Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.

Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

An alternative examinee-centered standard setting strategy

Patrick M Irwin, University of Nebraska - Lincoln

Abstract

This study examined an alternative examinee-centered standard setting method that is a modification of the Contrasting Groups method called the Modified Contrasting Groups method (MCGM). The alternative method was designed to be affordable, not too complex computationally, and done in a reasonable amount of time. The study investigated how well the alternative method produced results that are approximately equivalent to other standard setting methods. The results generated by the MCGM were compared to the results produced by the Angoff method, Contrasting Groups method, and the Borderline Group method. ^ Two variations of the alternative method (MCGM 1 and MCGM 2) were explored in the study. MCGM 1 and MCGM 2 are based on having teacher predictions of student performance and the students' actual performance or scores on an assessment. The MCGM 1 and MCGM 2 follow different but standard mathematical procedures that are easy to follow. The calculation for the MCGM 1 requires finding the median score for each performance category, then calculating the average of the two adjacent performance categories. MCGM 2 is based on combining the adjacent performance categories and calculating a median value of the scores in the two adjacent categories. ^ This study found that MCGM 1 and MCGM 2 produced cut scores higher than the Angoff and Contrasting Groups method on average, but lower than the Borderline Group method. The study also found the MCGM 1 and MCGM 2 consistently produced different cut scores using the same data sets. ^ The MCGM 1 and MCGM 2 appear to be viable standard setting methods for many school districts in Nebraska to use to meet state and federal reporting requirements. For some districts, either or both of these variations could be added to what is currently in place to provide a range of plausible cut scores. For other districts, this method could be adopted as the sole cut score method.^

Subject Area

Education, Tests and Measurements|Psychology, Psychometrics

Recommended Citation

Irwin, Patrick M, "An alternative examinee-centered standard setting strategy" (2007). ETD collection for University of Nebraska - Lincoln. AAI3245357.
http://digitalcommons.unl.edu/dissertations/AAI3245357

Share

COinS