Explorations of Individual Change Processes and Their Determinants: A Novel Approach and Remaining Challenges 

• In Event: 3-195 - Leveraging Machine Learning Methods for Research on Child Development in the Big Data Era

Integrative Statement

Exploratory approaches to uncovering differences in change trajectories have primarily revolved around growth mixture models (GMM; Muthen & Shedden, 1999). In GMMs, a categorical latent variable is introduced into the growth model and the different latent classes can have different parameters of change or fundamentally different change trajectories. Recursive partitioning approaches to growth modeling have recently gained the attention of social scientists (Brandmaier, von Oertzen, McArdle, & Lindenberger, 2013; Stegmann, Jacobucci, Serang, & Grimm, 2018), whereby observed covariates are used to partition the data in the search for groups of participants with different change trajectories. A comparison of these approaches (Jacobucci, Grimm, & McArdle, 2017) highlighted how GMMs are better able to capture divergent trajectories because it does not rely on observed covariates to partition the data; however, recursive partitioning clearly indicates how the groups are formed. In this talk, we discuss a new approach that combines the ability to explore change trajectories with GMMs and understand associations with observed covariates. This approach is based on a newly developed R package MplusTrees (Serang, Jacobucci, Stegmann, & Grimm, 2018) that performs recursive partitioning and allows for estimation using Mplus (Muthen & Muthen, 1997-2018). The new approach is illustrated with longitudinal reading data from the Early Childhood Longitudinal Study – Kindergarten 1998-1999 Cohort. The challenges of searching longitudinal data are discussed along with recommendations.

Authors

• Kevin Grimm, Arizona State University
  Presenting Author

• Ross Jacobucci, University of Notre Dame
  Non-Presenting Author

• Gabriela Stegmann, Arizona State University
  Non-Presenting Author

• Sarfaraz Serang, Utah State University
  Non-Presenting Author