The Factor Structure of the Satisfaction with Life Scale (SWLS): A Meta-Analytic Structural Equation Modeling

Abstract

This study aims to determine the most appropriate factor structure for the life satisfaction scale by using the meta-analytic structural equation modeling (MASEM) approach. For this purpose, we extracted 41 correlation matrices from 33 primary studies (N = 49316) in accordance with the inclusion criteria. In the first step of MASEM, we created the total correlation matrix according to the random-effects model. At this stage, we determined that there was a large and significant relationship between the scale items. In the second phase of MASEM, we established four models of SWLS (original single-factor model, modified single-factor model, two-factor model, and the first three-item model). As a result of the analysis, we determined that although the goodness-of-fit indices of the original single-factor model were at a “good” level, the model-data fit of modified-single-factor model and the two-factor model was better. However, we determined that the modified-single factor model was the most appropriate one since there was a high correlation (r = 0.92, p < .01) between the factors in the two-factor model, and its divergent validity could not be ensured. We determined that the first three-item model is a saturated model.