Abstract
This study explores the role of interpersonal discussion and social norms in a public health campaign, the BBC Condom Normalization Campaign, designed to promote conversation and change the public perception of condom use in India. Drawing upon the integrative model of behavioral prediction, attitudes, self-efficacy, subjective norms, and descriptive norms were predicted to relate to behavioral intentions to use condoms. It is important to note that the valence of discussion was hypothesized to relate to each of these more proximal predictors. The authors used structural equation modeling to test the model on 3 separate samples of Indian men between the ages of 15 and 49 years: (a) high-risk men who had sex with nonspouses; (b) low-risk, sexually inactive, unmarried men; and (c) low-risk, monogamous, married men. Results were similar for low- and high-risk audiences, with valence of discussion about condoms predicting condom-related attitudes, self-efficacy, and subjective and descriptive social norms with respect to condom use, which, in turn, predicted behavioral intent to use condoms. These findings underscore the need to take not only the frequency but also the valence of interpersonal discussion into account when assessing the effect of health campaigns. Implications for theory and design of future public communication campaigns are explored.
Acknowledgments
This research was made possible by a grant from the Bill and Melinda Gates Foundation. Joyee S. Chatterjee acknowledges the support of the University of Southern California Annenberg Graduate Fellowship grant that facilitated her role in the preparation of this article. The authors thank Wendy Meltzer and the editorial staff at the Journal of Health Communication.
Notes
1The campaign was part of the Bill and Melinda Gates Foundation Avahan initiative. In conjunction with the campaign, local organizations conducted interpersonal interventions with high-risk groups to promote condom use and also distributed condoms. Only the mass media campaign is considered here.
2We ran a variation in which we added categories of discussion partners as a separate construct. The addition of this variable did not change any of the beta coefficients, nor was the construct significantly related to campaign exposure or to attitudes. However, it did positively relate to self-efficacy, subjective norms, and descriptive norms.