Accounting for Interindividual Differences in Trajectories When Assessing Cross-Lag Relationships

• In Event: AERA Poster Session 7
  In Poster Session: Division D Section 2 Poster Session 2

Abstract

This paper explores the relationship between the Random Intercept Cross-Lag Panel Model (RI-CLPM) and Latent Growth Curve Model with Structured Residuals and Unspecified Growth Trajectory (LGCM-SR-UGT) showing that the RI-CLPM is nested in the LGCM-SR-UGT. This nested structure allows for straightforward model comparison to determine if inter-individual variability in change should be accounted for when fitting cross-lag panel models. A simulation study was conducted to evaluate consequences of ignoring inter-individual variability in change on cross-lag estimates when fitting RI-CLPM to data exhibiting such variability. On the whole, cross-lag paths and standard errors are underestimated with increasing inter-individual variability in change.

Authors

• Paul Wesley Scott, University of Pittsburgh

• Feifei Ye, The RAND Corporation