ࡱ> 7  bjbjUU "7|7|6_/l  8  '!!!!!!!!&&&&&&&$=( ]*H'9!!!!!'%!!A'%%%!T!!&%!&%%&&!t! @rN "&&,W'0'&*%F*&% SEQ CHAPTER \h \r 1Running head: REMEMBERING GAY/LESBIAN MEDIA CHARACTERSJournal of Homosexuality, 2007, 53(3), 19-34.Remembering Gay / Lesbian Media Characters: Can Ellen and Will Improve Attitudes toward Homosexuals?Jennifer M. Bonds-Raacke, University of North Carolina-PembrokeElizabeth T. Cady, Center for Engineering EducationRebecca Schlegel, University of Missouri-ColumbiaRichard Jackson Harris, Kansas State UniversityLindsey Firebaugh, Washburn University AbstractThree studies examined the priming effects of remembering portrayals of homosexual characters in the media. Study 1 had participants recall any memorable gay or lesbian media character and complete a survey about their perceptions of the character. Results were consistent with Greenbergs Drench Hypothesis with over two-thirds of participants recalling either Ellen or Will, and evaluative ratings for the characters recalled were generally positive. Study 2 was conducted to determine if participants viewed specific homosexual characters as being portrayed positively, negatively, or neutrally. With the exception of Ellen, in general, characters were viewed as positive portrayals but not overwhelmingly so. Finally, Study 3 was conducted to directly assess the effects of thinking about a positive or negative homosexual character on general attitudes towards gay men and lesbians. Results indicated that those recalling a positive portrayal later showed a more positive attitude toward gay men than those recalling a negative portrayal, and women had a more positive attitude than men toward gay men and lesbians. Such findings illustrate the importance of positive role models in the media as potential primes of social attitudes. Remembering Gay/Lesbian Media Characters: Can Ellen and Will Improve Attitudes toward Homosexuals?Although long nearly absent in the media, gay and lesbian characters have made their way into mainstream television and film in large numbers in the last decade or so. Until recently, however, such characters were hardly ever depicted. For example, the Production Code of 1934 formalized the voluntary exclusion of all gay and lesbian characters from Hollywood films (Russo, 1981), and such exclusion was adhered to when television emerged 15 years later. Not until the 1960s and 1970s did television shows occasionally deal with some gay and lesbian themes, although networks were still reluctant to introduce a regularly appearing homosexual character. Television and studio films have also struggled with depicting homosexual characters. Sometimes actors have been hesitant to accept gay roles, fearing that such roles would make it difficult to obtain straight roles later (Harris, 1999). In spite of attempts to exclude gay and lesbian characters and difficulties with casting homosexual characters, gay and lesbian characters have continued to emerge in television and film. Perhaps the most publicized homosexual character in the history of television was Ellen Morgan, the lead character of the program Ellen, played by Ellen DeGeneres. Although the episode featuring Ellen coming out set record ratings, the formerly highly rated sitcom was canceled soon after. Following the path set by Ellen, a new sitcom, Will and Grace, soon appeared on prime time featuring two gay characters played by Eric McCormick as Will Truman and Sean Hayes as Jack McFarland. Although we are now seeing gays and lesbians portrayed in the mass media with greater frequency than ever before, as yet we know little about the effect of these portrayals on a general public with largely negative attitudes towards homosexuals. The few experiments examining effects of media portrayals of homosexuals have shown some promising results. For example, Riggle, Ellis, and Crawford (1996) had participants view a documentary film depicting events surrounding the life and death of a prominent gay politician. Those participants viewing the film had a significant positive change in attitudes toward homosexuals. Similarly, Walters (1994) had two groups of participants complete measures of homophobia and empathy for homosexuals at the beginning and end of a school term. One group of participants was exposed to lectures on homophobia and homosexuality along with slides and video scenarios to demonstrate how gays and lesbians are stereotyped in the media. At the end of the term, this group of participants showed an increase in empathy for homosexuals and a decrease in homophobia, whereas the