edstan-package {edstan}

Stan for item response theory

Description

edstan attempts to make easy the fitting of standard item response theory models using rstan.

Details

A user will generally want to use the following functions (in order) to fit a model:

1. irt_data to format the data,

2. irt_stan to fit a model, and

3. print_irt_stan to view some results.

Additionally, labelled_integer is some times helpful for data formatting and stan_columns_plot creates a plots of convergence and other statistics by parameter vector. The package also includes six Stan models (see irt_stan for a list) and two example datasets (aggression and spelling).

It is expected that once a user is comfortable fitting pre-defined edstan models, they will write their own Stan models and fit them with stan, for which irt_stan is a wrapper.

See Also

Case studies for each of the edstan models have been published.

Rasch and two-parameter logistic models

http://mc-stan.org/documentation/case-studies/rasch_and_2pl.html

(Generalized) partial credit model

http://mc-stan.org/documentation/case-studies/pcm_and_gpcm.html

(Generalized) rating scale model

http://mc-stan.org/documentation/case-studies/rsm_and_grsm.html