We investigated methods to enhance the generalizability of large language models (LLMs) designed to classify dimensions of collaborative discourse during small group work. Our research utilized five diverse datasets that spanned various grade levels, demographic groups, collaboration settings, and curriculum units. We explored different model…

Collaborative dialogue offers rich insights into students' learning and critical thinking. This is essential for adapting pedagogical agents to students' learning and problem-solving skills in STEM+C settings. While large language models (LLMs) facilitate dynamic pedagogical interactions, potential hallucinations can undermine confidence, trust,…

Educators and researchers are interested in ways that ChatGPT and other generative AI tools might move beyond the role of "cheatbot" and become part of the network of resources students use for writing. We studied how high school students used ChatGPT as a writing support while writing arguments about topics like school mascots. We…

The development of large language models offers new possibilities for enhancing adaptive scaffolding of student learning in game-based learning environments. In this work, we present a novel framework for automatic plan generation that utilizes text-based representations of students' actions within a game-based learning environment, Crystal…

Recent advances in AI have opened the door for the automated scoring of open-ended math problems, which were previously much more difficult to assess at scale. However, we know that biases still remain in some of these algorithms. For example, recent research on the automated scoring of student essays has shown that certain varieties of English…

Short answer assessment is a vital component of science education, allowing evaluation of students' complex three-dimensional understanding. Large language models (LLMs) that possess human-like ability in linguistic tasks are increasingly popular in assisting human graders to reduce their workload. However, LLMs' limitations in domain knowledge…

The PISA assessment 2022 of creative thinking was a moonshot effort that introduced significant advancements over existing creativity tests, including a broad range of domains (written, visual, social, and scientific), implementation in many languages, and sophisticated scoring methods. PISA 2022 demonstrated the general feasibility of assessing…

We report four experiments investigating the effect of prosody on listening comprehension in 11-13-year-old children. Across all experiments, participants listened to short object descriptions and answered content-based questions about said objects. In Experiments 1-3, the descriptions were read in an emotionally positive or neutral tone of voice.…

Letters of recommendation from school counselors are required to apply to most selective universities. We use cutting-edge natural language processing techniques to algorithmically analyze a national dataset of over 600,000 student applications and counselor recommendation letters submitted through the Common Application. We examine how the length…

There is growing interest in the potential of Artificial Intelligence (AI) to assist in various educational tasks, including writing assessment. However, the comparative efficacy of human and AI-powered systems in this domain remains a subject of ongoing exploration. This study aimed to compare the accuracy of human raters (teachers and…

Long before ChatGPT, it was an open secret that students did not always read the books they were assigned in their English Language Arts (ELA) classes, relying instead on online study guides like SparkNotes. Via a retrospective survey, our exploratory study examined (1) the rate of SparkNotes use among high-school ELA students; (2) why students…

As intelligent textbooks become more ubiquitous in classrooms and educational settings, the need to make them more interactive arises. An alternative is to ask students to generate knowledge in response to textbook content and provide feedback about the produced knowledge. This study develops Natural Language Processing models to automatically…

Large language models (LLMs) have the potential to impact learning in the advanced learner virtual classroom through personalized learning and on-demand feedback. This study investigated whether an LLM added to virtual science course activities impacted student learning. Using the GPT4o model from OpenAI, the LLM was developed as a co-tutor to…

Machine and prompt-based Artificial Intelligence (AI) learning has made significant evolution profusely. In education, it has revitalized researchers and educators to scout out subsequent advantages for optimizing learning results. Chiefly, Generative AI has exhibited substantial potential as a tool for language augmentation. This study aims to…

Education is an important way to achieve global Sustainable Development Goals (SDGs), while classroom engagement and collective efficacy are key factors that influence SDG learning outcomes. However, students' in-depth thinking could be limited when they apply to search engines such as Google to support their learning of SDG-related topics. Large…