group denied such experiences showed no changes in attitudes. Such research investigating the influence of the media on attitudes toward homosexuals is important because gays and lesbians are the target of considerable prejudice manifested in a wide range of behaviors from verbal attacks to violent physical attacks. For example in 1984, the National Gay Task Force found due to their sexuality: (1) 90% of gay men and 75% of lesbians have been verbally harassed, (2) almost 50% of gay men and more than 33% of lesbians had been threatened with physical violence, and (3) 20% of gay men and 10% of lesbians had been physically assaulted (in Herek, 1988). In addition, the FBI reported that 13% of all hate crimes in 1995 were motivated by sexual orientation bias (FBI, 1995). Even on college campuses, lesbian and gay male undergraduates report being verbally insulted, chased or followed, and physically assaulted (DAugelli, 1992). However, future research is needed to determine what influence very popular homosexual characters in entertainment television have on attitudes towards homosexuality. Such portrayals are important in terms of Greenbergs (1988) Drench Hypothesis, which says that a few highly salient instances drench the viewer with influence, rather than a slow and even dripping of impact from many infrequently viewed characters. Additionally, attitude-change research is needed to investigate what can be done to decrease negative attitudes towards homosexuality (Herek, 1988). PurposeThe present research consisted of three studies. The purpose of Study 1 was to determine which gay and lesbian characters were most often thought of, indirectly testing the Drench Hypothesis, which would prevail if the majority of participants recalled a few highly salient and popular portrayals. In addition, Study 1 assessed how participants viewed these characters on a variety of personality dimensions (e.g., humorous, likeable). The purpose of Study 2 was to determine if participants viewed specific homosexual characters mentioned in Study 1 as being portrayed positively, negatively, or neutrally. Finally, Study 3 was conducted to directly assess the priming effects of thinking about a positive or negative homosexual character on general attitudes towards gay men and lesbians. All three studies for this research had participants remember media-watching from their own lives. Previous research has used this sort of methodology to investigate memories for frightening movies from childhood (Harrison & Cantor, 1999; Hoekstra et al., 1999), movies with sexual content from childhood (Cantor, Mares, & Hyde, 2003), memories for watching violent or romantic movies on dates (Harris, Hoekstra, Scott, Sanborn, Karafa, & Brandenburg, 2000; Harris et al., in press), and mediated sporting events (Bonds-Raacke & Harris, under review). This methodology was useful in the current studies for several reasons. Remembering a certain character from a regularly-watched TV show is likely to evoke many pleasant or unpleasant memories of watching that show. When asked to think of a character that is memorable to themselves, participants are likely to come up with those who have been most influential in their own thinking. Therefore, if relatively few characters are chosen by many participants, Greenberg's (1988) Drench Hypothesis would be supported, suggesting that a small number of exemplars have a potentially disproportionate amount of influence, due to the large audiences they draw. Another advantage of this methodology is its ecological validity. Unlike laboratory studies where the experimenter shows the participants a pre-selected film clip, here the participant chooses a character from a program that they have watched it in its entirety and on their own time and with their realistic attention level. Study 1MethodParticipants were 269 college students enrolled in a General Psychology course who were given partial course credit for their participation during the fall semester of 2001. Participants were first asked to think of a memorable gay/lesbian character in a TV show/movie and complete a survey for their perceptions of this character using 7-point Likert-scales for the dimensions of: serious/humorous, likeable/not likeable, mentally ill/mentally stable, safe/dangerous, moral/unmoral, honest/dishonest, responsible/irresponsible, kind/cruel, violent/nonviolent, and bad role model/good role model. Participants also responded to attitudinal statements assessing the extent to which they believed: (1) real-life gays/lesbians were like the character, (2) they