Joint Quantitative Brownbag

Speaker

Sacha Epskamo

Dr. Sacha Epskamo
Faculty of Social and Behavioural Sciences
University of Amsterdam

Sacha Epskamp is an Assistant Professor in Psychological Methods and Psychometrics at the University of Amsterdam. Dr. Epskamp’s worked on integrating network modeling techniques in the field of psychometrics, with applications in clinical, social and developmental psychology. In addition to theoretical work he has developed several software package to facilitate the use of their methods for empirical researchers including statcheck and semPlot. He has been awarded the Psychometric Dissertation Prize in 2018 on network psychometrics, and is Associate Editor for Psychometrika and Psychological Methods.

Title

Introducing psychonetrics, an R package for (dynamic) structural equation modelling and network psychometrics.

Abstract

In this talk, I introduce the R package psychonetrics (psychonetrics.org), which is a new open-source software package for structural equation modelling (SEM) and network psychometrics. The psychonetrics package currently includes multi-group (full-information) maximum likelihood estimation and weighted least squares estimation of (dynamic) latent variable models for cross-sectional, time-series and panel data. In addition, it includes the option for every variance-covariance matrix to be modelled as a Cholesky decomposition or a Gaussian graphical model (GGM; networks model of partial correlations). This allows for a large number of models that are closely related to SEM to be estimable: multi-level and meta-analytic GGMs (Epskamp et al., 2020), latent and residual network models (Epskamp et al., 2017), and latent-variable graphical vector-autoregression models for time-series and panel data (Epskamp, 2020). Finally, the package provides functionality to simplify exploratory estimation and model search for all included modelling frameworks. The talk will include a broad overview of the functionality of psychonetrics and empirical examples showcasing its potential.

Background Reading

  • Epskamp, S. (2020). Psychometric network models from time-series and panel data. Psychometrika, 85, 206-231. https://doi.org/10.1007/s11336-020-09697-3
  • Epskamp, S., Isvoranu, A.-M., & Cheung, M. (2020). Meta-analytic Gaussian network aggregation. PsyArxiv Preprint. https://doi.org/10.31234/OSF.IO/236W8
  • Epskamp, S., Rhemtulla, M. T., & Borsboom, D. (2017). Generalized network psychometrics: Combining network and latent variable models. Psychometrika, 82(4), 904-927. https://doi.org/10.1007/s11336-017-9557-x