The University of North Carolina at Greensboro

School of Education - Educational Research Methodology

ERM Software

ERM faculty members have made the following software packages available free for download. This software is offered “as is” with no guarantees. If you have any comments or questions about any software on this page, contact the author of that specific package.

Packages for PC:

CITAN: Classical Item Analyzer
Computes test scale reliability analyses, classical item analysis, and distractor analysis. Routines have been checked against other software (e.g., SPSS, Systat). Flexible application to many types of selected-response items. User’s manual and sample files included.
Download   Manual

DIFAS: Differential Item Functioning Analysis System
Windows-based program that computes odds ratio estimates of differential item functioning, differential test functioning, and differential step functioning effects, along with associated tests of significance. This includes the Mantel-Haenszel common log-odds ratio, the Breslow-Day test of trend in odds ratio heterogenity, and Liu-Agresti cumulative common log-odds ratio, and the relevant generalizations to step-level analyses for polytomous items.
Download  Manual

IRT-Lab: Software for Teaching and Research in Item Response Theory
Windows-based program that produces graphical representations of item response functions (item characterisitic curves) for dichotomous and polytomous items, item and test information functions, likelihood functions, and generates simulated responses to dichotomous and polytomous IRT models.
Download  Manual

Packages for R (Do you know about R? Click here for details.):

CTT: Classical Test Theory Functions
This package can be used to perform a variety of tasks and analyses associated with classical test theory (CTT): score multiple-choice responses, perform reliability analyses, conduct item analyses, and transform scores onto different scales.
Download   Manual

mixRasch: Mixture Rasch Models with JMLE
The included function will estimate a mixture Rasch model using joint maximum likelihood estimation (JMLE). The estimation is based on a mixture partial credit model. Step parameters can be constrained to estimate a mixture rating scale model. Estimating a model with only one latent class accomplishes a standard Rasch analysis with JMLE.

Connect with the School of Education!