Computer adaptive testing (CAT) is an increasingly common mode of test administration offering improved test security, better measurement precision, and the potential for shorter testing experiences. This article presents a new item selection algorithm based on a generalized objective function to support multiple types of testing conditions and…

Educational tests often have a cluster of items linked by a common stimulus ("testlet"). In such a design, the dependencies caused between items are called "testlet effects." In particular, the directional testlet effect (DTE) refers to a recursive influence whereby responses to earlier items can positively or negatively affect…

This study evaluates Large language models (LLMs)' performance on Chinese Postgraduate Medical Entrance Examination (CPGMEE) as well as the hallucinations produced by LLMs and investigate their implications for medical education. We curated 10 trials of mock CPGMEE to evaluate the performances of 4 LLMs (GPT-4.0, ChatGPT, QWen 2.1 and Ernie 4.0).…

Creating assessment questions aligned to standards is a time-consuming task for teachers, but large language models such as ChatGPT can help. Bryan Drost & Char Shryock describe a three-step process for using ChatGPT to create assessments: 1) Ask ChatGPT to break standards into measurable targets. 2) Determine how much time to spend on each…

In Spring 2020, the need for redesigning online assessments to preserve integrity became a priority to many educators. Many of us found methods to proctor examinations using Zoom and proctoring software. Such examinations pose their own issues. To reduce the technical difficulties and cost, many Zoom proctored examination sessions were shortened;…

This paper introduces AMMORE, a new dataset of 53,000 math open-response question-answer pairs from Rori, a mathematics learning platform used by middle and high school students in several African countries. Using this dataset, we conducted two experiments to evaluate the use of large language models (LLM) for grading particularly challenging…

Background: Multiple-choice item (MCI) assessments are burdensome for instructors to develop. Artificial intelligence (AI, e.g., ChatGPT) can streamline the process without sacrificing quality. The quality of AI-generated MCIs and human experts is comparable. However, whether the quality of AI-generated MCIs is equally good across various domain-…

This study aimed to reveal the differences in individuals' abilities, their standard errors, and the psychometric properties of the test according to the two methods of applying the test (electronic and paper). The descriptive approach was used to achieve the study's objectives. The study sample consisted of 74 male and female students at the…

VocQGen is an automated tool designed to generate multiple-choice cloze (MCC) questions for vocabulary assessment in second language learning contexts. It leverages several natural language processing (NLP) tools and OpenAI's GPT-4 model to produce MCC items quickly from user-specified word lists. To evaluate its effectiveness, we used the first…