Research Seminar
Fitting cross-lagged panel models with random effects: A linear mixed modeling approachBart JacobsKU Leuven | |
| Abstract: | An increasingly popular class of statistical models is that of the models for cross-lagged analyses. In such analyses, two variables, each measured at multiple time points, are studied together. In practice, these analyses are often set up within the framework of structural equation models. Important disadvantages of this approach include that it does not allow for random effects and that adding covariates may be a burden. To overcome these problems, we will present a linear mixed modeling approach, which allows for the same flexibility as structural equation models for cross-lagged analyses, but which has the advantage that extra covariates and random effects can be easily added. This approach will be illustrated by means of longitudinal data on depression and self-esteem. |
| Date: | Tue Feb 23, 12:15 pm - 1:15 pm |
| Place: | room 02.51 (Department of Psychology, Tiensestraat 102, 3000 Leuven) |