|Abstract: ||The theory of planned behavior is an expectancy-value model based on the
assumption that the planned behavior, in turn, is determined by attitude, subjective norms, and perceived behavior control (Ajzen, 1991). Recent studies have shown that a person’s sense of self-identity may influence one’s behavior (Terry, Hogg, & White,1999). And the relationship between self-identity and exercise behavior is positive (De Bruijn & Van den Putte, 2012). In addition, Jackson, Crawford, and Godbey(1993) proposed that the constraints that individuals perceive do not always lead to non-participation; people may use appropriate negotiated strategies to overcome the constraints. Some successful negotiation strategies of leisure constraints, such as individual’s motivation, personal control, and self-efficacy, are frequently used.
In this paper we plan to expand the theory of planned behavior by adding two
mediating variables of the exercise self-identity and the constraint negotiation strategy. That is, this study aims to investigate the mediating effects of the exercise self-identity and the constraint negotiation strategy on the relationship between three precursors in the theory of planned behavior and participation intention in Curves exercise.
Six scales were included in the questionnaire: attitude toward exercise, subject norm, perceived behavior control, exercise self-identity, the constraint negotiation,and behavioral intention. A seven-point scale was used from “1= very disagree” to “7=very agree.” We sent a preliminary questionnaire to 35 participants in one of Curves centers on a voluntary basis. And we received responses from 34 people (97%) within two weeks. Based on the analysis of the pilot study, the questionnaire is revised. Then, using all the members of Curves (only for female) as the population, a stratified sampling was used. In total, questionnaires were sent out to twenty Curves centers in the whole Taiwan by a regular mailing system. Each Curves center helped distribute the questionnaires to 30 members. In other words, 600 questionnaires were sent out to members. Then, we received responses from 555 females (92.5%). The range of age is between 14 and 70 with a mean age of 41. The Cronbach Alphas for the scales were between 0.782 and 0.959. Modification indexes were used to revise the scales in the confirmatory factor analysis. Then, the structural equation modeling (SEM) is used to examine the proposed model. The model fit is acceptable: The Normed Chi-Square index (NCI)=3.839, Goodness of fit index (GFI) = 0.911, root mean square error of approximation (RMSEA) = 0.072, Goodness of fit index (GFI) = 0.911, adjust good of fit index
(AGFI) = 0.875, Root Mean square Residual (RMR) = 0.072, Normed-fit index(NFI)= 0.928, Incremental fit index (IFI) = 0.945, Comparative Fit Index (CFI) =0.945. Most of the path coefficients are significant: 0.308 for attitude, 0.253 for exercise self-identity, and 0.173 for negotiation were positively significant with behavior intentions; 0.239 for attitude, 0.145 for subjective norm, and 0.426 for perceived behavior control were positively significant with exercise self-identity; 0.157 for attitude, 0.189 for subjective norm, and 0.245 for perceived behavior control were positively significant with negotiation. Some path coefficients are not significant. Variables of the subjective norm and perceived behavior control had no significant relationship with behavior intentions.
In terms of the mediation roles of the exercise self-identity and the constraint negotiation, the results indicated that both exercise self-identity and the constraint negotiation partially mediated the relationship between attitude and intention, but seemed to fully mediate the relationship between subject norm and intention and the relationship between perceived behavior control and intention. The mediating effect of the exercise self-identity seems to be stronger than that of the constraint negotiation. On the whole, this model explains 30% of variance in behavioral intentions (R square = 0.30). These findings suggest that the exercise self-identity and the constraint negotiation strategy can be mediators in the theory of planned behavior when applied into Curves exercise. In conclusion, it is majorly the active attitudes of the female participants to engage in the circuit exercise of Curves centers through the enhancement of psychological exercise self-identity and adoption of constraint negotiation strategy.