Validating the satisfaction with life scale among early adolescents: psychometric assessment using item response theory

Abstract

The Satisfaction with Life Scale (SWLS) has been extensively validated using Classical Test Theory, mostly relying on factor analytic methods among adult samples. The current study used Item Response Theory to validate SWLS among a sample of early adolescents in California and examine associations between SWLS and tobacco and cannabis use. Data were collected from ninth-grade students (N = 2552) attending 10 public high schools in Los Angeles, California in 2013. Nonparametric and parametric item response modeling were used to validate the SWLS. Binary logistic regression was used to examine the associations between SWLS and tobacco and cannabis use. Item discrimination indices were above 1.80, indicating all items functioned appropriately in terms of measuring the construct and separating participants of different levels of life satisfaction. The test information curve indicated that the SWLS was best for discriminating between respondents with low to average life satisfaction. Participants who reported high scores on the SWLS, compared with those who reported low scores, were significantly less likely to report lifetime e-cigarette, cigarette, and cannabis use. The SWLS possessed excellent psychometric properties among a sample of early adolescents in California. Findings may be useful for scholars and practitioners to screen for subjective well-being in early adolescence, and target interventions focused on improving adolescent health & well-being, which may help prevent substance use initiation or sustained use.