to the main character and to those of the opposite sex of the main character. There could be significant differences between male and females who view stimuli of the same and opposite sex. In this study, lesbian and gay individuals were grouped collectively and viewed both stimuli of male gay individuals and female lesbian individuals.

Due to ethical limitation within the pilot study participants were not ask if stereotypical images offended them. The study did not want to make people think of how an image could possibly hurt them as that violates the code of ethics of offending them which this study did not want to do. However, the study was interested in understanding why some stereotypical images could produce negative effects rather than positive effects. But due to the ethics it was opted not to ask participants why stereotypical images would offend them. For

future research, it would perhaps be ideal to hold a focus group about counter-stereotypical and stereotypical images to see how lesbian and gay individuals react within the boundaries of ethics.

Conclusion

The current study contains empirical evidence that lesbian and gay individuals who view counter-stereotypical media messages experience increased state self-esteem compared to their non-CSM peers. This is an important finding that is deserving of replication and extension. Replication will identify the robustness of this finding while extension will help us identify the boundary conditions of this effect. As gay and lesbian representation continues to rise in mass media, opportunities to support well-being among members of this community (and sensitivity among out-group members) should not be underestimated.

References

Berry, G. L. 2000. “Multicultural Media Portrayals and the Changing Demographic Landscape: The Psychosocial Impact of Television Representations on the Adolescent of Color.” Journal of Adolescent Health 27 (2): 57–60. doi:10.1016/S1054-139X (00)00133-6. Bidell, M. P. (2005). The sexual orientation counselor competency scale: Assessing attitudes,

skills, and knowledge of counselors working with lesbian, gay, and bisexual clients. Counselor Education and Supervision, 44(4), 267-279.

Blanton, H., Crocker, J., & Miller, D. T. (2000). The effects of in-group versus out-group social comparison on self-esteem in the context of a negative stereotype. Journal of

Experimental Social Psychology, 36(5), 519-530.

Bond, B. J., & Drogos, K. L. (2014). Sex on the shore: Wishful identification and parasocial relationships as mediators in the relationship between Jersey Shore exposure and emerging adults' sexual attitudes and behaviors. Media Psychology, 17(1), 102-126. Brown, W. J., & Basil, M. D. (1995). Media celebrities and public health: Responses

to'Magic'Johnson's HIV disclosure and its impact on AIDS risk and high-risk behaviors. Health Communication, 7(4), 345-370.

Cohen, J. (2001). Defining identification: A theoretical look at the identification of audiences with media characters. Mass communication & society, 4(3), 245-264.

Cohen, L. L., & Swim, J. K. (1995). The differential impact of gender ratios on women and men: Tokenism, self-confidence, and expectations. Personality and Social Psychology

Bulletin, 21(9), 876-884.

Craig, S. L., McInroy, L., McCready, L. T., & Alaggia, R. (2015). Media: A catalyst for resilience in lesbian, gay, bisexual, transgender, and queer youth. Journal of LGBT

Youth, 12(3), 254-275.

Davies, P. G., Spencer, S. J., Quinn, D. M., & Gerhardstein, R. (2002). Consuming images: How television commercials that elicit stereotype threat can restrain women academically and professionally. Personality and Social Psychology Bulletin, 28(12), 1615-1628.

Denzer-King, A. (2016). Effects of Stereotyped Representation in Television on Female and People of Color Viewers’ Self-Esteem.

Dumitrescu, A. L., Zetu, L., & Teslaru, S. (2012). Instability of self‐esteem, self‐confidence, self‐ liking, self‐control, self‐competence and perfectionism: associations with oral health status and oral health‐related behaviours. International journal of dental hygiene, 10(1), 22-29.

Erba, J. (2018). Media Representations of Latina/os and Latino Students' Stereotype Threat Behavior. Howard Journal of Communications, 29(1), 83-102.

Fogliati, V. J., & Bussey, K. (2013). Stereotype threat reduces motivation to improve: Effects of stereotype threat and feedback on women’s intentions to improve mathematical ability. Psychology of Women Quarterly, 37(3), 310-324.

Gerbner, G. (1998). Cultivation analysis: An overview. Mass Communication and Society, 1(3-4), 175-194.

Gillig, T., & Murphy, S. (2016). Fostering support for LGBTQ youth? The effects of a gay adolescent media portrayal on young viewers. International Journal of Communication, 10, 23.

Hastie, R., Schroeder, C., & Weber, R. (1990). Creating complex social conjunction categories from simple categories. Bulletin of the Psychonomic Society, 28(3), 242-247.

Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social psychology, 60(6), 895.

Hewner, M., & Knobelsdorf, M. (2008, November). Understanding computing stereotypes with self-categorization theory. In Proceedings of the 8th international Conference on

Computing Education Research (pp. 72-75). ACM.

Hoffner, C., & Buchanan, M. (2005). Young adults' wishful identification with television characters: The role of perceived similarity and character attributes. Media

psychology, 7(4), 325-351.

Hogg, M. A., & Terry, D. I. (2000). Social identity and self-categorization processes in organizational contexts. Academy of management review, 25(1), 121-140.

