Skip to contents

easyRaschBayes

easyRaschBayes provides functions to reproduce classic Rasch analysis features using Bayesian item response theory (IRT) models fitted with brms. It supports both dichotomous Rasch models and polytomous partial credit models, and exposes the full posterior distribution for all output.

For more materials on Rasch analysis, see the vignette for my (frequentist) package easyRasch.

Installation

Install the stable version from CRAN:

install.packages("easyRaschBayes")

Or install the development version from GitHub:

install.packages("remotes") # if you don't have `remotes` installed
remotes::install_github("pgmj/easyRaschBayes")

Main functions

Function Description
infit_statistic() Conditional infit / outfit statistics
item_restscore_statistic() Item–rest score associations with Goodman & Kruskal’s gamma
plot_residual_pca() Residual PCA contrast plot for dimensionality assessment
q3_statistic() Yen’s Q3 residual correlations for local dependence evaluation
plot_ipf() Item category probability function curves
plot_targeting() Person-item map (Wright map)
dif_statistic() Differential Item Functioning (DIF) analysis
fit_statistic_pcm() Posterior predictive item fit for polytomous models
fit_statistic_rm() Posterior predictive item fit for dichotomous models
RMUreliability() Reliability via Relative Measurement Uncertainty (RMU)

Usage 

library(brms)
library(easyRaschBayes)

# Fit a Bayesian Rasch model (dichotomous)
fit <- brm(
  response ~ 1 + (1 | item) + (1 | id),
  data   = my_data,
  family = bernoulli(),
  chains = 4, cores = 4
)

# Conditional item infit
infit <- infit_statistic(fit)
infit_post(infit)

# Local dependence with Yen's Q3
q3_results <- q3_statistic(fit)
q3_post(q3_results)

# Response category functioning
plot_ipf(fit)

# Person-item map
plot_targeting(fit)

References

Bürkner, P.-C. (2020). Analysing Standard Progressive Matrices (SPM-LS) with Bayesian Item Response Models. Journal of Intelligence, 8(1). doi:10.3390/jintelligence8010005

Bürkner, P.-C. (2021). Bayesian Item Response Modeling in R with brms and Stan. Journal of Statistical Software, 100, 1–54. https://doi.org/10.18637/jss.v100.i05

Credits

This started as a side project to update the code to assess item fit included in Bürkner (2020). With the help of Claude Opus 4.6 things expanded to try to reproduce core Rasch analysis steps. Most of the code in this package is produced by the LLM.

Magnus Johansson is a licensed psychologist with a PhD in behavior analysis. He works as a research specialist at Karolinska Institutet, Department of Clinical Neuroscience, Center for Psychiatry Research.

License

GPL (>= 3)