Based on the assumption that the variables used in diffusion of information studies in developing nations are subject to Western bias, theoretically complex, difficult to compare across cultures, and difficult to operationalize, a study was undertaken to test the predictive ability of simpler demographic variables. The data were drawn from a nationwide sample of 4,863 Ghanians and were collected by information agency interviewers of that country using a quota scheme to approximate census figures. Analysis of results showed that, using a modification of the D. Lerner and E. Rogers and L. Svenning model, variables from the biosphere (age and sex), the culture (education and occupation), the environment (road class and town size), and the use of mass media explained as much or more of the variance in political knowledge of the respondents as did studies using psychological variables such as empathy, cosmopoliteness, socioeconomic status, or achievement motivation. The findings suggest the use of a causal process model and the simpler demographic variables for market analysis and testing campaign effectiveness when a full-scale sociometric analysis is not feasible, especially in developing nations where little is yet known about the dynamics of the interview situation or about cross-cultural comparability of data. (Author/FL)