The theory of planned behavior in message design : testing the effectiveness of statistical evidence in messages about genetically modified foods

Abstract

The Theory of Planned Behavior (TPB) identifies three constructs, attitudes, subjective norms, and perceived behavioral, which are useful in predicting behavioral intentions toward an attitude object (Ajzen, 1985). A 3 (evidence type) x 4 (message topic) between-subjects design was used in this study to test the utility of the TPB in predicting behavioral intentions toward genetically modified (GM) foods and to examine the influence of statistical evidence on attitudes and behavioral intentions toward GM foods. Participants (N = 431) completed measures of math anxiety and math self-efficacy prior to a reading a risk message pertaining to GM foods or a control message. Participants then completed measures of message comprehension, attitudes, perceived behavioral control, subjective norms, and behavioral intentions toward GM foods as well as measures of their math ability and demographic information. Messages were collapsed based on pilot results that indicated they were equivalent and results supported the utility of the TPB in predicting behavioral intentions. There was not a main effect for evidence type on attitudes toward GM foods. Evidence type was significantly related to comprehension such that presentations that included a bar graph were comprehended more than the percentage formats. Significant negative correlations were found to exist between math self-anxiety and the variables of math self-efficacy and math ability. Math ability was positively correlated with comprehension of messages. However, comprehension was not a significant predictor of negative attitudes toward GM foods, perhaps due to a sleeper effect. Overall, these and other results are discussed in terms of the importance of using theory to guide message construction and the significance of the math competency construct. Limitations of the study and implications for future research are also discussed.