Combining Intention with Investment to Predict Withdrawal Behavior

The present study assessed the extent to which the withdrawal behaviors for two samples of Americans (254 health care workers and 45 ROTC students) could be predicted using self-reported information about withdrawal intention and investment size. Hierarchical regression analyses were performed for each sample, and the results were consistent: For both samples, the main effect of intention (entered first) contributed significantly to the prediction of withdrawal behavior. Neither the main effect of investment (entered second) nor the Intention × Investment interaction effect (entered third) contributed significantly to the prediction of withdrawal behavior for either sample. These findings are inconsistent with prior research, which showed that all three parameters (intention, investment, and their interaction) made significant contributions to the prediction of participation in music-related activities. We propose that the inconsistent findings relate to the degree of personal relevance of the criterion measures.