The Satisfaction with Life Scale (SWLS): Psychometric Evaluation Among Managers Reveals Two Novel Subscales and Suboptimal Performance of the Shortened Three-Item Version (SWLS 3)

BackgroundLife satisfaction is widely used as a social indicator to direct governmental interventions that target individuals' quality of life. Issues of dimensionality and equivalent functioning of common measures such as the Satisfaction with Life Scale (SWLS) may negatively implicate inferences and decisions derived from those measures.MethodsUsing a cross-sectional design and a convenience sample of 255 Polish managers (mean age = 48.9 ± 8.2 years, 22.6% females), this study evaluated the SWLS through confirmatory factor analysis (CFA) and multigroup CFA for construct validity and measurement invariance. Its convergent, discriminant, predictive, and criterion validities were assessed through average variance extracted (AVE), heterotrait-monotrait (HTMT) ratio of correlations, SWLS correlations with its subscales and with measures of personality traits, COVID distress, and general organizational trait.ResultsData did not support the one-factor and two-factor (past and present life satisfaction) SWLS or the SWLS3. A novel two-factor SWLS of congruence between ideal and real self (items 1, 2, and 5) and achievement (items 3 and 4) displayed the best fit (χ2 = 7.38, DF = 4, CMIN/DF = 1.85, p = 0.117, CFI = 0.996, TLI = 0.991, RMSEA = 0.058, SRMR = 0.0136), invariance across gender and marital status, high convergent validity (AVE = 0.667 and 0.842), high internal consistency (α = 0. 851 and 0.914), acceptable discriminant validity (HTMT = 0.911), good predictive validity as noted by correlations with its subscales (r = 0.758 and 0.885, p <0.01), and adequate criterion validity—the SWLS and congruence were negatively associated with COVID-19 distress (r = -0.158 and -0.149, p <0.05), congruence positively correlated with age (r = 0.126, p <0.05), achievement correlated with openness to experience (r = 0.191, p <0.01), and all SWLS measures positively correlated with general organizational trait (r range = 0.263 to 0.356, p <0.01). ConclusionThe findings expand our understanding of how managers conceptualize life satisfaction: congruence (between ideal and real self) and achievement. A revised phrasing of item 5 is necessary as contributed to scalar variance across younger and older managers. More investigations are needed to confirm the findings.