ETS scientists developed a human-centered AI (HAI) framework that combines data on how students interact with assessments--such as task navigation and time spent--with their performance, providing deeper insights into student performance in large-scale assessments.

The rise of the digital era has led to a decline in handwriting as the primary mode of communication, resulting in negative effects on handwriting literacy, particularly in complex writing systems such as Chinese. The marginalization of handwriting has contributed to the deterioration of penmanship, defined as the ability to write aesthetically…

This study explores Hong Kong higher education students' perceptions of E-assessment systems, focusing on factors shaping acceptance of E-examinations over traditional formats. Quantitative analysis of 107 respondents reveals significant positive correlations between diverse pre-exam guidance (e.g., tutorials) and key system features (e.g.,…

This systematic review synthesizes 14 peer-reviewed studies from 2015 to 2023, focusing on the assessment methods and delivery of digital microcredentials. Microcredentials provide specialized, focused content and recognize professional learning or competency in specific skills. This paper defines digital microcredentials as those offered in an…

Example-generation tasks have been suggested as an effective way to both promote students' learning of mathematics and assess students' understanding of concepts. E-assessment offers the potential to use example-generation tasks with large groups of students, but there has been little research on this approach so far. Across two studies, we…

In computerized adaptive testing, item exposure control methods are often used to provide a more balanced usage of the item pool. Many of the most popular methods, including the restricted method (Revuelta and Ponsoda), use a single maximum exposure rate to limit the proportion of times that each item is administered. However, Barrada et al.…

This study compares how the generative AI (GenAI) large language model (LLM) ChatGPT performs in grading university exams compared to human teachers. Aspects investigated include consistency, large discrepancies and length of answer. Implications for higher education, including the role of teachers and ethics, are also discussed. Three…

Driven by technological innovations, continuous digital expansion has transformed fundamentally the landscape of modern higher education, leading to discussions about evaluation techniques. The emergence of generative artificial intelligence raises questions about reliability and academic honesty regarding multiple-choice assessments in online…

Assessment and institutional research offices have too much data and too little time. Standard reporting often crowds out opportunities for innovative research. Fortunately, advancements in data science now offer a clear solution. It is equal parts technique and philosophy. The first and easiest step is to modernize data work. This column…

Teachers, examiners and assessment experts know from experience that some candidates annotate exam questions. "Annotation" includes anything the candidate writes or draws outside of the designated response space, such as underlining, jotting, circling, sketching and calculating. Annotations are of interest because they may evidence…

Arithmetic fluency is regarded as a foundational math skill, typically measured as a single construct with pencil-and-paper-based timed assessments. We introduce a tablet-based assessment of single-digit fluency that captures individual trial response times across several embedded experimental contrasts of interest. A large (n = 824) cohort of…

In recent years, the application of explainability techniques to automated essay scoring and automated short-answer grading (ASAG) models, particularly those based on transformer architectures, has gained significant attention. However, the reliability and consistency of these techniques remain underexplored. This study systematically investigates…

Overview: This systematic review explores the utilization of artificial intelligence (AI) for assessment, grading, and feedback in higher education. The review aims to establish how AI technologies enhance efficiency, scalability, and personalized learning experiences in educational settings, while addressing associated challenges that arise due…

The purpose of this study is to describe and evaluate a multilingual automated essay scoring (AES) system for grading essays in three languages. Two different sentence embedding models were evaluated within the AES system, multilingual BERT (mBERT) and language-agnostic BERT sentence embedding (LaBSE). German, Italian, and Czech essays were…

Item compromise has long posed challenges in educational measurement, jeopardizing both test validity and test security of continuous tests. Detecting compromised items is therefore crucial to address this concern. The present literature on compromised item detection reveals two notable gaps: First, the majority of existing methods are based upon…