Electronic Journal of Human Sexuality, Volume 2, July 1, 1999


Attitudes Toward The Media's Portrayals of Gays and Lesbians

Peter B. Anderson, Akram Fakhfakh, Mary Alice Kondylis
University of New Orleans, Department of Human Performance and Health Promotion, New Orleans, Louisiana


       There are a variety of ways that the media affects and tracks the attitudes that the public has toward homosexuals.  Examples include dissemination of information, both accurate and inaccurate, marketing, television shows, and public opinion polls.

       Although we frequently hear references to "the gay community", Manning (1996) reminds us that multiculturalism is not about defining groups; it is about accepting individuals.  The writer argues that the gay community is merely a convenient, all-purpose mythic entity.  His contention is that this myth allows the gay media to present the idea of a shared purpose and identity and also provides the liberal media with an easy way to achieve political correctness.

       The media spent a great deal of time and effort in their coverage of the 1992-1993 controversy of gays in the military.  The findings of a recent study investigating their selection and use of unofficial sources indicate that decisions made in the selection of these unofficial sources were a result of the journalists' personal understanding of newsworthiness, their personal definition of expertise, and the particular news frames chosen to structure individual stories.  Additionally, it was discovered that almost no
attempt was made to obtain the views of more neutral parties; it was preferable to present a "raging debate" to the public.  Findings also suggest that former officers and public officials were used to explain the military's viewpoint and political outsiders to represent gay plaintiffs.  Through this tactic, balance appeared to have been achieved in a way that legitimized the views of one side while marginalizing the views of the other (Steele, 1997).

       In contrast, a place where gays have not only not been silenced, but actually encouraged to tell all is in the increasingly popular medium of daytime talk shows.  Talk show hosts encourage public voyeurism, on-camera fights and vivid descriptions of those whose identities differ from the norm.  All this increases their ratings and makes more money for the producers of the show, and they do succeed where others have failed by giving voice to those who are otherwise systematically silenced.  According to author Joshua Gamson, this is the only spot in mainstream media culture where it is possible for non heterosexuals to speak for themselves (Gamson, 1995).

       According to Wilkerson (1994), even the medical profession may be considered part of the media, since its more traditional attitudes are reflected in both policy and the dissemination of information.  There are numerous examples of the perpetuation of homophobic attitudes such as the treatment of people with HIV, "moralistic interpretations of people with AIDS," certain conceptions about ways in which HIV is transmitted, media representation of AIDS, and even the way in which medicine's "objectivity" reinforces a moral view inimical to gay men, lesbians and bisexuals.  The author concluded that the notion of medicine's "objectivity" and traditional conceptions of the ethics of health care hinder an appreciation of the ways in which medicine presumes and perpetuates homophobic values.

       In a 1995 study, college undergraduates read two newspaper articles about an "innocent victim" of AIDS.  When asked to explain the implicit message, very few could do so without prompting.  However, when asked directly, most respondents agreed that they understood the implicit message was that this "innocent victim" was different from other people with AIDS who are somehow blameworthy.  The few who did identify the subtext on their own were those with a more positive attitude towards homosexuals (Schellenberg, Keil & Bem, 1995).

       Riggle, Ellis & Crawford (1996) conducted a study of the impact of "media contact" on attitudes toward gay men.  College students were asked to view a documentary film of the life of a prominent gay politician.  Participants completed the Attitudes Toward Homosexuals Scale either prior to or after the film.  Results showed that viewing the film had a significant positive effect on attitudes.  However, the results of another 1996 study showed that increased media attention aimed at dissemination of information regarding the nature of HIV/AIDS may actually increase homophobic attitudes and anxiety concerning the disease and those affected (All & Fried, 1996).

       Yang (1997) examined changes in attitudes towards homosexuality in the U.S. through trends in public opinion polls.  The data were categorized by  issues related to homosexuality which included legal status, morality, acceptability, causes, familiarity with self identified homosexuals, as well as views on both military and nonmilitary occupations, civil rights, marriage and adoption rights, and AIDS.  The writer concluded that public attitudes have shifted in a liberal direction across a number of issues over time, but that levels of change vary and are dependant upon the particular issues in question.

       The present study examined volunteers perceptions of media's presentation of homosexuality, the impact of the media's presentations, and their personal attitudes and relationships with gays and lesbians.  It was hypothesized that gender, age, and level of religiosity would predict differences in each of the dependant variables.  Specifically, it was hypothesized that older, more religious men would demonstrate significantly more negative attitudes towards gays and lesbians than younger, less religious women.



       Of the 200 surveys completed, 2 had sufficient missing data and were eliminated from analysis, other missing data was accounted for on each item.  The respondents (125 women and 65 men - 8 respondents did not reply to this question) had an age range of less than  20 (2.0%) to over eighty (0.5%).  Age frequencies were 39.9% below 40, 39.4% between 40 and 49 and 20.7% over 50 years old [Median = 40 to 49, SD = 12.1, Range = <20 to >80].   The respondents were mostly white (79.3%) and black (14.6%) with 1.5% Asian and 3.0% other.

     Most, 52.5% defined themselves as religious, with 20.7% defining themselves as very or extremely religious.  On average they read a newspaper or magazine 5 days each week, watched TV 3 days each week, have cable (80%), and watch a movie 5 or more times each month.


       The instrument was 3 pages long and consisted of 10 demographic (e.g., age, race, sex) and media exposure (e.g., how many days a week do you read the newspaper, how many times a month do you watch a movie, do you have cable) questions and 23 items related to attitudes about portrayals of gays and lesbians in the media, the impact of those portrayals, and personal attitudes and relationships with gays and lesbians (see Appendix A).  The questionnaire was tested for face and content validity and pilot tested with a group of 30 volunteers at the airport prior to the implementation of data collection.