would enjoy having this character as a friend, (3) the characters family accepted their homosexuality well, and (4) the characters friends accepted their homosexuality well. ResultsOver two-thirds of participants listed Ellen (47%) from Ellen and The Ellen Show or Will (20%) from Will & Grace as the most memorable gay or lesbian character. See Table 1 for a complete listing of recalled characters. Participants also indicated these characters were relatively humorous, likeable, mentally stable, safe, honest, kind, responsible, and nonviolent. These characters were judged in the middle of the scale for moral/immoral and good role model/bad role model (See Table 2). Participants agreed that that the characters friends accepted their homosexuality well and were ambivalent about whether (1) real gay and lesbian people were like the character, (2) they would enjoying having this character as a friend or acquaintance, and (3) if the characters parents accepted their homosexuality well (See Table 3). DiscussionResults were consistent with Greenbergs Drench Hypothesis, in that the majority of participants recalled a few highly salient and popular portrayals. In addition, participants were generally positive in their ratings of the characters. However, future research is needed to determine if recalling such characters served as a prime and influenced attitudes positively. Also, investigating the influence of negative portrayals on participants ratings will aid in explaining how media portrayals affect individuals attitudes towards gays and lesbians.Study 2MethodIn Study 2, 86 different participants from Study 1, but from the same population, completed a questionnaire during fall semester 2001. Participants were given a list of 130 characters and asked to indicate whether they viewed the character in a positive, negative, or neutral way by circling their choice. Participants were instructed to respond only to the character and not to the actor portraying him or her. If the person was a sports figure, singer, news anchor, etc., they were to respond to the person. If the character was unfamiliar to the participants, they were asked to circle the Dont Know alternative. The characters listed included the 9 most frequently recalled homosexual characters from Study 1, plus 2 homosexual characters recalled by one participant each from Study 1, and an additional 119 other characters who were not homosexuals. These additional characters were included as norming for another study and to disguise that information was being gathered on homosexuals in the media. ResultsIn general, the nine gay or lesbian characters were viewed by more participants as positive than as negative, with the exception of Ellen, the most recalled character from Study 1. In this study, 24.9% of participants felt she was a positive portrayal, 39.5% felt she was negative, and 23.3% felt she was neutral. More participants indicated that Will, the second most recalled character from Study 1, was portrayed more positively than Ellen. Will was seen as positive by 51.2% of participants, negative by 8.1%, and neutral by 16.3%. A complete listing of characters with percentage for type of portrayal can be found in Table 1. DiscussionAlthough characters were viewed as more positive than negative, the characters were not seen by the majority of participants as being overwhelmingly positive. Thus, it is still difficult to determine if recalling the gay or lesbian character in Study 1 served as a prime, which influenced participants responses to subsequent scales on attitudes towards gays and lesbians. What is clear from Study 2 is that different participants view the same characters in very different ways. Thus, Ellen Morgan may be a positive portrayal to some viewers but certainly not all. Therefore, to determine if different types of portrayals have an effect on individuals attitudes towards homosexuals, a third study was conducted. Study 3MethodTwo groups of participants were used for Study 3, conducted in fall of 2002. Group one (N=65) recalled a positive portrayal of a homosexual media character, while group two (N=49) recalled a negative portrayal. Specifically, group one was instructed to think about a very memorable character in a TV show or movie that they remember who was portrayed very positively and who was clearly identified as gay or lesbian. Participants were instructed that by positive we mean someone who is presented in such a way that many people would tend to admire and or like this person. Finally, participants were told that they did not