The pervasive issue of cheating in educational tests has emerged as a paramount concern within the realm of education, prompting scholars to explore diverse methodologies for identifying potential transgressors. While machine learning models have been extensively investigated for this purpose, the untapped potential of TabNet, an intricate deep…

Benefiting from item preknowledge (e.g., McLeod, Lewis, & Thissen, 2003) is a major type of fraudulent behavior during educational assessments. This paper suggests a new statistic that can be used for detecting the examinees who may have benefitted from item preknowledge using their response times. The statistic quantifies the difference in…

The probability of test takers changing answers upon review of their initial choices is modeled. The primary purpose of the model is to check erasures on answer sheets recorded by an optical scanner for numbers and patterns that may be indicative of irregular behavior, such as teachers or school administrators changing answer sheets after their…

David Foster is the founder of Kryterion, an Internet test administration company, and currently serves there as chief scientist and executive vice president. He is the author of numerous articles for industry trade journals and textbooks and sits on the Council for the International Test Commission. In this interview, Foster talks about his…

The use of value-added modeling (VAM) in school accountability is expanding, but deciding how to embrace VAM is difficult. Various experts say it's too unreliable, causes more harm than good, and has a big margin for error. Others assert VAM is imperfect but useful, and provides valuable feedback. A closer look at the models, and their use,…

This dissertation proposes a new factor-analytic technique for detecting cheating on exams. Person-fit statistics have been developed to assess the extent to which examinees' response patterns are consistent with expectation, with expectation defined in the context of some model. Response patterns that are inconsistent with expectation are said to…

Computer-based assessments are becoming more commonplace, perhaps as a necessity for faculty to cope with large class sizes. These tests often occur in large computer testing venues in which test security may be compromised. In an attempt to limit the likelihood of cheating in such venues, randomised presentation of items is automatically…

Careful management of a testing program can greatly limit a teacher's opportunity to cheat while administering standardized tests to students and can increase the likelihood that such cheating will be detected. The Austin Independent School District's systemwide testing staff's plan for controlling cheating has three basic premises: (1) plan and…

Sound testing practices and the high-quality information that can result are helpful to those who have oversight, responsibility, or interest in American education. To the extent that tests provide high-quality information, they form the basis for making accurate judgments about individual students. It is equally true, however, that factors which…

Since recent rises in standardized test scores have once again raised the possibility of cheating, there are a number of things which a school district should do to minimize the problem. These involve: (1) informing staff, students, and parents about appropriate and inappropriate test-taking and test-improvement skills; (2) providing staff and…

Due to a concern regarding the validity of Chicago public elementary schools standardized test scores on the Iowa Tests of Basic Skills, a test audit was conducted. The audit was limited to 40 schools. Two schools were selected from each of the 20 administrative districts, with each school testing one seventh- and one eighth-grade class. These…

Discusses student cheating on examinations and research/term paper assignments. Offers methods and techniques to prevent or minimize cheating. (SR)

In the context of reviewing an article for this journal (van der Linden & Sotaridona, this issue, pp. 283-304) the topic of unconditional and conditional hypothesis testing came under consideration. While this is hardly a new issue (consider, for example, arguments regarding the chi square vs. Fisher exact test of independence for a 2 x 2…

A statistical test for detecting answer copying on multiple-choice tests based on Cohen's kappa is proposed. The test is free of any assumptions on the response processes of the examinees suspected of copying and having served as the source, except for the usual assumption that these processes are probabilistic. Because the asymptotic null and…