ABSTRACT
Objective
This study aimed to assess evidence of validity and reliability of Exercise Dependence Scale-Revised (EDS-R).
Methods
A non-probabilistic sample consisted of 709 exercise practitioners from all over Brazil, aged 12 to 73 years (M = 25.7; SD = 8.43), most part male (55.1%) and single (75.5%). Participants answered EDS-R and demographic issues. About factor structure, procedures of exploratory and confirmatory factor analysis were used. Additionally, to assess measurement reliability, Cronbach’s alpha were calculated (internal consistency) and r Pearson correlations.
Results
Exploratory analysis (principal axis factoring) suggested an explained variance of 62.7% for a seven-factor structure. Total internal consistency of the scale was 0.88, with Cronbach’s alpha indices ranging from 0.68 to 0.89 among the factors. The correlation coefficients ranged from 0.10 to 0.65, supporting interrelated dimensions. Confirmatory factor analysis corroborated a seven-factor structure, based on satisfactory indicators of model goodness of fit [χ2 (168) = 288.21; p < 0.001, χ2/df = 1.78, GFI = 0.93, CFI = 0.95 and RMSEA = 0.04 (90% CI = 0.037 to 0.055)].
Conclusions
These findings supported the psychometric adequacy of EDS-R, according exploratory and confirmatory analyzes, in line with the theoretical model of exercise dependence.
Dependency; exercise; validation studies; psychometric parameters
