Publications
The success of responsive science teaching to elicit, foreground, and pursue the substance of students' ideas depends on how teachers can recognize and respond to these ideas and connect them to disciplinary perspectives. One approach to supporting responsive teaching is to design instruction that facilitates students' scientific sensemaking and invites teachers to consider possible student ideas. Generating students' ideas, however, is labor-intensive as it requires understanding of learning content and instructional contexts. In this work, we explore the capacity of several large language models (LLMs), including OpenAI's GPT-5-mini, GPT-4o, Claude's Sonnet 4, Gemini 2.5 Flash, Mistral 7B, and Llama-4-17B, to simulate 8820 students' ideas in science learning activities across domains (life sciences, physics, and chemistry) and grade levels (elementary, middle, and high school) to support instruction planning. Findings indicate that the LLM-simulated responses are realistic and are mostly at the target grade levels for knowledge scope and readability. We observe variations in performance across LLMs and grade levels, with LLMs producing more within-scope ideas for high school lessons and overly complex ideas for lower grades. Interviews with six teachers using the LLM-generated ideas with their own lesson plans reveal that the simulated student responses are overall realistic and align with the lessons' objectives. Responses are most useful when they spark teachers' sensemaking about what students know and suggest relevant instructional scaffolds. We discuss how to improve LLMs' simulations of student thinking and promote responsive science teaching with AI-generated insights.
An important goal of science instruction is to promote students' understanding and evaluation of social issues related to science. Our study used the socioscientific issues (SSI) context of water conservation to explore evidence evaluation and certainty of assumptions among undergraduate students (n = 217) enrolled in a science literacy course at a research-intensive university. We investigated students' ability to use evidence and certainty of assumption about water conservation, as well as the predictability of these traits based on personal characteristics, including personal epistemic beliefs, emotion, open-mindedness, metacognition, and stakeholder identity. We also studied students' ability to provide reasons for their certainty of assumptions. We found that most students were fairly certain in their assumptions and identified themselves as strong stakeholders in water conservation. However, most students could not provide reasons for their assumptions. The application of regression models indicated that students' stakeholder identity, open-mindedness, and regulation of cognition were significant predictors of their certainty of assumption, while emotions, epistemic beliefs, and knowledge of cognition were non-significant predictors. We also found that students' evidence use was predicted only by their knowledge of cognition, while other personal characteristics were non-predictors. Pedagogical implications and future research avenues are discussed in the papers.
Supporting meaningful student engagement is an important goal of educators and educational researchers. However, there are many variations in definitions and measures of engagement, making it challenging to recognize and design for deep engagement. To address this challenge, the purpose of this work is to interrogate the ways that different theoretical and methodological choices impact how we describe student engagement. Methodology We applied two student engagement coding schemes to classroom video from two computer- supported collaborative learning (CSCL) contexts. We compare how each coding scheme made different types of engagement more or less visible and explore potential decision points and synergies. Findings The coding schemes differed in many ways, including their theoretical grounding, level of analysis, and inclusion of various dimensions of engagement. We argue that each of these methodological choices orient researchers to video data differently, which has implications for how researchers make sense of group work, next analytical steps, and claims. There are also substantial challenges that researchers may encounter when using engagement coding schemes outside of the contexts in which they were developed, including achieving reliability and the relevance of indicators of engagement to the novel context. Discussion This study contributes to our understanding of how to make sense of engagement in classroom video and the synergies between different coding choices and theoretical perspectives.
This study analyzed the relation between self-efficacy (SE) and sense of belonging (BL) in introductory physics courses by measuring these variables at two time points, early in the semester and midsemester. The study was performed in the introductory, calculus-based mechanics class at a large university in the United States and includes 1373 complete records. An extensive set of Observed Prior Causes (OPCs), including Force and Motion Conceptual Evaluation pretest scores, college and high school preparation variables, high school physics and math variables, personality variables, and demographics, were found to be predictive of SE or BL. Early in the semester, SE and BL were highly correlated ( r = 0.56 ); after controlling for OPCs, r = 0.37 of this correlation remained. SE fully explained the relation of BL to midsemester test average, while OPCs fully explained the relation of SE to test average. Changes to SE and BL in response to course feedback in the form of test averages were also explored. Test average explained little of the relation between early and midsemester SE or BL. The analysis supported a model where test average predicted midsemester SE, which in turn predicted midsemester BL. Mediation analysis found both directed and correlational paths were important in explaining the complex relation of these variables. Some moderation effects were detected, but were generally small.
