Abstract
Data regarding social support, optimism, affective tone, affective distress, and immune function were collected on fifty HIV-1 seronegative gay men. These data were utilized in a heuristic context to compare the relative merits of regression analysis versus causal modeling via path analysis. The heuristic model was that social support, optimism, and affective tone would predict affective distress, which would, in turn, predict immune function. Results were mixed, in that the psychosocial variables were interrelated as anticipated, but none of the psychosocial variables were related to immune function. Implications of these findings are discussed, and guidelines for the utilization of causal modeling are presented.