For different steps in the assessment cycle the TBA Centre provides R packages, including shinyReCoR for scoring short text responses using Natural Langue Processing (NLP), LogFSM for analyzing log data using finite-state machines and ShinyItemBuilder for using CBA ItemBuilder items in assessment based on R.

Scoring of Short Text Responses

The R-package shinyReCoR provides a Shiny application for automatically coding text responses using R. The app is specialized in short text responses from assessments such as PISA, but can be employed in other contexts as well. The app’s complete version will allow you to construct your own text corpora from Wikipedia, compute tailored semantic spaces, apply automatic spelling correction, adapt the workflow to your needs by switching to R at any time, export the classifier so that it can be used as a service during assessment, choose other test languages, and much, much more.

* Contact: Dr. Fabian Zehner
* Development: Nico Andersen & Dr. Fabian Zehner
* Further Links: ReCo project; https://www.reco.science/

Analyses of Log Data

The R-package LogFSM illustrates the use of finite-state machines to extract low-level features from log data, emphasizing reproducibility and validity for log data analyses (i.e., the construction of process indicators). The tool can be used to analyze log data in standard formats (such as the XES-standard) and provides methods to convert data gathered with different software (e.g., with the IRTlib Deployment Software and CBA ItemBuilder tasks) into standard formats.

* Contact & Development: PD Dr. Ulf Kröhne
* Further Links: https://www.logfsm.com

Computer-based Assessments with R (and Shiny)

The R-package ShinyItemBuilder allows to use assessment content created with the CBA ItemBuilder to be used, together with R in Shiny applications. This allows to run various types of innovative assessments. ShinyItemBuilder is well-suited for many small-scale applications (with a limited number of concurrent test-takers) and allows using easily accessible hosting environments (such as shinyapps.io) for online testing. Feedback can be provided to test-takers using markdown and R, and IRT-based test assembly is supported using one of the available packages for computerized adaptive testing (CAT) and multi-stage testing (MST) in R.

* Contact: PD Dr. Ulf Kröhne
* Development: Ulf Kröhne & Felix Wagner
* Further Links: https://github.com/kroehne/ShinyItemBuilder

Other Tools