Multilevel Rasch modeling: Does misfit to the Rasch model impact the regression model?
Publication Date
2020
Document Type
Article
Abstract
Multilevel Rasch models are increasingly used to estimate the relationships between test scores and student and school factors. Estimating the item parameters and the regression parameters in a single model should yield more accurate estimates of the coefficients, standard errors, and proportion of variance at each level than estimating the students' scores in the first step and then treating these scores as fixed. Multilevel Rasch models may be preferred to multilevel IRT models due to simplicity and broader software options. In this simulation study, response data were generated to follow one, two, and three parameter logistic (1PL, 2PL, 3PL) models, but the Rasch model was used to estimate the latent regression parameters. When the response functions followed 2PL or 3PL models, the proportion of variance explained in test scores by the simulated student or school predictors was estimated accurately with a Rasch model. Proportion of variance within and between schools was also estimated accurately. The regression coefficients were misestimated unless they were rescaled out of logit units. However, item-level parameters, such as DIF effects, were biased when the Rasch model was violated, similar to single-level models.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
DeMars, C. E. (2020). Multilevel Rasch modeling: Does misfit to the Rasch model impact the regression model? Journal of Experimental Education, 88, 605-619. https://doi.org/10.1080/00220973.2019.1610859
