Assessing causality in the cognitive mediation model: A panel study of motivations, information processing, and learning during campaign 2000

Abstract

This two-wave national panel study was designed to test the causal claims of the "cognitive mediation model." The data indicate strong support for the following causal relationships predicted by the model: (a) surveillance motivations influence information processing, (b) information processing influences knowledge, and (c) motivations influence knowledge only indirectly through information processing. However, additional analyses demonstrated that these variables are not related in a simple unidirectional causal pattern. Instead, panel analyses found that most of these relationships are mutually causal. Future research should consider the reciprocal nature of relationships between information processing and knowledge, particularly as it relates to the study of the knowledge gap hypothesis.