have to like everything about the character, but the overall impression was to be positive. Group two was also asked to think about a very memorable character who was clearly identified as gay or lesbian, but one who was portrayed very negatively. Participants were instructed that by negative we mean someone who is presented is such a way that many people would dislike and or have little respect for this person. Again, participants were instructed that they did not have to dislike everything about the character, but their overall impression must be negative. Next, participants completed a two-part survey. Part one was identical to the survey used in Study 1 in which participants rated their perceptions of the character on various dimensions using Likert-scales and responded to four attitudinal statements. In part two of the survey, participants completed one of the versions of the Attitudes Toward Lesbian and Gay Men (ATLG) Scale (Herek, 1988). The particular version used consisted of 40 statements, 20 statements about gay men (ATG subscale) and 20 about lesbians (ATL subscale), in which participants indicated their level of agreement or disagreement to the statements on a Likert-scale ranging from 1 (strongly agree) to 7 (strongly disagree). The subscales contained identical statements (only varying in the reference to gay men or lesbians) to allow for a direct comparison of scores. Scores for each subscale could range from 20-140, with higher scores indicating a more negative attitude. Sample statements included: female/male homosexuality is an inferior form of sexuality and female/male homosexuality is a perversion. Results See Table 4 for a list of characters recalled for the positive and negative conditions. A two-way MANOVA was conducted to determine the effects of gender and character evaluation (i.e., positive portrayals and negative portrayals) on the two dependent variables of attitudes towards gay men (ATG scale) and attitudes towards lesbians (ATL scale). MANOVA results indicated that character evaluation [Pillais Trace = .086, F (2, 109) = 5.12, p < .01] significantly affected attitudes toward gay men, and gender [Pillais Trace = .219, F (2, 109) = 15.26, p < .001] significantly affected attitudes towards both gays and lesbians (See Table 5). Univariate ANOVAs were conducted as follow-up analyses. ANOVA results indicated that attitudes toward gay men [F (1, 110) = 5.43, p<.05, h2 = .047, power = .637] significantly differed by character evaluation. Specifically, those recalling a positive portrayal later showed a more positive attitude toward gay men then those recalling a negative portrayal. ANOVA results also indicated that attitudes towards gay men [F (1, 110) = 19.50, p<.001, h2 = .151, power = .992] and attitudes toward lesbians [F (1, 110) = 4.88, p < .05, h2 = .042, power = .591] significantly differed by gender. Specifically, women s attitudes towards both lesbians and gay men were more positive (See Table 5). Independent t-tests were also conducted to examine differences between perceptions of characters (e.g., serious/humorous, likeable/not likeable) based on whether the media portrayal was positive or negative. As expected, the positively portrayed characters were found to be more humorous, likeable, mentally stable, safe, moral, honest, responsible, kind, and nonviolent then the negatively portrayed characters (See Table 2). Finally, participants attitudes towards whether: (1) real gay and lesbians are like the character, (2) they would enjoy having the character as a friend, and (3) the characters friends and family accepted their homosexuality were measured (See Table 3). Results indicated that those recalling a positive portrayal agreed more that they would enjoy having this character as a friend [t (154) = -4.53, p < .001] and that the characters friend accepted their homosexuality well [t (151) = -4.36, p < .001] compared to those recalling a negative portrayal. A follow-up study was also conducted to assess participants attitudes toward homosexuals under a control condition. Specifically, 86 participants were asked to think of a memorable media character who was single (i.e., not married) with no mention of sexual orientation. After recalling the character, participants rated their perception of the character on various dimensions before completing the ATLG scale. Participants were asked to recall any unmarried character to ensure that they engaged in the same type of memory task as those participants who were asked to recall a positive or negative portrayal of a homosexual