Hutter, R. R., & Crisp, R. J. (2005). The composition of category conjunctions. Personality and

Social Psychology Bulletin, 31(5), 647-657.

Jennings-Walstedt, J., Geis, F. L., & Brown, V. (1980). Influence of television commercials on women's self-confidence and independent judgment. Journal of Personality and Social

Psychology, 38(2), 203.

Kunda, Z. (1990). The case for motivated reasoning. Psychological bulletin, 108(3), 480.

Marx, David M., Sei Jin Ko, and Ray A. Friedman. "The “Obama effect”: How a salient role model reduces race-based performance differences." Journal of Experimental Social

Psychology 45, no.

Mastro, D. (2009). Racial/ethnic stereotyping and the media. Media processes and effects, 377-391.

Mastro, D. E., & Kopacz, M. A. (2006). Media representations of race, prototypicality, and policy reasoning: An application of self-categorization theory. Journal of Broadcasting &

Electronic Media, 50(2), 305-322.

McKinley, C. J., Mastro, D., & Warber, K. M. (2014). Social identity theory as a framework for understanding the effects of exposure to positive media images of self and other on intergroup outcomes. International Journal of Communication, 8, 20.

McInroy, L. B., & Craig, S. L. (2017). Perspectives of LGBTQ emerging adults on the depiction and impact of LGBTQ media representation. Journal of Youth Studies, 20(1), 32-46. Morin-Messabel, C., Ferriere, S., Martinez, F., Devif, J., & Reeb, L. (2017). Counter-stereotypes

and images: an exploratory research and some questions. Social Psychology of

Education, 20(1), 1-13.

McKinley, C. J., Mastro, D., & Warber, K. M. (2014). Social identity theory as a framework for understanding the effects of exposure to positive media images of self and other on intergroup outcomes. International Journal of Communication, 8, 20.

Prati, F., Crisp, R. J., & Rubini, M. (2015). Counter-stereotypes reduce emotional intergroup bias by eliciting surprise in the face of unexpected category combinations. Journal of

Experimental Social Psychology, 61, 31-43.

Pedulla, D. (2014). The positive consequences of negative stereotypes: Race, sexual orientation, and the job application process. Social Psychology Quarterly, 77(1), 75–94.

Ramasubramanian, S. (2011). The impact of stereotypical versus counterstereotypical media exemplars on racial attitudes, causal attributions, and support for affirmative action. Communication Research, 38(4), 497-516.

Rosenberg, M. (1965). Rosenberg self-esteem scale (RSE). Acceptance and commitment therapy.

Shrauger, J. S., & Schohn, M. (1995). Self-confidence in college students: Conceptualization, measurement, and behavioral implications. Assessment, 2(3), 255-278.

Simon, S., & Hoyt, C. L. (2013). Exploring the effect of media images on women’s leadership self-perceptions and aspirations. Group Processes & Intergroup Relations, 16(2), 232-245.

Tiggemann, M., & Zaccardo, M. (2015). “Exercise to be fit, not skinny”: The effect of fitspiration imagery on women's body image. Body image, 15, 61-67.

Tajfel, H. (1972). Social categorization. English manuscript of ‘‘La cate´gorisation sociale.’’ In S. Moscovici (Ed.), Introduction a` la psychologie sociale (Vol. 1, pp. 272–302). Paris: Larousse

Turner, J. C. (1975). Social comparison and social identity: Some prospects for intergroup behaviour. European journal of social psychology, 5(1), 1-34.

Verkuyten, M., & Brug, P. (2004). Multiculturalism and group status: The role of ethnic identification, group essentialism and protestant ethic. European Journal of Social

Psychology, 34(6), 647-661.

Wilcox, K., & Laird, J. D. (2000). The impact of media images of super-slender women on women's self-esteem: Identification, social comparison, and self-perception. Journal of

Research in Personality, 34(2), 278-286.

Appendix Appendix A. Scales

A.1. Self-Confidence

Below are a number of statements that reflect common feelings, attitudes, and behaviors. Please read each statement carefully and think about whether you agree or disagree that it applies to you. Try to respond honestly and accurately, but it is not necessary to spend much time deliberating about each item. Please think about how the items applies to you during the last 2 months unless some other time period is specified. Indicate your degree of agreement with each statement survey. From 1 (strongly disagree) to 7 (strongly agree)

1. I often feel unsure of myself even in situations I have successfully dealt with in the past.

2. I lack some important capabilities that may keep me from being successful. 3. Much of my time I don’t feel as competent as many of the people around me. 4. I have fewer doubts about my abilities than most people.

5. When things are going poorly, I am usually confident that I can successfully deal with them.

6. I have more confidence than most people I know.

7. If I were more confident about myself, my life would be better.

A.2. Appeal scale

The following question are about the attitude you have toward this advertisement. The answers are completely anonymous, so please answer as truthfully as you can. Please answer using the following scale.

When I see this message about Brookelle Jaffrary / Thomas Kennington I feel …

negative        positive

bad        good

unfavorable        favorable

Not nice        nice

I do not

like it        I like it

A.3. Emotional Appeal scale

To what extent do you agree with the following statements regarding the astronaut advertisement? Please answer as truthfully as you can. From 1 (strongly disagree) to 7 (strongly agree)

1. In my opinion this advertisement has a very strong appeal to my emotions. 2. This advertisement contained a lot of information.

