dexter: An R Package to Manage and Analyze Test Data
Open Access
- 28 April 2023
- Vol. 5 (2), 350-375
- https://doi.org/10.3390/psych5020024
Abstract
In this study, we present a package for R that is intended as a professional tool for the management and analysis of data from educational tests and useful both in high-stakes assessment programs and survey research. Focused on psychometric models based on the sum score as the scoring rule and having sufficient statistics for their parameters, dexter fully exploits the many theoretical and practical advantages of this choice: lack of unnecessary assumptions, stable and fast estimation, and powerful and sensible diagnostic techniques. It includes an easy to use data management system tailored to the structure of test data and compatible with the current paradigm of tidy data. Companion packages currently include a graphical user interface and support for multi-stage testing.Keywords
This publication has 30 references indexed in Scilit:
- Conditional Statistical Inference with Multistage Testing DesignsPsychometrika, 2013
- Profile Analysis: A Closer Look at the PISA 2000 Reading DataScandinavian Journal of Educational Research, 2012
- An introduction to ROC analysisPattern Recognition Letters, 2005
- Randomization-based inference about latent variables from complex samplesPsychometrika, 1991
- Weighted likelihood estimation of ability in item response theoryPsychometrika, 1989
- Characterizing the manifest probabilities of latent trait modelsPsychometrika, 1983
- Adaptive EAP Estimation of Ability in a Microcomputer EnvironmentApplied Psychological Measurement, 1982
- A rasch model for partial credit scoringPsychometrika, 1982
- A goodness of fit test for the rasch modelPsychometrika, 1973
- Estimating item parameters and latent ability when responses are scored in two or more nominal categoriesPsychometrika, 1972