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…

Errors in examination papers and other assessment instruments can compromise fairness. For example, a history question containing an incorrect historical date could be impossible for students to answer. Incorrect instructions at the start of an examination could lead students to answer the wrong number of questions. As there is little research on…

In educational and psychological measurement, a person-fit statistic (PFS) is designed to identify aberrant response patterns. For parametric PFSs, valid inference depends on several assumptions, one of which is that the item response theory (IRT) model is correctly specified. Previous studies have used empirical data sets to explore the effects…

Educational testing has been criticized for its disconnect from modern cognitive science and its limited role in improving instruction and student learning. Reform efforts emphasize the need for testing to provide specific diagnostic insights into students' skills and knowledge. Cognitive diagnosis (CD), an emerging paradigm in educational…

Response-time models are of increasing interest in educational and psychological testing. This article focuses on the lognormal model for response times, which is one of the most popular response-time models, and suggests a simple person-fit statistic for the model. The distribution of the statistic under the null hypothesis of no misfit is proved…

Response-time models are of increasing interest in educational and psychological testing. This paper focuses on the lognormal model for response times (van der Linden, 2006), which is one of the most popular response-time models, and suggests a simple person-fit statistic for the model. The distribution of the statistic under the null hypothesis…

Binary item response theory (IRT) models are widely used in educational testing data. These models are not perfect because they simplify the individual item responding process, ignore the differences among different response patterns, cannot handle multidimensionality that lay behind options within a single item, and cannot manage missing response…

A common interpretation of achievement test results is that they provide measures of achievement that are much like other measures we commonly use for height, weight, or the cost of goods. In a limited sense, such interpretations are correct, but some nuances of these interpretations have important implications for the use of achievement test…

It is demonstrated that the popular three-parameter logistic model can lead to markedly inaccurate individual ability level estimates for mixture populations. A theoretically and empirically important setting is initially considered where (a) in one of two subpopulations (latent classes) the two-parameter logistic model holds for each item in a…

Choosing whether to test via computer is the most difficult and consequential decision the designers of a testing program can make. The decision is difficult because of the wide range of choices available. Designers can choose where and how often the test is made available, how the test items look and function, how those items are combined into…

Lloyd Bond comments here on the Focus article in this issue of "Measurement: Interdisciplinary Research and Perspectives". The Focus article is entitled: "How Task Features Impact Evidence from Assessments Embedded in Simulations and Games" (Russell G. Almond, Yoon Jeon Kim, Gertrudes Velasquez, and Valerie J. Shute). Bond…

Testing is an issue of increasing importance. While for many teachers language learning should be communicative; in fact, they expect their students to provide evidence of their knowledge. Thus, there is a clear mismatch on the approach to language teaching and language testing. As a consequence, there is an evident need change the testing…

Educational and psychological constructs are normally measured by multifaceted dimensions. The measured construct is defined and measured by a set of related subdomains. A bifactor model can accurately describe such data with both the measured construct and the related subdomains. However, a limitation of the bifactor model is the orthogonality…

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…

This paper provides a history of ETS's role in developing assessment instruments and psychometric procedures for measuring change in large-scale national assessments funded by the Longitudinal Studies branch of the National Center for Education Statistics. It documents the innovations developed during more than 30 years of working with…