3. This advertisement really did hold my attention. A.4. Identification scale

The following questions are about measuring what you think about the astronaut in the advertisement. The answers are completely anonymous, so please answer as truthfully as you can. From 1 (strongly disagree) to 7 (strongly agree)

1. I like the Brookelle Jaffary/Thomas Kennington

2. I do not have any feelings about the Brookelle Jaffary/Thomas Kennington 3. I can easily relate to Brookelle Jaffary/Thomas Kennington

4. I do not think Brookelle Jaffary/Thomas Kennington is not easily understood. 5. I think I could be a good friend with the Brookelle Jaffary/Thomas Kennington 6. I have no doubt the Brookelle Jaffary/Thomas Kennington and I would work well

7. The Brookelle Jaffary/Thomas Kennington is a personal role model. 8. I am personally inspired by the Brookelle Jaffary/Thomas Kennington. A.5. State self-esteem scale

This items are designed to measure what you are thinking at this moment. There is of course, no right answer for any statement. The best answer is what you feel is true of yourself at the moment. Be sure to answer all of the items, even if you are not certain of the best answer. Again, answer these questions as they are true for you RIGHT NOW.From 1 (strongly disagree) to 7 (strongly agree)

1. I am currently worried about whether I am regarded as a success or failure 2. I feel self-conscious

3. I feel displeased with myself

4. I am worried about what other people think of me 5. I feel inferior to others at this moment

6. I feel concerned about the impression I am making 7. I am worried about looking foolish

8. If I were more confident about myself, my life would be better 9. Right now, I am satisfied with myself

10. Right now, I think I am no good at all.

11. I feel that I have a number of good qualities, right now

12. Right now, I feel that I am able to do things as well as most other people. 13. At this moment, I feel I do not have much to be proud of

14. I certainly feel useless at this time

15. I feel that I'm a person of worth, at least on an equal plane with others right now 16. I wish at this moment I could have more respect for myself

17. All in all, at this moment, I am inclined to feel that I am a failure. 18. I take a positive attitude toward myself right now

Appendix B. Stimuli

B.1. Treatment Make Up Artist

B.3. Treatment Skier

B.4. Control Skier

Appendix C. Rotated Component matrix. C.1. Table 1

Factor loadings for exploratory factor analysis with varimax rotation of State Self-Esteem.

1

Component

2 3

I am currently worried about

whether I am regarded as a success or failure

I feel displeased with myself

I am worried about what other people think of me

.81 .79

I feel inferior to others at this moment I feel concerned about the impression I am making

I am worried about looking foolish If I were more confident about myself, my life would be better

Right now, I am satisfied with myself. Right now, I think I am no good at all. I feel that I have a number of good qualities, right now

Right now, I feel that I am able to do things as well as most other people. At this moment, I feel I do not have much to be proud of

I certainly feel useless at this time I feel that I'm a person of worth, at least on an equal plane with others right now I wish at this moment I could have more respect for myself

All in all, at this moment, I am inclined to feel that I am a failure.

I take a positive attitude toward

myself right now .89 .83 .84 .79 .86 .83 .87 .79 .93 .81 .81 .88 .85 .76 .54 C.2. Table 2

1

Component 2 I often feel unsure of myself even in

situations I have successfully dealt with in the past

I lack some important capabilities that may keep me from being successful

.92

.88

Much of my time I don’t feel as competent as many of the people around me

I have fewer doubts about my abilities than most people

When things are going poorly, I am usually confident that I can successfully deal with them

I have more confidence than people I know If I were more confident about myself, my life would be better

.83 .82 .81 .86 .91 C.3. Table 3

Factor loadings for exploratory factor analysis with varimax rotation of Self-Confidence.

Component

1 2

I like Brookelle Jaffary

I do not have any feelings about Brookelle Jaffary

I can easily relate to Brookelle Jaffary. I do not think Brookelle Jaffary is easily understood.

I think I could be a good friend with the Brookelle Jaffary .90 .70 .80 .81 .90

would work well together

I am personally inspired by Brookelle Jaffary

Brookelle Jaffrary is a personal role model I like Thomas Kennington

I do not have any feelings about Thomas Kennington

I can easily relate to Thomas Kennington I do not think Thomas Kennington is easily understood.

I think I could be a good friend with the Thomas Kennington

I have no doubt the Thomas Kennington and I would work well together

I am personally inspired by Thomas Kennington

Thomas Kennington is a personal role model .90 .87 .82 .86 .88 .81 .75 .85 .87 .85

Appendix D. Conceptual Model D.1. Figure One

Moderation Conceptual Model

Condition (control/treatment State Self-Esteem Self-Confidence 4.77*** -1.07***

p < .05*, p < .01**, p < .001***

D.2. Figure One

Moderation-Mediation Conceptual Model

p < .05*, p < .01**, p < .001*** Condition (control/treatment) Identification Self-Confidence State Self-Esteem .62* .59* -.35*** 1.35***