An empirical comparison of within-subjects and between-subjects expectancy theory models

One of the most consistent criticisms of expectancy theory research for the prediction of effort is that it has not been tested using the within-subjects choice model that the theory requires. The choice model proposed by the theory generates a motivational force score (MFS) for each of several effort levels for each subject and predicts that each will choose the level of effort which has the highest MFS for that subject. This study operationalizes the choice model and compares it with a difference model (derived from subtracting the MFS for low effort from the MFS for high effort) and a single-alternative model (MFS for high effort only). Subjects were 74 undergraduates who estimated valences and expectancies for the outcomes of six general student activities. The difference model, not the choice model, was generally the best predictor, while the single-alternative model predicted least well. Within-subjects predictions were of greater magnitude than between-subjects predictions. The results indicated that sufficiently designed future studies should gather data for three levels of effort (high, medium, and low) and then compare the models to determine the most effective predictor for that particular situation.