Search
Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Room
Browse By Unit
Browse By Session Type
Help
About Vancouver
Personal Schedule
Sign In
The purpose of this study is to provide a valid approach and evidence of scale comparability for accommodated tests by using differential person functioning (DPF) as types of modeling-data misfit (person-fit). Although differential item functioning (DIF) is often investigated to rule out test unfairness as an explanation for the group difference, the unequal sample size and the size of samples have a significant effect on the statistical power of DIF detection procedure. One of the advantages of person-fit analysis is that even with a small sample sizes, it can be applied to investigate the scalability of examinees within different subgroups. This study demonstrates that person-fit analysis is appropriate for scale comparability study regardless of the size of the subgroups.
Dong Gi Seo, Michigan Department of Education
Shiqi Hao, Michigan Department of Education
Steven Guy Viger, Michigan State University