Joseph Krajcik’s research while affiliated with Michigan State University and other places

We discuss transforming STEM education using three aspects: learning progressions (LPs), constructed response performance assessments, and artificial intelligence (AI). Using LPs to inform instruction, curriculum, and assessment design helps foster students’ ability to apply content and practices to explain phenomena, which reflects deeper science understanding. To measure the progress along these LPs, performance assessments combining elements of disciplinary ideas, crosscutting concepts and practices are needed. However, these tasks are time-consuming and expensive to score and provide feedback for. Artificial intelligence (AI) allows to validate the LPs and evaluate performance assessments for many students quickly and efficiently. The evaluation provides a report describing student progress along LP and the supports needed to attain a higher LP level. We suggest using unsupervised, semi-supervised ML and generative AI (GAI) at early LP validation stages to identify relevant proficiency patterns and start building an LP. We further suggest employing supervised ML and GAI for developing targeted LP-aligned performance assessment for more accurate performance diagnosis at advanced LP validation stages. Finally, we discuss employing AI for designing automatic feedback systems for providing personalized feedback to students and helping teachers implement LP-based learning. We discuss the challenges of realizing these tasks and propose future research avenues.

In science education, a three-dimensional learning progression describes students' knowledge-in-use development of disciplinary core ideas, crosscutting concepts, and science and engineering practices over time so that knowledge becomes more sophisticated, allowing learners to apply their knowledge in new and compelling situations. Researchers put efforts into developing and using LPs in learning systems to support student knowledge-in-use. With the rapid development of digital technologies, artificial intelligence (AI) can serve as a partner to provide potential solutions to support LP-based learning systems further. This chapter addresses this essential research agenda: 1) positioning theoretical perspectives and a conceptual model for integrating AI into LP in learning systems; 2) discussing our current work on integrating AI into knowledge-in-use LPs with automatic scoring information, personalized feedback, and instructional support in learning systems; and 3) envisioning the future research work with recommendations for AI-empowered LP-based learning systems.

Three-dimensional learning progressions (3DLPs) have received wide-scale attention in depicting learning pathways that students travel as they progress toward knowledge-in-use in making sense of phenomena and solving problems. Using a design-based research approach, this study put efforts into developing a 3DLP of matter, interaction, and energy at the middle school level and presents the essential design principles for developing 3DLPs. Our chapter offers an illuminating exploration of our design-based research, shedding light on the fundamental design principles and systematic processes involved in developing 3DLPs. By identifying design principles and establishing a systematic process, we unlock new perspectives and approaches to shape the future of designing and implementing 3DLPs to support student knowledge-in-use development.

This study explores the adaptations made by elementary school teachers to promote student Productive Disciplinary Engagement (PDE) during modeling activities within an online learning context. PDE means students make collective intellectual progress by using disciplinary practices and ideas to formulate possible solutions or answers. Given the abrupt transition to remote education instigated by the COVID-19 pandemic, student engagement and hands-on learning opportunities have faced significant challenges. This study adopts a collective case study design with an interpretivist perspective, investigating three fourth-grade teachers over the 2020-2021 school year. We collected multiple data sources, including professional learnings, observations, teacher interviews, and student artifacts. The study identified four significant themes pertaining to teacher adaptations that corresponded with increased students' PDE: leverage technology tools for constructing models; maximize student-centered choice and position students as epistemic agents in modeling; incorporate family and community resources for modeling; and utilize students' diverse knowledge and expertise in modeling. These findings contribute to our understanding of effective strategies for promoting student engagement and the practical application of scientific modeling in online learning environments.

One reason for the widespread use of the energy concept across the sciences is that energy analysis can be used to interpret the behavior of systems even if one does not know the particular mechanisms that underlie the observed behavior. By providing an approach to interpreting unfamiliar phenomena, energy provides a lens on phenomena that can set the stage for deeper learning about how and why phenomena occur. However, not all energy ideas are equally productive in setting the stage for new learning. In particular, researchers have debated the value of teaching students to interpret phenomena in terms of energy forms and transformations. In this study, we investigated how two different approaches to middle school energy instruction—one emphasizing energy transformations between forms and one emphasizing energy transfers between systems—prepared students to use their existing energy knowledge to engage in new learning about a novel energy‐related phenomenon. To do this, we designed a new assessment instrument to elicit student initial ideas about the phenomenon and to compare how effectively students from each approach learned from authentic learning resources. Our results indicate that students who learned to interpret phenomenon in terms of energy transfers between systems learned more effectively from available learning resources than did students who learned to interpret phenomena in terms of energy forms and transformations. This study informs the design of introductory energy instruction and approaches for assessing how students existing knowledge guides new learning about phenomena.