About the workshop
Recent years have seen a rise of the network approach to psychology: modeling human behavior as a complex system of interacting psychological, biological, and sociological components. This approach has been used in many areas of psychological research, such as mental health, attitude formation, and intelligence. While it originated from a complex-systems point of view, it has since inspired new psychometric modeling frameworks implemented in open-source and accessible software — now known as the field of network psychometrics. Network psychometrics is both an alternative to and complementary with traditional psychometric modeling techniques such as structural equation modeling (SEM).
This workshop introduces attendees to the core ideas of network psychometrics
and teaches how to perform psychometric network modeling in R.
The morning session provides a general introduction to network
analysis of cross-sectional data: the theoretical foundations of the network
approach, undirected network models, estimating network models from data (using
the bootnet R package and the graphical JASP interface), and
assessing the stability, comparability, and replicability of estimated networks.
The afternoon session discusses how network models can be used
in a psychometric modeling framework that extends to confirmatory analyses,
latent variable modeling, multi-group designs, and longitudinal data, using the
psychonetrics R package for SEM and psychometric network modeling.
Schedule
Morning — Introduction to Network Psychometrics (Adela Isvoranu)
| Time | Topic |
|---|---|
| 9:00 – 9:45 | Part 1: Welcome & theoretical foundations of network psychometrics — introductions, aims of the day, materials & setup; what is a network?; history and background: psychology as a complex system and the network approach to psycho(patho)logy; applications in mental health research and beyond |
| 9:45 – 10:45 | Part 2: Undirected network models & estimation — the Gaussian graphical model; five interpretations of network edges; estimating network models from data: modeling framework (GGM / Ising / MGM), estimation framework (multivariate vs. univariate), and model selection (thresholding, LASSO regularization, model search); which method to choose?; R code examples (bootnet) |
| 10:45 – 11:00 | Coffee break |
| 11:00 – 11:30 | Part 3: Network stability, comparison & replicability — bootstrapped edge-weight accuracy; centrality and its stability (CS-coefficient); comparing networks across groups |
| 11:30 – 12:20 | Practical session — work through Exercise Set 1 (estimating and analyzing network models) with instructor support |
| 12:20 – 12:30 | Part 4: Outlook — networks for longitudinal data — temporal and contemporaneous networks with graphicalVAR and mlVAR |
| 12:30 – 1:30 | Lunch (on your own — local establishments) |
Afternoon — Psychometric Network Modeling with psychonetrics (Sacha Epskamp)
| Time | Topic |
|---|---|
| 1:30 – 2:00 | Part 5: Introduction to psychonetrics — network psychometrics meets SEM; the psychonetrics workflow (specify → run → inspect → modify → compare); saturated GGMs, pruning, and stepwise model search; confirmatory network analysis and fit measures |
| 2:00 – 2:30 | Practical 1 + coffee break — work through Exercise Set 2 (GGMs in psychonetrics) with instructor support; refreshments are served alongside |
| 2:30 – 3:00 | Part 6: Psychometric network models — testing networks on new data; latent variable models and latent network models; multi-group analysis and invariance testing; network temperature; meta-analytic network aggregation; new methods (some experimental) |
| 3:00 – 3:30 | Practical 2 — work through Exercise Set 3 (latent network models & multi-group analysis) with instructor support |
| 3:30 – 4:15 | Part 7: Longitudinal network models — n = 1 time-series (VAR / graphical VAR); temporal, contemporaneous and between-person structures; panel data models (panelvar / panelgvar); comparison with RI-CLPM models (ri_clpm); dynamic latent variable models (dlvm1) |
| 4:15 – 4:45 | Practical 3 — work through Exercise Set 4 (longitudinal network models) with instructor support |
| 4:45 – 5:00 | Part 8: Future outlook & closing — new developments in psychonetrics (Blume-Capel Ising models, multi-level latent variable models); Exploratory Graph Analysis; further reading; closing & Q&A |
Materials
Slides — morning session
- Part 1: Theoretical Foundations of Network Psychometrics
- Part 2: Undirected Network Models & Model Selection
- Part 3: Network Stability, Comparison & Replicability
- Part 4: Outlook — Network Models for Longitudinal Data
Slides — afternoon session
- Part 5: Introduction to psychonetrics
- Part 6: Psychometric Network Models
- Part 7: Longitudinal Network Models
- Part 8: Future Outlook & Related Work
Exercises & data
- Exercise sets with solutions and data (zip)
- Set 1 — Estimating and analyzing network models (morning practical session, 11:30)
- Set 2 — GGMs in psychonetrics (afternoon Practical 1, 2:00)
- Set 3 — Latent network models & multi-group analysis (afternoon Practical 2, 3:00)
- Set 4 — Longitudinal network models (afternoon Practical 3, 4:15)
AI workshop assistant
- Set up your preferred AI (ChatGPT, Claude, …) as a workshop assistant — companion documents and prompts for each session, including help with installing R and working through the exercises.
Recommended reading
- Isvoranu, A. M., Epskamp, S., Waldorp, L. J., & Borsboom, D. (Eds.). (2022). Network Psychometrics with R: A Guide for Behavioral and Social Scientists. Routledge.
- Epskamp, S. (2020). Psychometric network models from time-series and panel data. Psychometrika, 85(1), 206–231.
- Epskamp, S., Borsboom, D., & Fried, E. I. (2018). Estimating psychological networks and their accuracy: A tutorial paper. Behavior Research Methods, 50(1), 195–212.
Preparation
You can attend the full workshop without running any software. If you would like to follow along with the practical parts and exercises on your own laptop, please install the following before the workshop:
install.packages(c("qgraph", "bootnet", "psychonetrics",
"graphicalVAR", "mlVAR", "dplyr",
"NetworkComparisonTest", "semPlot"))
- Optionally: JASP (free), for the point-and-click network module used in the morning demonstration
Need help installing R? Our AI workshop assistant setup can walk you through it step by step.
Instructors
Adela Isvoranu
Dr. Isvoranu's research focuses on pathways from mental health to mental illness and fuzzy boundaries between psychopathological conditions, arguing for complexity-based conceptualizations of psychopathology that focus on symptoms and their interactions. She is the main author of the textbook Network Psychometrics with R: A Guide for Behavioral and Social Scientists.
Sacha Epskamp
Dr. Epskamp's PhD dissertation laid the groundwork for the field of network
psychometrics — estimating network models from psychological datasets and
equating these to established psychometric modeling techniques. He has
implemented these methods in several widely used open-source software packages,
including qgraph, bootnet, and
psychonetrics, and received the dissertation prize of the
Psychometric Society.