The motivation for physical activity measure-revised: A within- and between-network construct validation study

Abstract

The purpose of this study was to evaluate select aspects of construct validity associated with scores from the Motivation for Physical Activity Measure-Revised (MPAM-R; Ryan et al., 1997). Using a within- and between-network approach to construct validation, two independent samples of physically active women (N1 = 321; N2 = 342) completed the MPAM-R plus assessments of psychological need satisfaction and behavioral regulations for exercise based on Self-Determination Theory (SDT; Deci & Ryan, 2002). Global model fit of the 5 factor, 30-item MPAM-R measurement model in the calibration sample was suboptimal (CFI/IFI = 0.85; RMSEA = 0.10; SRMR = 0.08). Iterative removal of MPAM-R items trimmed the 5-factor MPAM-R measurement model to 20 items that appeared tenable in the samples used for model calibration (CFI/IFI = 0.93; RMSEA = 0.09; SRMR = 0.05) and model cross-validation (CFI/IFI = 0.93; RMSEA = 0.08; SRMR = 0.06). Score reliability estimates with the 20-item model ranged from 0.66 to 0.93 across the samples. Greater psychological need satisfaction and more autonomous exercise motives were positively correlated with more intrinsic motives assessed with the trimmed MPAM-R items (r12 = 0.14 to 0.68) rather than the extrinsic motive of appearance (r12 = 0.01 to 0.06). Overall, this study offers initial support for a reduced item MPAM-R that is consistent with SDT which may yield further insights into the motivational dynamics of physical activity behaviour.

Acknowledgments: This study was supported by a grant provided by the Social Sciences & Humanities Research Council of Canada