摘要：The Assessment of Quality of Life - 6D scale (AQoL-6D) is a self-report instrument designed to provide a sensitive multidimensional evaluation of health related quality of life. The current paper assesses the construct, concurrent and convergent validity of the AQoL-6D in a combined longitudinal population sample drawn from across urban, regional and remote areas of Australia. The AQoL-6D was administered within the Hunter Community Study and the Australian Rural Mental Health Study over time (mean years lag = 3.90, SD = 1.30). Observations with sufficient data were used to confirm the construct validity of the AQoL-6D domains and higher-order structure using confirmatory factor analyses (CFA, N = 7915). The stability of this structure across cohorts and over time was assessed using multi-group CFA. Additionally, the concurrent validity (against the SF-36) and convergent validity of AQoL-6D domains and factors were assessed. The construct validity of the AQoL-6D domains was considered satisfactory. Two higher-order factors, representing the physical and psychological components of quality of life were identified (CFA model fit: RMSEA = .07, SRMR = .03; TLI = .96, CFI = .98). These factors displayed group and temporal invariance, as well as concurrent and convergent validity against a range of measures. Recommendations for the derivation of summary scores are provided, together with a provisional set of norms. The AQoL-6D is a useful tool for assessing quality of life impairment in epidemiological cohort studies, both cross-sectionally and over time. It displays appropriate levels of construct, concurrent and convergent validity. Conceptualisation of higher-order factors as representing the physical and psychological aspects of quality of life impairment may increase the sensitivity and appeal of the AQoL-6D, particularly for studies examining predictors of and changes in social and psychological outcomes.
关键词：Life Satisfaction ; Factor Score ; Domain Score ; Standardise Root Mean Square Residual ; Confirmatory Factor Analysis Model
