Findings from analyses presented in this paper, using data from a Roper survey, suggest that the role of attitudes toward homosexuals should be at the center of future explorations of the relationship between the media coverage of AIDS and public opinion. While the available data are limited, our analyses raise the possibility that anti-gay attitudes constrain the ability of the media to effectively communicate information about risk factors and how the disease is transmitted. Researchers need to explore the possibility that anti-gay attitudes stand between media information and public knowledge and public opinion.According to Singer, Rogers, and Corcoran (1987), the first U.S. poll on AIDS was conducted in June 1983. Since then, polls on AIDS and issues related to the epidemic have become increasingly frequent. Surveys have explored awareness, perceived importance of AIDS as a national issue, knowledge about transmission routes and affected groups, attitudes toward people with AIDS, opinions about policy issues such as spending, testing, and quarantine, and behavior in response to AIDS and its perceived risks, including precautions taken and changes in sexual practices.Researchers have discussed these poll findings primarily in the context of agenda-setting models, looking at the media's impact on the public's perception of the issue. This is true of Singer, Rogers, and Corcoran (1987) and, as their title suggests, of Dearing, Rogers, and Fei's paper "The agenda-setting process for the issue of AIDS" (1988). However, the latter authors do not find that "traditional agendasetting" models explain the processes they found in their analysis.