Stability in components and predictors of subjective well-being (SWB): implications for SWB structure

Abstract

A little investigated explanation for the high temporal stability in SWB scores is the stability of predictors. In the current investigation, over 400 participants were followed over a 48-month period during which SWB and predictor measures were taken at the beginning, the middle, and the end of the study. Predictors of SWB included demographic variables, domain satisfactions and stresses, current hassles and uplifts, and personality characteristics and styles. Component measures of SWB included both long- and short-term affect measures. As expected, long-term components of SWB yielded average temporal stability coefficients of 0.60 while short-term components averaged coefficients of only 0.35. With the exception of demographic variables, that fail to achieve significant predictive power and daily hassles, that have small predictive power, only personalty factors and domain satisfactions, averaged over all domains, reached the stability of long-term SWB components. When structural equation modelling was employed with significant SWB predictors, including prior SWB scores, top-down and bi-directional models provided the best fit for the data. These results are more consistent with our earlier hypothesis that SWB has both trait- and state-like properties than with one that attributes SWB stability solely to environmental and personality variables.