Does the transtheoretical model need an attitude adjustment? Integrating attitude with decisional balance as predictors of stage of change for exercise

Objectives. To compare conceptually similar decision-making components from the theory of planned behavior (attitude) and the transtheoretical model (pros and cons) to determine the extent to which attitude towards exercise adds to the prediction of stage of exercise behavior beyond that of pros and cons. Method. A sample of college undergraduates (N=223) were given stage of change, attitude and decisional balance measures regarding their exercise behavior. Structural equation modeling (SEM) was used to test the underlying measurement structure of the decision-making components, while a series of discriminant function analyses (DFAs) was performed using a combination of pros, cons, and cognitive and affective attitudes as predictors of membership in one of the five stages of change. Results. SEM determined that a correlated four-factor model, which included pros and cons and two attitude subscales, provided the best fit to the data. The DFAs revealed that the addition of attitude components to pros and cons significantly increased the overall explained variance across the stages of change from 32% to 56%, and improved the predictive ability of pros and cons from 31.2% to 48.2%. Conclusions. Although conceptually related, pros, cons and attitude were not closely linked at a construct-measurement level. Furthermore, the addition of attitude to pros and cons increased the overall explained variance across stages of change and improved the predictive ability of pros and cons alone. The measurement model and DFA results taken in combination strongly suggest that the addition of cognitive and affective attitudes may strengthen the decision-making aspect of the trans theoretical model. (C) 2001 Elsevier Science Ltd. All rights reserved.