Abstract
Lim, H., & Wells, C. S. (2020). irtplay: An R package for online item calibration, scoring, evaluation of model fit, and useful functions for unidimensional IRT. Applied Psychological Measurement, 44(7–8), 563–565. https://doi.org/10.1177/0146621620921247
The R package, irtplay, has been removed from the Comprehensive R Archive Network (CRAN) repository due to issues related to intellectual property (IP) violations as of September 13, 2022.
In the light of this removal, we introduce irtQ (Lim & Wells, 2023), a newly refined R package now available on the CRAN. The irtQ has been designed as a successor to the irtplay package and meticulously developed to ensure fully compliance with industry IP standards. For those who relied on the irtplay, the irtQ can serve as its direct replacement.
While the irtQ retains much of the core functionality of irtplay, there are certain notable changes: 1. Functionality Differences: • For the est_score function, users should now provide the character strings ‘ML’ and ‘MLF’ as method arguments to implement the maximum likelihood estimation and maximum likelihood estimation with fences, respectively. This replaces the earlier ‘MLE’ and ‘MLEF’. • The shap_df function has revised argument names. Specifically, ‘par.dc’, ‘par.py’, and ‘empty.par’ are now ‘par.drm’, ‘par.prm’, and ‘default.par’, respectively. • The plm function has been renamed to prm, with its argument ‘pmodel’ now renamed to ‘pr.model’. 2. Optimizations: A number of functions in the irtQ (e.g. est_irt, est_score, lwrc) have been optimized to offer more efficient and quicker computations compared to their counterparts in the irtplay.
We sincerely apologize for any confusion or inconvenience caused by the removal of irtplay. We deeply value your understanding and continued trust.