      As a result of the pilot test, the timing of visits to the waiting lounges and  minor adjustments in the wording of the questionnaire were made . All attitudinal questions were answered on a 5 point likert-type scale ranging from strongly disagree to strongly agree, with agree, disagree, and neutral responses available.  Behavioral questions (e.g., have you seen the movie "Philadelphia") were posed in a "yes" or "no" format.


       After approval by the Human Subjects Committee at the University of New Orleans, the survey was conducted at the New Orleans International Airport on both days of a weekend between 10 AM and 4 PM..  A male investigator in his twenties went to different waiting areas 30 to 40 minutes before a scheduled departure to ask volunteers to complete the survey while they were waiting for their flight.  Because some of the questionnaire items were related to perceptions of the effect that gays and lesbians seen on TV and in the movies would have on children, the investigator selected people who were traveling with children and asked them to volunteer to complete the questionnaire.


       Pearson correlations revealed that gender was not related to any of the outcome variables.  Younger respondents were less likely to support the statement that too much sex on TV causes perversion (p<.001).  Very religious respondents were more likely to support the statements that the media glorifies homosexuality (p<.001) and the media promotes homosexuality (p=.001).  They are also more likely to state that gays and lesbians are presented in too positive a way by the media (p=.004), too much sex on TV leads to promiscuity (p=.01), and watching gays on TV leads to bisexuality (p=.003).  In addition, they are less likely to welcome gays or lesbians in their homes (p=.01), hire a gay or lesbian employee (p=.001), allow their child to have a gay or lesbian teacher (p=.001), pediatrician (p=.002), physical education teacher (p=.002), or friend (p=.01).

       Multiple regression analysis revealed significant predictive value for level of religiosity on 10 of 13 dependant variables (see Table 1 and Appendix A) .  Only on the items, "it would be alright for my child to have a gay or lesbian friend" (where religiosity predicted 3.5% of the variance and entered in the second step of the regression analysis after gender), "too much sex on TV leads to promiscuity ( and the item, "too much sex on TV causes perversion" (no variables entered in to the regression equation) was level of religiosity not a significant predictor for the outcome variable.  In every case being very or extremely religious predicted more negative responses to attitudinal and behavioral items.  Gender was predictive for two items: "watching gays on TV leads to bisexuality" (explaining 3.8% of the variance, F (2,187) = 8.15, p<.001, endorsed more by men than women) and "it would be alright for my child to have a gay or lesbian friend" (explaining 4.1% of the variance, F (1,188) = 10.15, p<.001, endorsed more by women than men).  Age was not predictive for any item.

 Table 1

       1.   The Mass Media glorifies gays and lesbians.
       2.   The Mass Media are promoting homosexuality.
       4.   Are gays and lesbians presented too positively by the media?
       6.   Too much sex on TV leads to promiscuity.
       7.   Too much sex on TV causes perversion.
       9.   Watching gays on TV leads to bisexuality.
       15.   Gays are responsible for the AIDS epidemic.
       17.   Gays and lesbians are welcome in my home.
       18.   I would hire a gay or lesbian employee.
       19.   It would be alright for my child to be taught by a gay or lesbian teacher.
       20.   It would be alright for my child to have a gay or lesbian physical education teacher.
       21.   It would be alright for my child to have gay or lesbian friends.
       22.   I would take my child to a gay or lesbian pediatrician.


       The results were only partially supportive of the hypothesis.  Age was not predictive for any of the questionnaire items and gender was only predictive for two items and explained a very small percentage of the variance for each.  Self reported level of religiosity was predictive for most items and explained between 2.1% and 7.0% of the variance for each (typically between 5% and 7%).

       Ancillary analysis revealed that respondents who had seen the movie Philadelphia were significantly more likely to hold positive attitudes towards gays and lesbians and to behave in more accepting ways.  These results could be a result of selection bias or a residual impact of the movie.  Previous research has demonstrated both positive and negative relationships between media and attitudes toward homosexuals (All & Fried, 1996; Riggle, Ellis, & Crawford, 1996).  Future research designed to pinpoint the significant variables that account for these conflicting results will advance our understanding of the role of the media in shaping important social attitudes.

       The sampling technique used in this study was very unique and contributed some positive as well as potentially negative methodological opportunities.  Most studies are conducted with a population of college students with an average age of approximately 21 years old.  Few studies are conducted with populations of older adults, especially in such a distinctive setting.  The advantages of this sampling procedure include involving a unique sample, reducing our reliance on a college population, and expanding our future research efforts to include new samples that have not been previously explored.  The disadvantages include the fact that this was still a convenience sample and that we can not generalize our results beyond the respondents themselves.

       The results of this study support the notion that individual beliefs and values, specifically their level of religious beliefs, are the determining factor in their rating of the media's portrayal of homosexuality.  In this study more religions respondents were more likely to hold negative attitudes towards portrayals of gays and lesbians in the media, to believe that portrayals of homosexuality or sex in the media would lead to negative consequences, to be rejecting of gay or lesbian teachers and pediatricians, to keep known homosexuals out of their lives and homes, and to believe that homosexuals are responsible for the AIDS epidemic.  Results of previous research on the relationship between media portrayals of homosexuality and personally held attitudes are equivocal (All & Fried, 1996; Riggle, Ellis, & Crawford, 1996).  There is recent evidence that environmental factors (i.e.,. stress) correlate with homosexual frequency within cultures (Barber, 1998).  Could the extremely negative portrayals of homosexuality in the past have helped suppress some homosexual behaviors in addition to open communication about homosexuality?  Likewise, could more positive portrayals negate this suppression and support even greater  increases in openness about homosexuality and self reported homosexual behavior?  These and other important questions need further research.

Go to References

Return to Front Page