Mplus 8 offers the possibility to analyze intensive longitudinal data, such as obtained with ecological momentary assessments, experience method sampling, ambulatory assessments, and daily diaries, using dynamic SEM (DSEM). DSEM includes N=1 time series analysis that can be used to model the dynamics within a particular individual over time, for instance the reciprocal effects that momentary stress and rumination may have on each other. Additionally, DSEM includes multilevel extensions of time series models, such that at level 1 a time series model is used to model the within-person dynamics of a process, while at level 2 we can model individual differences in the parameters that capture these dynamics. DSEM can handle multivariate outcomes as well as latent variables, and random effects can be both predicted from but also predictors of level 2 variables. In this workshop we will begin by covering what dynamic multilevel modeling is and why this is of interest, and then we will explore some of the possibilities that DSEM in Mplus has to offer. In addition, I will give some live demonstrations of these new modeling options in Mplus version 8 (prior to its actual release, which is now expected in take place in the second half of April).
Some basic knowledge of Mplus is preferable for this course. You can obtain this knowledge from the one-day course Mplus: How to get started?, which is taught before the Multilevel conference, on April 11. For more information and registration, click here.
Associate Professor – Methodology and Statistics