When Do Measured Attitudes Reflect Past Behavior?

Abstract

Building on current literature on the effect of measurement order, this study aimed to examine the effect of measurement order on the correlation among measured constructs of attitude, cognition, and past behavior. We particularly focus on the effect of past behavior, as a knowledge activator, on the attitude formation since research also demonstrates that past behavior, in addition to attitudes, can be a good predictor of future behavior. The current study suggested that when cognitions were measured prior to attitudes, the results were consistent with the attitude construction process. Perhaps the most interesting finding is that the self-reported behavior predicts the self-reported attitudes only when the measurement of the former precedes the latter. Measuring behavior upfront, prior to attitudes, can weaken the attitude-behavior consistency.

Building on current literature on the effect of measurement order, this study aimed to examine the effect of measurement order on the correlation among measured constructs of attitude, cognition, and past behavior. We particularly focus on the effect of past behavior, as a knowledge activator, on the attitude formation since research also demonstrates that past behavior, in addition to attitudes, can be a good predictor of future behavior. The current study suggested that when cognitions were measured prior to attitudes, the results were consistent with the attitude construction process. Perhaps the most interesting finding is that the self-reported behavior predicts the self-reported attitudes only when the measurement of the former precedes the latter. Measuring behavior upfront, prior to attitudes, can weaken the attitude-behavior consistency.