Description
Data in psychotherapy research are often generated in settings where the use of (generalised) linear models provides a flexible approach to formalise research questions and drive the design of quantitative studies. Both in observational as well as randomised studies data can occur nested in participants or services, have longitudinal components, and/or time-varying predictors (eg. time as a common case). In all these situations mixed effects models are a useful framework and related analytic strategies or subclasses of this modelling approach (eg. hierarchical linear models) are often applied as a default. The workshop is roughly separated into three parts. The workshop will first briefly introduce participants to the formal background of mixed effects models and their relationship to other commonly applied multivariate and multivariable techniques. This will quickly lead into the second part, where the implementation of these models in R will be presented. The main part of the workshop will focus on classic teaching examples as well as examples from the presenter's past and ongoing research to introduce a range of practical use cases for mixed models in psychotherapy research. While the workshop is planned with enough content to benefit people not using R, for use of the applied examples, workshop syntax etc. basic knowledge of R is a requirement. The workshop will end with the discussion of further topics such as non-linear extensions, missing data, and power calculations.Period | 22 Sept 2022 |
---|---|
Event title | 9th Society for Psychotherapy Research Europe Chapter Meeting: Therapist Responsiveness: Challenges and Opportunities |
Event type | Conference |
Location | Rome, ItalyShow on map |
Related content
-
Activity
-
How to use simulation methods for power analyses and sample size calculations in cluster-randomised designs
Activity: Talk or presentation types › Oral presentation