Longitudinal associations between exercise identity and exercise motivation : a multilevel growth curve model approach

Saved in:
Bibliographic Details
Title translated into German:Längsschnittuntersuchungen von Bewegungsidentität und Bewegungsmotivation : ein mehrstufiger Wachstumskurvenmodellansatz
Author:Ntoumanis, N.; Stenling, A.; Thøgersen-Ntoumani, C. ; Vlachopoulos, S.; Lindwall, M.; Gucciardi, D.F.; Tsakonitis, C.
Published in:Scandinavian journal of medicine & science in sports
Published:28 (2018), 2, S. 746-753, Lit.
Format: Publications (Database SPOLIT)
Publication Type: Journal article
Media type: Electronic resource (online) Print resource
ISSN:0905-7188, 1600-0838
Online Access:
Identification number:PU201803001747

Author's abstract

Past work linking exercise identity and exercise motivation has been cross-sectional. This is the first study to model the relations between different types of exercise identity and exercise motivation longitudinally. Understanding the dynamic associations between these sets of variables has implications for theory development and applied research. This was a longitudinal survey study. Participants were 180 exercisers (79 men, 101 women) from Greece, who were recruited from fitness centers and were asked to complete questionnaires assessing exercise identity (exercise beliefs and role-identity) and exercise motivation (intrinsic, identified, introjected, external motivation, and amotivation) three times within a 6 month period. Multilevel growth curve modeling examined the role of motivational regulations as within- and between-level predictors of exercise identity, and a model in which exercise identity predicted exercise motivation at the within- and between-person levels. Results showed that within-person changes in intrinsic motivation, introjected, and identified regulations were positively and reciprocally related to within-person changes in exercise beliefs; intrinsic motivation was also a positive predictor of within-person changes in role-identity but not vice versa. Between-person differences in the means of predictor variables were predictive of initial levels and average rates of change in the outcome variables. The findings show support to the proposition that a strong exercise identity (particularly exercise beliefs) can foster motivation for behaviors that reinforce this identity. We also demonstrate that such relations can be reciprocal overtime and can depend on the type of motivation in question as well as between-person differences in absolute levels of these variables.