character. Scores for the control condition were similar to those found for participants recalling a negative portrayal of a homosexual character. Specifically, the mean score for the ATL was 75.04 and for the ATG the mean score was 82.57. Thus, it seems that thinking of a positive portrayal served as a prime and influenced individuals attitudes towards homosexuality in a positive way, while thinking of a negative portrayal did not substantially change peoples attitudes towards homosexuality (See Table 5). DiscussionThe results obtained from Study 3 demonstrate a basic priming effect from thinking about a media character, thus suggesting the huge potential influence of the entertainment industry on the general publics attitudes toward homosexuality. Specifically, it seems that recalling a positive portrayal of a homosexual character from the media can contribute to a positive change in attitudes towards real gay men. Such findings remind us of the importance of positive role models in the media for affecting social change. General DiscussionOverviewThe present studies have shown a priming effect on social attitudes from merely thinking for a few minutes about a character seen on television. This is striking for several reasons. One, the effect comes from remembering and thinking about the character; participants do not even have to view anything in the context of the experiment. Just thinking about a character for a few minutes is enough to affect attitudes measurably.Second, the effect was more positive than negative. Thinking about a positive character led to a more positive attitude toward gay men, but thinking about a negative character did not significantly change the attitude from that of a control group thinking about an unrelated character. The fact that the effect worked only in one direction also argues against a task demand explanation that people feel they should respond to the attitude scale in a way consistent with the character they thought about, since such a task demand should have worked in both a positive and negative direction. Finally, in line with Greenberg's Drench Hypothesis, these results suggest the potentially enormous influence of a few positive role model media characters on attitudes toward that social group. It may be that the show Will and Grace, with its huge audiences, is doing more to improve attitudes toward gay men than any amount of explicit social teaching in schools, families, churches, and elsewhere. These results are not really surprising considering the huge success of entertainment education efforts in many developing countries. Cooperative efforts of governments, broadcasters, and physicians have produced dramas with the specific purpose of advancing the social status of women, increasing condom use and other HIV-protective behaviors, or lowering the rate of domestic violence. Many of these shows, such as Hum Log (We People) in India or Soul City in South Africa, have become very popular shows and have been followed by substantial attitudinal and lifestyle changes (Singhal & Rogers, 1999, 2002). Even in the U.S., there has been cooperation between the Center for Disease Control and producers of shows like ER in writing scripts to encourage healthier behaviors (Stolberg, 2001).Character RecalledAmong the characters recalled by participants in Studies 1 and 3, it is interesting that a small number of characters were recalled with high frequency. It is likely that these characters were consistently making an impact on viewers. It is important to note that while Will and Ellen were the most frequent exemplars in both studies, Will surpassed Ellen in Study 3 as the most frequently reported character. This could be due to the fact that there was a one year time difference between the two studies and Ellens show had been off the air for several seasons, while Will and Grace was still on the air and had recently entered syndication. Thus, it is not surprising that Will would have an increasingly large impact, while Ellens impact would have decreased. Additionally, the fact that so many participants recalled Ellen or Will is consistent with Greenbergs (1988) Drench Hypothesis. It is quite apparent from the findings that a select few leading characters are highly salient in peoples minds (e.g., drenching the viewer), while other supporting characters even if from popular shows, like Carol from Friends, have less effect on the viewing population. This is especially evidenced by the fact that Ellen (the most highly publicized homosexual character in history) was recalled