ABSTRACT
Piecewise growth models are one approach to subdivide a series of measurements into meaningful segments, and to summarize important aspects of change in each segment. In piecewise modeling, the different periods may be hypothesized to have different growth patterns for individuals in the same sample. Because the mean intercept for the constrained model is different than the mean intercept for the piecewise model, the scaling for the slope mean also differs. Although the piecewise model allows for a comparison of the different growth trends, these trends share a common intercept. Unlike the piecewise approach, the interrupted time series experiment (ITSE) and simple change latent growth modelings are capable of detecting simultaneous differences in both level and slope, and providing the researcher with tests of significance for the two necessary indicators of the effect of the intervention. The primary disadvantage of ITS designs is that results are susceptible to internal validity threats.
