Modelling change in perceived quality of life (PQOL)

Almost all previous research on PQOL has examined static relationships. This paper, based on two waves of an Australian panel study (N=677), investigates change. We estimate a model in which changes in PQOL are seen as depending directly on changes in satisfaction with particular domains of life (marriage, job, etc.), which in turn are affected by life events occurring within these domains. This model accords with a “bottom-up” rather than a “top-down” theory of PQOL or well-being (Diener, 1984). Chief interest lies in the strong relationships between domain specific events and changes in domain satisfactions. In previous research life events have been directly linked to PQOL, or to physical or mental illness. The different result here arises from estimating a more appropriate model with an additional link (changes in domain satisfactions) in the chain. From a methodological standpoint, the paper suggests that life events inventories can be valuable research instruments and that recent validity criticisms (e.g. Schroeder and Costa, 1984) have been overstated. The second part of the paper deals with relationships between personality traits, social support and life events. We test hypotheses about links between (1) personality and subsequent life events (2) the “buffering” effects of personality and social support on the impact of adverse events (3) the effects of events on personality. Personality traits apparently do affect the subsequent incidence of life events. Extraversion is associated with favourable subsequent events, and neuroticism with adverse events. Internal locus of control apparently wards off adversity. An alternative hypothesis that these relationships are due to biassed reporting (i.e. personality traits are associated with degrees of bias) appears to be false. Contrary to findings from cross-sectional data, our panel results indicate that neither personality traits nor social support “buffer” the impact of adverse events on PQOL. Finally, life events appears to have small but statistically significant effects in modifying adult personality.