ABSTRACT
The main purpose of this chapter is to introduce an approach to investigating potential heterogeneity in higher-order trajectories of global domains (both as a CFM and a FCM). This approach involves combining second-order growth curve modeling (Chapters 3 through 6) and growth mixture modeling (GMM) (Chapters 7 and 8) procedures. Accordingly, in this chapter, two extensions to a GMM involving second-order growth curves (known as second-order growth mixture models, SOGMMs) are discussed. First, this chapter discusses a GMM extension of a curve-of-factors model (CFM). This chapter also discusses a GMM extension of a factor-of-curves model (FCM). Because both FCM and CFM take measurement errors in manifest variables are into account, they may provide a more precise test of the heterogeneity that exists within the sample population compared to a LGCM with composite measures. Furthermore, this chapter introduces a multidimensional growth mixture model (MGMM). Typically, this model does not include second-order growth factors. Instead, primary growth factors of multiple subdomains are used together as indicators of latent classes.
