Project
Bayesian Methods for Network Analysis of Longitudinal Psychometric Data

My PhD project focuses on the development of new statistical models and evaluation methodologies for Bayesian psychometric networks, with particular emphasis on intensive times series data (e.g., ecological momentary assesments or EMA) [1]. Specifically, the project aims to extend existing frameworks of Markov Random Fields (MRFs) [2] within a Bayesian context [3], addressing both methodological innovations and practical applicability.
The project builds upon and extends the foundational research conducted by the Bayesian Graphical Modeling Lab which has developed core methodologies for cross-sectional network analysis. These earlier contributions consist of the development of network models that fit psychometric data, development of model-averaged tests for network topologies [4, 5]. In the next phase, the project will be extended by modeling longitudinal psychometric data and properties of the models, includingheterogeneity between and within persons. This should result in two main developments:

1. The development of explanatory dynamic networks: Evaluate the support for the inclusion of exclusion of covariates in explanatory models using BMA. Development of stochastic search methods for variable selection for the collection of dynamic MRFs.

2. The development of time varying regression: Modeling change in individual’s behavior using time-varying covariates and assess their support using BMA.

To ensure that these methodological advances are accessible to applied researchers, the project also includes the development of user-friendly research software. Specifically, new functionality will be implemented in open-source R packages (e.g. by continuing the development of the ‘bgms’ R-package) and integrated into the open-source research software JASP.
Overall, this project aims to contribute methodological advancements of the field of psychometrics, with implications that extend to related domains in psychological and behavioral research.

[1] Shiffman, S., Stone, A. A., & Hufford, M. R. (2008). Ecological momentary assessment. Annu. Rev. Clin. Psychol., 4(1), 1-32.
[2] Kindermann, R., & Snell, J. L. (1980). Markov random fields and their applications (Vol. 1). American Mathematical Society.
[3] Marsman, M., van den Bergh, D., & Haslbeck, J. M. (2025). Bayesian analysis of the ordinal Markov random field. Psychometrika, 90(1), 146-182.
[4] Marsman, M., Waldorp, L., Sekulovski, N., & Haslbeck, J. (2024). A Bayesian Independent Samples T Test for Parameter Differences in Networks of Binary and Ordinal Variables (No. f4pk9_v1). Center for Open Science.
[5] Sekulovski, N., Keetelaar, S., Huth, K., Wagenmakers, E. J., van Bork, R., van den Bergh, D., & Marsman, M. (2024). Testing conditional independence in psychometric networks: An analysis of three bayesian methods. Multivariate Behavioral Research, 59(5), 913-933.

Supervisors
Dr. Maarten Marsman
Dr. Lourens Waldorp

Financed by
NOW – VIDI

Period
1 September 2025 – 31 August 2029