Chapter 13 - Statistical Models for Analyzing Stability and Change in Happiness

Abstract

Based on a state-trait theory of happiness, basic concepts of the measurement of change are introduced—in particular, the distinction between direct versus indirect change assessment as well as between individual versus mean change. Latent variables models for measuring state variability and trait change are presented (latent state model, latent change model, latent autoregressive model, latent state-trait model, latent state-trait autoregressive model, latent state-trait change models for continuous and discontinuous change), and some guidelines for choosing an appropriate model are formulated. This chapter shows how the assumption of measurement invariance over time can be tested using these models, both for continuous as well as categorical observed variables. The chapter also presents an application of the models to the first four waves of the Household, Income, and Labour Dynamics in Australia (HILDA) Survey, and it illustrates how time-varying dummy variables representing life events can be included in the models.