Mark Stemmer (FAU Erlangen-Nürnberg)

Person-centered methods using R: Log-linear models and Configural Frequency Analysis

While variable-centered psychology examines means or covariances of scale values of psychological constructs, the person-centered approach is concerned with combinations of characteristics that a person combines and their frequencies of occurrence. Characteristic combinations or configurations that occur significantly more frequently than predicted under the null hypothesis are called Types. Configurations that occur significantly less frequently than postulated under the null hypothesis are referred to as Antitypes (Stemmler, 2020). The methods presented are also suitable for smaller samples or for cases where the statistical assumptions for multivariate methods such as interval level data, normal distributions or homogeneous error variances are not fulfilled.

Besides an introduction to the open-source software R, the R package confreq is presented. It allows not only the analysis of configurations with the CFA, but also the analysis of log-linear models as well as the two-sample CFAs, which can be seen as a counterpart to the t-test for contingency tables. In the latest confreq version, covariates can also be used. All methods are demonstrated with small numerical examples and practiced on the user's own laptop.

The person-centered approach will not be presented here as an either-or alternative to the variable-centered approach, but it is intended to expand the thinking and evaluation possibilities in psychological research.

Prerequisites for participation: Only basic knowledge in multivariate methods is required; at the same time, the course offers an introduction to the software R!

Literature of interest:

Stemmler, M. (2018). Personenorientierte Entwicklungspsychologie als eine alternative Methode zum variablenorientierten Ansatz. In B. Kracke & P. Noack (Hrsg.), Handbuch der Entwicklungs- und Erziehungspsychologie. Heidelberg: Springer Verlag. doi:10.1007/978-3-642-54061-5_26–1

Stemmler, M. (2020). Person-centered methods: Configural Frequency Analysis (CFA) and Other Methods for the Analysis of Contingency Tables (second edition). Series: Springer Briefs in Statistics. New York: Springer Publishing Company.

Stemmler, M. & Heine, J.-H. (2016). Using Configural Frequency Analysis as a Person-Centered Analytic Approach with Categorical Data. International Journal of Behavioural Development, 41, 632–646. doi: 10.1177/0165025416647524.

Stemmler, M., Heine, J.-H. & Wallner, S. (2019). Analyzing Tree Structures With Configural Frequency Analysis and the R-Package confreq. Psychological Test and Assessment Modeling, 61, (4), 419–433.

Target Audience:

Participants should have knowledge of inferential as well as multivariate statistics. The participants will work on their own laptops with R version 4.0 (or newer) and an R editor (preferably R-Studio) already installed.