A Meta-Analysis of the ELM's Argument Quality × Processing Type Predictions

One of the key dual-process model predictions is that audiences will be more persuaded by strong persuasive arguments than weak and that this difference in persuasiveness will be larger when they are processing centrally rather than peripherally. A series of meta-analyses were conducted (k = 134) to assess this claim and explore moderators. The data were generally consistent with the hypothesized interaction. The effects tended to be smaller when pre-post designs were used rather than post-test only. Assessments of the strength of the inductions did not tend to be associated with the size of the effects associated with those inductions.