Probing Theories of Individual Well-being: A Comparison of Quality-of-life Models Assessing Neighborhood Satisfaction

Abstract The extensively documented approach to quality of life proposed by Campbell, Converse, and Rodgers (1976) is translated into a testable theoretical model. Herein labeled the "first-order structural model," this approach springs from a popular interpretation of Lewinian concepts. Critical examination of this interpretation leads to several modifications of the measurement and causal structures hypothesized by Campbell et al., each of which is submitted to test. Likewise, an alternative theoretical approach based on a second, less well known interpretation of the same Lewinian concepts, herein labeled the "second-order structural model," is tested. These models are evaluated by structural equation techniques on a sample of 419 respondents dealing with neighborhood satisfaction. Analyses indicate that both the first-order and second-order structural models yield statistically acceptable accounts of neighborhood satisfaction, but that the second-order model performs better on several modeling criteria. The implications for using structural equation techniques to evaluate competing theory-motivated models are discussed.