The Random Intercept Cross-Lagged Panel Model

The random intercept cross-lagged panel model (RI-CLPM) is rapidly gaining popularity as a structural equation modeling approach to longitudinal data. It allows for the decomposition of observed scores into within-person dynamics and between-person differences.

This website is a supplement to “Three Extensions of the Random Intercept Cross-Lagged Panel Model” by Mulder and Hamaker (under review). It contains Mplus syntax and lavaan code for specifying the basic RI-CLPM and the following three extensions:

  1. including a time-invariant predictor and outcome,
  2. doing multiple group analysis, and
  3. using multiple indictors for variables.

Use the top menu to navigate to the Mplus syntax or lavaan code. The lavaan (R) tab contains additional code for performing the \(\bar{\chi}^{2}\)-test (chi-bar-square test) in R. This test is used for comparing nested models where the more parsimonious model is based on constraining parameters on the bound of the parameter space (e.g., constraining a variance to 0).


Data

You can find simulated example datasets (1189 units, 5 repeated-measures) on Github to get hands-on experience with this modeling approach. The example data are motivated by Narmandakh et al. (2020) who obtained five waves of data from 1189 adolescents on their sleep problems and anxiety over 15 years.


Contact

Questions can be addressed to the first author of the corresponding article, Jeroen Mulder.


References

Narmandakh, Altanzul, Annelieke M. Roest, Peter de Jonge, and Albertine J. Oldehinkel. 2020. “The bidirectional association between sleep problems and anxiety symptoms in adolescents: a TRAILS report.” Sleep Medicine 67 (March): 39–46. doi:10.1016/j.sleep.2019.10.018.