**UPDATE 2024-04-15: Version 0.1-5**
Unfortunately, the last update contained some console debugging messages, which have been eliminated now. We used the opportunity to fix a bug and add two features:
- FIX: Debug messages removed
- FIX: A color bug associated with `dsel=`
- NEW: breaks option for the person parameter histogram (following the default behavior of the generic `hist()` function)
- NEW: Another nice default color scheme (`colorful1`)
**UPDATE 2024-04-06: Version 0.1-4**
This is mainly a bugfix version with a few new features:
- NEW: Select specific categories to draw (option `csel=`)
- NEW: Highlight categories (option `highlight=`)
- NEW: Return object contains thetas actually used for drawing (`$theta`)
Package Description
-------------------
RMX is an R package drawing an extended Person-Item Map termed **PIccc**. It provides the function plotPIccc(), which processes the return objects of **eRm**, **psychotools**, **ltm**, **mirt**, and **TAM** and supports currently the following uni- and multidimensional models (to the extent supported by the respective package):
|Model| eRm | ltm | TAM | mirt |psychotools|
|--- |:---:|:---:|:---:|:----:|:---:|
|RM | :) | :) | :) | :) | :) |
|2PL | | :) | :) | :) | :) |
|3PL | | :) | :) | :) | :) |
|4PL | | | | :) | :) |
|PCM | :) | :) | :) | :) | :) |
|RSM | :) | | :) | :) | :) |
|GPCM | | :) | :) | :) | :) |
|GRM | | :) | | :) | |
|GRSM | | | | :) | |
|NRM | | | | :) | |
In contrast to the "classical" PI-Map, the PIccc draws not only the item location parameters, but several functions: the CCC (*hence the name*), the TCC, the Category, Item, and Test Information Functions, and the standard error curve. Moreover, it shows the item category frequencies barcharts. Many graphical options allow for fine-tuning the diagram (see the gallery below).
The current version is 0.1-3.
For details, pls. refer to:
**Kabic, M. & Alexandrowicz, R.W. (2023). \
RMX/PIccc: An Extended Person--Item Map and a Unified IRT Output for eRm, psychotools, ltm, mirt, and TAM. \
*psych*, 5 (3), 948--965, \
DOI: https://doi.org/10.3390/psych5030062**
[**---> Article RMX/PIccc**][1]
The article is part of the special issue "Computational Aspects and Software in Psychometrics II" edited by Alexander Robitzsch:
https://www.mdpi.com/journal/psych/special_issues/Computational_Psychometrics
Here an example of a PIccc:
![An example of a PIccc:][2]
The following link contains a zip-packed html file with many worked examples, each showing the code for all supported packages:
[**---> RMX plotPIccc Gallery**][3]
Download and unzip the file to open it in your favorite browse. On top, you will find a table of contents linking to the various sections, each demonstrating an option. Moreover, you will find a set of tabs for each diagram showing how to produce it using eRm, psychotools, ltm, mirt, or TAM. We hope, this little gallery inspires you for informative diagrams for you analyses!
Another interesting graphical option for Rasch model analysis might be the GMX package, which provides extended graphical model checks for the RM, the PCM, and the RSM (supporting the packages eRm and psychotools), see
- [GMX package (here on osf.io/2ryd8)][4]
- [GMX article 1 (PTAM)][5]
- [GMX article 2 (PTAM)][6]
In case of further questions, remarks, or (perish the thought!) bug report, pls. contact rainer.alexandrowicz [at] aau [dot] at.
[1]: https://www.mdpi.com/2624-8611/5/3/62
[2]: https://osf.io/download/dgfy3/
[3]: https://osf.io/download/yh8b5/
[4]: https://osf.io/2ryd8/
[5]: http://
[6]: https://www.psychologie-aktuell.com/fileadmin/Redaktion/Journale/ptam_2023-2/PTAM__2-2023_4_kor.pdf