Black and Hispanic faculty - underrepresented minorities (URMs) within academia - face career barriers that come to a crux in promotion and tenure decisions. Leveraging a natural experiment in choice architecture within a dataset of 1804 promotion and tenure decisions across six universities, we find that joint (906 faculty) vs. separate (898 faculty) evaluation reduces racial disparities in faculty outcomes. Specifically, in joint evaluation, an analysis of the simple slopes finds that Black and Hispanic faculty receive, on average, 9% fewer negative votes at the department level than in separate evaluations when controlling for research productivity, school, gender, rank, discipline, department size, and grant acquisition. Using moderated mediation analyses, we calculate that this translates into a 16.2% increase in the likelihood of a Black/Hispanic faculty member receiving a promotion. In a survey of 289 professors who have served on promotion and tenure committees (i.e., the key P&T decision-makers), we find that only 17% of faculty expect joint evaluation to improve underrepresented minority faculty outcomes and, conversely, 43% expect separate evaluation to improve underrepresented minority faculty outcomes. This natural experiment suggests that altering evaluation mode or simulating joint evaluation mode could help address academia's underrepresentation problem, but not in the way decision-makers expect.
Empowering Data Literacy: Fostering Self-Regulation and Adaptive Agency Through Personal Informatics
Data literacy vital for workforce and everyday life. Personal Informatics (PI): Analysis and collection ofpersonal data. Teaching data science with personal data = students encouraged and empowered to gain competency. Reflecting on habits supports well-being improvement.
Emotions play a central role in shaping learning within digital environments. Although their effects may depend on how students' emotional experiences manifest into concrete behaviors, the links between these dimensions remain underexplored. This study investigates the most common behaviors during episodes of boredom, confusion, frustration, and engaged concentration in an educational game, as well as associations with situational interest, self-efficacy, prior knowledge, and learning gains, using interaction logs and sensor-free affect detectors. Results show that boredom is linked to off-task roaming, both consistently associated with lower motivation and learning. In contrast, behaviors during engaged concentration, frustration, and especially confusion vary widely, shaped by motivational traits and prior knowledge and offering diverse associations with learning. Concrete regulatory responses in these states—such as systematizing findings with in-game tools, skimming domain content to resolve doubts, or testing hypotheses—are positively associated with learning and motivation, reflecting students' ability to regulate emotions and address cognitive challenges. However, less constructive responses, such as aimless wandering, were tied to lower knowledge and motivation, underscoring the need for additional support. These findings extend existing affective theory by underscoring the importance of considering the behavioral dimension when analyzing students' emotions in digital learning environments.
In this article, we review the literature highlighting early science, technology, engineering, and math (STEM) learning opportunities in U.S. Latine family contexts. We describe and synthesize research on Latine families' STEM-related beliefs and practices (i.e., STEM activities and STEM talk), as well as patterns related to the research areas, methodologies, and theoretical foundations among STEM learning studies focused on U.S. Latine families with young children (0–8 years old). Studies emphasize family interactions and strengths-based approaches to illustrate how Latine children are learning about STEM within the context of their families. This work positions Latine caregivers as leaders in promoting early STEM learning and partners in addressing gaps in Latine participation in STEM. We argue that more research is needed to show the wealth of STEM-related knowledge and skills that can be built on to promote meaningful and culturally sustaining learning experiences for children. This paper makes an important contribution to the literature by centering the experiential knowledge of Latine families, who are underrepresented due to the culture of power in STEM education.
Children possess remarkable proportional quantity skills; yet, their proportional estimations are often misguided by absolute quantities. Recent efforts have aimed to both quantify individual differences and identify possible erroneous strategies in proportional reasoning. We examined—across four time points over two years—the strategies pre-K-to-third-grade students used to estimate visually presented proportional quantities on a line and how their initial strategy related to their proportional reasoning growth and math ability. Results showed that individual differences are captured by strategy profiles that vary in the attended proportional or absolute quantity. Children who initially focused on proportional information maintained this strategy throughout the study and had stronger math ability than children who focused on one or two absolute quantities. Importantly, the absolute quantities children focused on were distinctly related to their estimation growth and math ability. These findings underscore the importance of considering quantitative and qualitative descriptions of children's proportional reasoning skills.


