The reciprocal relationship between husbands and wives’ marital forgivingness: a two-wave cross-lagged latent difference score analysis of ten-year data

Cross-lagged latent difference score (LDS) models complement cross-lagged regression models, but are better suited to detecting differences in intrapersonal change and examining the relation between changes in different variables across time. In this article we present cross-lagged LDS models as a method for conceptualizing and measuring change in two-wave dyadic data. The statistical analysis of these models is illustrated using data on marital forgivingness collected from 61 couples at two time points separated by a 10-year interval. The results support the view that cross-lagged LDS models can be an appropriate means to analyze within-person change over two occasions in the context of nonindependent couple data. This is true even when sample size prohibits the estimation of cross-lagged LDS models through common factors using multiple indicators. Conditions that increase model reliability in the absence of multiple indicators are described.