The factorial validity and psychometric properties of the Polish version of Exercise Dependence Scale-Revised (EDS-R)

Introduction
Physical exercise is a hallmark of physical and mental well-being. The benefits of re­gular exercise of moderate intensity were proven in case of various conditions, disorders and diseases. However, in some cases, excessive exercise may lead to loss of control over the behaviour and to suffering due to various negative consequences. This condition is referred to as exercise dependence. The purpose of this study was to verify the factorial structure, internal validity, reliability and criterion validity of the 21-item Exercise Dependence Scale-Revised (EDS-R) in a sample of Polish recreational exercisers selected according to the convenience method.

Material and methods
In studies 1 and 2 there were 625 participants (52% female) who exercised recreationally from 3 to 5 times a week. In both studies, the participants filled in a Polish version of the EDS-R and answered demographic questions and questions about the frequency and duration of their exercise activity.

Results
In study 1, an Exploratory Factor Analysis was conducted, a five-factor solution was found. The internal consistency for the subscales and the total scale were high. In study 2, a Confirmatory Factor Analysis was performed though neither the five or seven-factor solution yielded an acceptable fit.

Discussion
Study 1 showed promising results suggesting that the five-factor EDS-R has a very good fit to the Polish data. The results of the study 2 are inconclusive as neither of the tested models fit the data. It is important to note that some international studies showed a lack of scale invariance both between different nationalities and different ages persons.

Conclusions
Further studies are necessary to investigate the factorial structure of EDS-R-PL and, until this is confirmed, the scale and the results should be analysed with caution.