Investigating the Effect of Differential Rapid Guessing on Population Invariance in Equating.

ABSTRACT: Score equating is an essential tool in improving the fairness of test score interpretations when employing multiple test forms. To ensure that the equating functions used to connect scores from one form to another are valid, they must be invariant across different populations of examinees. Given that equating is used in many low-stakes testing programs, examinees' test-taking effort should be considered carefully when evaluating population invariance in equating, particularly as the occurrence of rapid guessing (RG) has been found to differ across subgroups. To this end, the current study investigated whether differential RG rates between subgroups can lead to incorrect inferences concerning population invariance in test equating. A simulation was built to generate data for two examinee su

SUBMITTER: Deng J 

PROVIDER: S-EPMC9483216 | biostudies-literature | 2022 Oct

REPOSITORIES: biostudies-literature