ABSTRACT
In high-stakes educational achievement testing, data forensics has risen as a key approach for supporting the integrity and interpretability of student test scores. Item answer-changing analysis, also called erasure analysis, has emerged as a method of flagging potential cheating behavior. Erasure analysis via scanned paper and pencil tests involves examining the instances of student answer-changing behavior on multiple choice tests, looking for aberrant or larger than expected numbers of answer changes or of wrong-to-right answer changes. Aberrant patterns of answer changes may be indicative of changes made to items by someone other than the student themselves (Qualls, 2001). While answer-changing has been studied in the context of paper-pencil tests for decades, little is known about the construct related to computer-based tests.
