In search of reliable persuasion effects: I. A computer-controlled procedure for studying persuasion

Suggests that the research literature on persuasion is in a state of disarray, in part due to widespread use of nonstandardized procedures for collecting data and of statistically weak (low power) designs. The present study used (a) a computerized procedure as an aid to standardization and (b) within-Ss designs to obtain high power. This procedure has some resemblance to mass-media settings in which audiences are exposed to a large number of persuasive appeals in a short period. Data from 8 experiments with a total of 616 Ss (undergraduates and university newspaper respondents) demonstrate the possibility of studying effects of traditionally interesting independent variables with this procedure. Most importantly, reliable evidence of a rapid decay of persuasion within a single experimental session was found for a subset of the communication topics that were used. This indicated potential applications of the computer-controlled procedure to the study of variables that might influence the rate of decay of persuasion. However, the occurrence of rapid decay for only a subset of the communication topics necessitates caution in applying the procedure. Advantages and disadvantages of the reliance on within-Ss designs are considered. (24 ref)