Effect of call-in political talk radio shows on their audiences: Evidence from a multi-wave panel analysis

This article examines the effect of political talk radio (PTR) shows on listeners' political attitudes during the 1996 presidential election in the United States. Two complementary approaches for analyzing panel data (cross-lagged correlations and a fixed-effects conditional logit model) were employed to answer the question of whether the messages of PTR influence the audience or whether the audience is merely selecting sources consistent with its pre-existing political biases. In this context, the current study compares the usefulness of media exposure, on one hand, and political knowledge, on the other, as a tool for uncovering media effects on political attitudes. Three major conclusions result from the panel analyses. First is that the impact of PTR on political attitudes over time is very small at best. Second, when there is evidence of causal association between attitudes toward political figures and media reception, the direction of effect is from attitudes to reception (measured either by PTR exposure or political knowledge). Third, using political knowledge as a surrogate for media reception, instead of simple exposure, did not improve the number or magnitude of the observed associations between PTR messages and attitudes toward political leaders.