This NSF-funded project, Inclusive Data Science Education for Rural Elementary Students: A Research Practice Partnership for Agile Learning, was aimed at increasing participation in STEM education through developing and implementing computational data science units in rural elementary classrooms. Over four years our research team engaged in a research-practice partnership (RPP) with a rural elementary school in South Carolina to co-create, with teachers, elementary-level data science curriculum that was customized based on the context and needs of the school community. Our work together focused on providing relevant data science units to underserved (rural, special education) students. As such, our team researched data science curricula development with more than 30 rural elementary teachers, including two virtual teachers, and 600 students in their classrooms. We hosted in-person week-long workshops for two summers that were focused helping teachers use Universal Design for Learning (UDL) to co-create interest-based computational data science units for their students. Working with our educational partners, we were able to provide learning opportunities responsive to the needs of rural schools, the participating teachers and their students. During this project, we helped teachers align the data science units to topics the students cared about, requisite educational standards and everyday classroom instruction, while embedding computational thinking skills in the learning activities. We also learned about the important assets that rural teachers, their students, and families contribute to developing engaging instruction and learning with data science.

 We investigated “How can data science curricula be effectively designed, customized and integrated into inclusive, rural elementary classroom instruction to advance CT learning?” Our research plan focused on the impact of the data science curricula on both teachers and their students by examining ways elementary teachers and researchers co-designed a data science curriculum aligned with student interest and STEM practices. We examined elementary teachers’ successes and challenges during iterative implementation cycles of the data science curriculum, and we explored how they used UDL in their instructional practices. Using surveys, we measured and found significant increases in elementary teachers' confidence, self-efficacy, and interest in teaching data science through the co-design and implementation of data science curricula. Similarly, we found that the data science curricula impacted elementary students' data and computational problem-solving practices and confidence in using data science. This increase was also notable in students with disabilities. Teachers consistently used UDL to plan and implement the data science units in their classrooms, and their approach towards integrating both CT and UDL increased in the second year of data science unit implementations after they developed a strong community of practice to support expanding and improving on their instructional strategies.

 There were several key outcomes from this project. The project advanced knowledge of how to co-create interest-based data science units for elementary-aged students and resulted in the creation of a teacher and student-friendly data science instructional framework to guide curricular design.  Our research team also developed a novel data visualization tool, Groova, to assist elementary teachers and students in analyzing and visualizing data in order to tell data stories. Groova was developed based on teacher’s desire for a free, developmentally appropriate tool for young children, and revised with teacher input in the third year of the project.  An important key outcome of this project was the generation of more than more than 20 data science instructional units developed by K-5 teachers, with exemplars available on our project website datapopups.com. Teachers also developed and identified associated resources such as instructional presentations, curated data sets, and engaged students with STEM experts in their local community who use data in their everyday work. The grant resulted in six published journal articles and eight conference presentations to date.

This projects engaged more than 600 underserved youth (rural youth, students of color, students with disabilities), their teachers and some community members in data science and CT,  which may lead to early pathways towards STEM careers, broadening participation in STEM.  We provided mentoring and hands-on experience for six interdisciplinary (Learning Sciences, Special Education, Teacher Education) graduate students working on the project over four years; these graduate students may influence the next generation of educational researchers interested in data science, rural populations, and students with disabilities. The curriculum and associated resources that are available on our website, datapopups.com, and our free data visualization tool for early elementary students, Groova (available at Groovadata.com). An unexpected outcome was the continuation and expansion of the data science PD and implementation by instructional coaches at the school to other teachers and grade levels after our research team stepped back. We served as off-site guides versus facilitators, and the units were extended to grades K-1.

Last Modified: 10/07/2024
Modified by: Danielle Herro