studysim: generate simulated medical studies for jamovi

A jamovi module for teaching epidemiology and clinical research. Pick a clinical scenario, an outcome, a study design and a sample size, and get a plausible dataset in a new jamovi window. A hidden answer key gives the true parameters and what the analyses should find.

• Tutorial: https://victor-moreno.github.io/jamovi-studysim/
• Scenario catalogue: https://victor-moreno.github.io/jamovi-studysim/scenarios.html (65 scenarios, most replicating a real study)

Analyses (menu: Simulate)

• Simulated study: one dataset from a scenario, outcome and design (observational, trials, diagnostic), with optional realism and biases, and its answer key.
• Compare designs: one scenario under several designs, estimates against the truth.
• Library: lists, checks, explores and exports scenarios.
• Derive scenario: estimates a new scenario from a real dataset.

Install

Download the .jmo for your computer from the Releases page (x64 for Intel/AMD, arm64 for Apple silicon) and in jamovi choose Modules (⊕) → Sideload. Requires jamovi 28.4 or newer; no other packages.

Developers

Rscript tools/build-library-js.R      # after editing inst/library
bash tools/build-jmo.sh current       # dist/*.jmo
bash tools/build-ai-skill.sh          # dist/studysim-scenario-skill.zip
Rscript -e 'pkgload::load_all("."); testthat::test_dir("tests/testthat")'

The engine (R/engine_*.R, R/derive.R) is plain R, independent of jamovi. tests/ui/*.py and tutorial/screenshots/capture.py drive jamovi in Docker with the jamovi-automation helpers; the tutorial is a Quarto site in tutorial/, rendered to docs/.

Acknowledgements

Designed and built together with Claude Code, over several rounds of design, implementation and testing in jamovi. Ideas come from simstudy, simDAG, PRIME-CVD and The Islands; no code was copied. Survival models use the R package survival by Terry Therneau; mixed models in the answer key use nlme (Pinheiro, Bates and the R Core Team). Scenario parameters were estimated from datasets distributed with the R packages survival (Therneau), MASS (Venables and Ripley), datasets (R Core Team), HSAUR3 (Everitt and Hothorn), TH.data (Hothorn), pROC (Robin et al.), medicaldata (Higgins), aplore3 (Braglia) and NHANES (Pruim); we thank the original investigators, cited in each scenario.

This module has been developed with support of the Instituto de Salud Carlos III (ISCIII), “Programa FORTALECE del Ministerio de Ciencia e Innovación”, through the project number FORT23/00032 and the Consortium for Biomedical Research in Epidemiology and Public Health (CIBERESP), action Genrisk.

Licence

GPL (>= 2).