with such great frequency even after her show has been absent from the airwaves for many years. Such findings demonstrate the potential influence that one character from one show can have on the viewing population. It is also interesting that some of the same characters were often recalled in both the positive and negative conditions in Study 3. For example, Jack and Ellen were commonly listed as both a negative portrayal and a positive portrayal, suggesting that different people evaluate the same characters very differently. For example, one person might view a stereotypical effeminate portrayal of a gay man as negative due to it perpetuating a stereotype, whereas another person might view the portrayal as positive due to the characters likeable and humorous personality. Effects of PrimingIn examining the results from Studies 1 and 3, the results can be compared since the surveys used identical questions. Specifically, in both studies, participants rated the character on several dimensions and responded to four attitudinal statements. When comparing responses of the ratings on the character dimensions (See Table 2), an interesting pattern emerges. Those participants asked to recall any memorable homosexual media character (Study 1) and those asked to recall a positive portrayal of a homosexual media character (Study 3) had similar ratings on all the dimensions but one (i.e., bad role model/good role model). In general, the ratings for the dimensions were relatively positive with both groups of participants finding the character to be: humorous, likeable, mentally stable, safe, honest, kind, responsible, and nonviolent. Yet, those participants recalling a negative portrayal of a homosexual media character in Study 3 were more ambivalent on their ratings for these dimensions. Therefore, when evaluating a homosexual media character, thinking of a negative portrayal influences how individuals perceive the character. However, for the dimension of good role model/bad role model, the ratings were similar between participants recalling any memorable media character and participants recalling a negative portrayal of a homosexual character. Again, participants were ambivalent about whether the character was a good or bad role model. So, it appears that thinking of a character can influence how participants view the character on specific traits, but may have less influence on a judgment about whether the character is a good role model or not. Although these ratings were ambivalent and not really strongly negative, the less positive ratings for the role model item suggests some latent prejudice that comes out in the more general role model rating than in more specific trait evaluations. It is also worth noting the comparison of the responses on the attitudinal questions between Studies 1 and 3 (See Table 3). For example, more participants recalling a negative portrayal doubted that they would enjoy having this character as a friend or acquaintance and doubted that the characters friends accepted their homosexuality well, compared to those recalling any memorable character and those recalling a positive portrayal. ATLG ScoresScores on the ATL and ATG subscales for the current study replicate findings of Herek (1988) with men expressing a more negative attitude toward homosexuals (especially gay men) than women do. It could be that in this society heterosexuality is so strongly linked with masculinity that by rejecting homosexual men who are breaking gender norms, heterosexual men are affirming their masculinity. As Herek (1988) recommended, attitude change research is needed despite whatever the reasons may be for negative attitudes towards homosexuals. The current studies contribute to this line of research by demonstrating how media can be used in a prosocial manner to change attitudes towards homosexuality, although future research is still needed. Limitations and Future ResearchAlthough results from the present studies do much to advance the knowledge of the potential influence the entertainment industry has on viewers social attitudes, there are limitations to having participants remember media characters. The major limitation is that memory is necessarily retrospective and unverifiable. Clearly there is decay of memory over time, as well as distortions in memory arising either from encoding failure or retrieval error. While these shortcomings must be acknowledged, there is good reason to think that this limitation is not a fatal flaw. In the present research, it is the effects of ones memory on attitudes that are