Presenters: Robert Gallop, Ph.D., West Chester University
Longitudinal data is extremely common in clinical research, whether from intervention studies or observational research.
Hierarchical linear models (HLM; also called multilevel or mixed effects models) are powerful regression-based models that have a number of advantages for longitudinal data.
This AMASS will introduce the basics of HLM for longitudinal data analysis in clinical research, with particular focus on psychotherapy research. In addition, this presentation will cover some of the complexities that can arise
with clinical data and how best to address them, including (a) impact of missing data in the HLM, (b) determination of the complexity of the covariance structure and its implication on model interpretation, (c) issues with centering,
(d) model diagnostics for HLM, (e) model formation including implementation dependent on the treatment of time and distribution of the outcome, and (f) model estimation.
Data from psychotherapy research settings will also be discussed as examples. Finally, this session will offer some caveats and advice for recognizing various complexities and proper procession
to ensure accurate implementation of the HLM and the interpretation of the results. Analysis examples will be presented using SPSS, though resources for other software packages will be provided.
You Will Learn:
1. How to estimate and interpret results from HLM analyses of longitudinal psychotherapy data.
2. How to assess for and fit more complex models for change, such as quadratic, cubic, and piecewise.
3. How to write SPSS commands for HLM via examples including sample syntax and screen captures of pull-down menus.
Gallop, R., & Tasca, G.A. (2009). Multilevel modeling of longitudinal data for psychotherapy researchers: II. The complexities. Psychotherapy Research, 19, 435-452.
Hedeker, D., & Gibbons, R.D. (2006). Longitudinal data analysis. Hoboken, NJ: Wiley.
Raudenbush, S.W., & Bryk, A.S. (2002). Hierarchical linear models (2nd ed.). Thousand Oaks, CA: Sage.
Tasca, G.A., & Gallop, R. (2009). Multilevel modeling of longitudinal data for psychotherapy researchers: I. The basics. Psychotherapy Research, 19, 429-437.
Presenters: Abby Lynn Braitman, Ph.D., Old Dominion University
Mplus is data analysis software focusing on latent variable modeling but includes a wide array of analysis capabilities beyond structural equation modeling (SEM).
Because of its ability to analyze complex data over a wide range of conditions, Mplus is increasingly popular as a statistical toolbox with clinical psychology researchers.
At the same time, Mplus is almost entirely driven by command-line syntax, which may be a hurdle for users unfamiliar with code-based analysis.
This AMASS will focus on familiarizing attendees with Mplus, including an overview of its function and characteristics, as well as basic and intermediate commands.
Topics covered will include path analysis, SEM, mediation, bootstrapping, latent growth modeling, and mixture modeling (i.e., latent class/profile analysis).
We will review both input and output as well as troubleshooting options. It is assumed that participants are familiar with the theory behind each analysis covered (regression, SEM, latent growth modeling, etc.) as the focus
of the presentation is on how to use Mplus to accomplish these analyses. Having a copy of Mplus is not required for participation, and those who might be considering Mplus as data analysis software are encouraged to attend.
You Will Learn:
1. How to execute several analyses using Mplus (e.g., path analysis, SEM, mediation, latent growth models, mixture modeling).
2. How to activate specific coding options (e.g., bootstrapping, technical output).
3. How to read and interpret Mplus output.
Muthén, L.K., & Muthén, B.O. (1998-2012). Mplus User's Guide (7th ed.). Los Angeles, CA: Muthén & Muthén.
[PDF freely available via Mplus website: https://www.statmodel.com/]