of primary interest, not the accuracy of the way the character is remembered. Thus how accurately participants remember a particular character is not critical in this case. This research is primarily interested in memory for the experience of viewing, with the memories themselves being of interest, whether or not they are fully accurate. Furthermore, the crucial experimental comparisons are by gender or priming condition, which should not differ in overall accuracy of memory. Future research is still needed to examine participants familiarity with the character. It could be that the more exposure the participant has with the character the more likely the character is to influence the participants subsequent attitudes toward homosexuals. Exposing participants directly to film clips of media portrayals of homosexual characters should also be considered. Finally, researchers should consider investigating participants reactions to extremely negative portrayals of homosexuals in the media, to non-stereotypical portrayals of homosexuals, to the actors and actresses depicting homosexual characters, and to homosexual media portrayals over time. ReferencesBonds-Raacke, J. M. & Harris, R. J. (under review). Autobiographical memories of mediated sporting events. Cantor, J., Mares, M.-L., & Hyde, J. S. (2003). Autobiographical memories of exposure to sexual media content. Media Psychology, 5, 1-32. DAugelli, A. R. (1992). Lesbian and gay male undergraduates experiences of harassment and fear on campus. Journal of International Violence, 7, 383-395. Greenberg, B. S. (1988). Some uncommon television images and the Drench Hypothesis. In S. Oskamp (Ed.), Television as a social issue (pp. 88 102). Newbury Park, CA: Sage. Harris, R. J. (1999). A cognitive psychology of mass communication (3rd ed.). Mahwah, NJ: Lawrence Erlbaum Associates. Harris, R. J., Hoekstra, S. J., Scott, C. L., Sanborn, F. W., Karafa, J. A., & Brandenburg, J. D. (2000). Young mens and womens different autobiographical memories of the experience of seeing frightening movies on a date. Media Psychology, 2, 245-268.Harris, R. J., Hoekstra, S. T., Sanborn, F. W., Scott, C. L., Dodds, L. A., & Brandenburg, J. D. (in press). Autobiographical memories for romantic movies seen on a date. Harrison, K., & Cantor, J. (1999). Tales from the screen: Enduring fright reactions to scary media. Media Psychology, 1, 97-116.Herek, G. M. (1988). Heterosexuals attitudes toward lesbians and gay men: Correlates and gender differences. The Journal of Sex Research, 25, 451 477. Hoekstra, S. J., Harris, R. J., & Helmich, A. L. (1999). Autobiographical memories about the experience of seeing frightening movies in childhood. Media Psychology, 1, 117-140.Riggle, E. D. B., Ellis, A. L., & Crawford, A. M. (1996). The impact of media contact on attitudes towards gay men. Journal of Homosexuality, 31 (3), 55 69. Russo, V. (1981). The celluloid closet: Homosexuality in the movies. New York: Harper & Row.Singhal, A., & Rogers, E.M. (1999). Entertainment-Education: A communication strategy for social change. Mahwah NJ: Lawrence Erlbaum Associates.Singhal, A., & Rogers, E.M. (2002). A theoretical agenda for entertainment-education. Communication Theory, 12, 117-135.Stolberg, G. S. (2001). CDC injects dramas with health messages. Retrieved June 26, 2001, from www.nytimes.com/2001/06/26/health/26CDC.htmlUnited States Department of Justice, Federal Bureau of Investigation (1995). Uniform crime reports: Hate crime. Retrieved June 5, 2003, from http://www.fbi.gov/ucr/hatecm.htmlWalter, A. S. (1994). Using visual media to reduce homophobia: A classroom demonstration. Journal of Sex Education and Therapy, 20 (2), 92 100. Table 1Percentage of Participants Recalling Gay or Lesbian Media Characters Study 1: Study 2:Character Show/Movie Overalla Positiveb Negativec Neutrald Dont Knowe______________________________________________________________________________Ellen Ellen 47.0 24.9 39.5 23.3 12.3Will Will and Grace 20.0 51.2 8.1 16.3 24.4Jack Will and Grace 8.0 40.7 7.0 15.1 37.2Andrew Philadelphia 4.0 43.0 4.7 11.6 40.7Carol Friends 3.0 26.7 11.6 29.1 32.6George My Best Friends Wedding 3.0 50.0 7.0 11.6 31.4Simon As Good As It Gets 2.0 32.6 9.3 15.1 43.0Amy Chasing Amy 2.0 25.6 3.5 7.0 63.9Carter Spin City 1.0 66.3 2.3 4.7 26.7a Character recalled b Viewed in positive wayc Viewed in a negative wayd Viewed in a neutral waye Did not know character Table 2Means for Likert Scale Ratings for Portrayals of Gay/Lesbian Characters Study 1: Study 3:Perceptions Meana Positiveb Negativec TC \l1 "Serious (1)/Humorous (7) 5.46 5.53 4.43***Likeable (1)/Not Likeable (7) 2.82 2.47 4.04***Mentally ill (1)/Stable (7) 5.77 5.77 4.72***Safe (1)/Dangerous (7) 2.25 2.16 3.45***Moral (1)/Immoral (7) 3.48 2.94 4.39***Honest (1)/Dishonest (7) 2.46 2.38 4.00***Responsible (1)/Irresponsible (7) 2.55 2.63 3.57***Kind (1)/Cruel (7) 2.04 2.07 3.43***Nonviolent (1)/Violent (7) 1.79 1.84 3.00***Bad Role Model (1)/Good Role Model (7) 4.40 5.03 4.41a From Study 1b Those recalling a positive portrayal c Those recalling a negative portrayal *** p<.001 for differences in ratings between recalling a positive and negative portrayal Table 3Comparison of Mean Likert Ratings on Attitudinal Statements from Studies 1 and 3 (1 = Strongly Agree, 5 = Strongly Disagree) Mean RatingReal homosexuals are like character Study 1 2.79 Study 3: Positive Portrayal Recall 3.00 Study 3: Negative Portrayal Recall 3.13Would enjoy having character as friend Study 1 2.73 Study 3: Positive Portrayal Recall 2.38* Study 3: Negative Portrayal Recall 3.48*Characters friends accepted homosexuality Study 1 1.87 Study 3: Positive Portrayal Recall 1.78* Study 3: Negative Portrayal Recall 2.76*Characters parents accepted homosexuality Study 1 2.93 Study 3: Positive Portrayal Recall 2.85 Study 3: Negative Portrayal Recall 2.94*Significant difference between positive and negative conditions at p < .001 Table 4Characters Recalled: Study 3Condition Characters Recalled PercentagePositive Will Truman 28Ellen DeGeneres 18Jack McFarland 9 Others (each < 2%) 45Negative Ellen DeGeneres 9Jack McFarland 7Rosie ODonnell 7Others (each <2%) 77 T able 5Mean Scores (Range 20-140) for ATL and ATG Subscales for Study 3Higher Scores: more negative attitude TC \l1 "Character Evaluation Gender Follow-Up TC \l1 "ATL Positive 70.06 ATL Men 78.80 ATL 75.04 TC \l1 "Negative 75.31 Women 66.60ATG Positive 74.32 ATG Men 92.68 ATG 82.57Negative 87.01 Women 68.84  Although participants in this and subsequent studies were not asked their own sexual orientation, past research at this university generally shows a minuscule percentage indicating themselves to be homosexual, so the best assumption must be that almost all participants were heterosexual or unrecognized (perhaps even to themselves) homosexuals. Media CharactersMedia CharactersOg$$$$((((((( )++.".2$24477@@DDDnFpFrFGGGPPRRWWYYYZZZ![#[j[|[]]__ccReUemmop$x8xxxJyfyyyyyz{|3|||n}H*OJQJj0J2U66] jU[MNO}~1e$da$ . !$1$  ? gq $$$((++..".2$2a3$a$d`$da$da344777C<@@@C-H LPPR,p@ P !$d ,p@ P !$d`/P @ P !$d$da$dd`RRReTVj[}[^maccPkmmopt9wEww ,p@ P !$d`,p@ P !$d$,p@ P !$da$wwAxxx?12@AhjxAފO{9_hi|}CJ>*H* jU>*H*6[ހ;t}Á1CggSg  !$ ^ `,p@ P !$$,p@ P !$d&d P /p@ P !$d,p@ P !$d ,p@ P !$d` ӂb=΄+D_y!,p@ P !$&d P $,p@ P !$d&d P ,p@ P !$d,p@ P !$y?o׆kOOO,p@ P !$d$,p@ P !$d+$,p@ P !$1$$,p@ P !$d&d P ,p@ P !$dh,p@ P !$ 3cƇ1@hEqA,p@ P !$$,p@ P !$d&d P ,p@ P !$dAVފ"P{8A^}$,p@ P !$dh&d P ,p@ P !$dh$,p@ P !$d&d P ,p@ P !$d^_njߌ$9<D$,p@ P !$0dh^`0,p@ P !$d$,p@ P !$dh^`,p@ P !$dh'DryP(,p@ P !$dh&d P `!$,p@ P !dh`+$,p@ P !dh1$ +$$,p@ P !dh1$$,p@ P !$0dh^`0ˍ΍%&' < jUj0J2U>* jU>*ww !$$!$da$$!$a$$!$d$&#$+Da$.$1$a$1,p@ P !$1$$,p@ P !$dh&d P  $,p@ P !$dh&d P  !$!$d / =!"#$%3 i4@4 NormalCJ_HmH sH tH <A@< Default Paragraph FontpOp _level1O/hhp@ P !$h1$]^h`lOl _level2L,p@ P !$1$]^`lO"l _level3L,88p@ P !$81$]^8`jO2j _level4I)p@ P !$1$]^`jOBj _level5I)p@ P !$1$]^`fORf _level6F& pp@ P !$p1$]^p`fObf _level7F& @ P !$ 1$]^ `dOrd _level8C# @ @ P !$@ 1$]^@ `dOd _level9C# P !$ 1$]^ `pOp _levsl1O/hhp@ P !$h1$]^h`lOl _levsl2L,p@ P !$1$]^`lOl _levsl3L,88p@ P !$81$]^8`jOj _levsl4I)p@ P !$1$]^`jOj _levsl5I)p@ P !$1$]^`fOf _levsl6F& pp@ P !$p1$]^p`fOf _levsl7F& @ P !$ 1$]^ `dOd _levsl8C # @ @ P !$@ 1$]^@ `dOd _levsl9C!# P !$ 1$]^ `pO"p _levnl1O"/hhp@ P !$h1$]^h`lO2l _levnl2L#,p@ P !$1$]^`lOBl _levnl3L$,88p@ P !$81$]^8`jORj _levnl4I%)p@ P !$1$]^`jObj _levnl5I&)p@ P !$1$]^`fOrf _levnl6F'& pp@ P !$p1$]^p`fOf _levnl7F(& @ P !$ 1$]^ `dOd _levnl8C)# @ @ P !$@ 1$]^@ `dOd _levnl9C*# P !$ 1$]^ `@O@ WP9_Heading 1 +1$dCJ>*HOH WP9_Heading 2 ,1$<OJQJCJ56(O( Default Para<O< WP9_Header.1$!$.O. WP9_Page Number<O< WP9_Footer01$!$4O4 Footnote Tex11$CJ,O!, Footnote RefH* ʼn^aʼnMNO}~1e? gq $$''**"*.$.a/00333C8<<<?AE*J5J>LQLZL N]PU"U$X[P]c]dffii6nppFqYqqDrrr7sssttu_uuu,>>>?%?6T=TTUqqvv2w8w6x=xxxByJy{{{{{{{{{{LU6ÉƉO|1&+o0..0*1667"7??@@uFxFVVWWpqFqNqmqoqqqqqDrHrUrrrrrrNssttttuu u]uuuuu?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_abcdefghijklmnopqrstuwxyz{|}Root Entry FrN1Table`*WordDocument"SummaryInformation(vDocumentSummaryInformation8~CompObjjObjectPoolrNrN  FMicrosoft Word Document MSWordDocWord.Document.89q