Embodying Data Points a Within Real-world Phenomenon

• In Event: How Data Science Education Can Enable Exploration of Real World Phenomena

Abstract

Objectives
This study examines how physically embodying data points and manipulating data objects about a real-world phenomenon can support fourth graders in developing an understanding of data and data science. This study asks: how does exploring a real-world phenomenon through physical experience and digital data visualizations shift elementary students’ understanding of data and data science.

Perspective(s)
This study builds on three strands of literature to examine the role of physical experience in elementary data literacy: embodied cognition (Danish et al., 2020; Nathan, 2021), lived experiences and emotion from data feminism (D'Ignazio & Klein, 2020), and the humanistic approach to data science (Lee et al., 2021). The study focuses on developing students’ awareness of the ubiquity of data (Fagerlund et al., 2024), interpreting data in context (Lee & Wilkerson, 2018), and “data moves” such as filtering and grouping (Erickson et al., 2019).

Methods
Eighteen 4th graders from a full-inclusion classroom in a charter school in the Midwestern U.S. participated in this five-days study. On Day 1, students drew creative data visualizations using self-collected peer conversation data. Day 2 and 3 focused on a dataset of 20 animals from a local shelter. Students first role-played those animals for an adoption event and then worked in groups to sort data cards with data about the animals based on different criteria. On Day 4, students explored the CODAP (The Concord Consortium, n.d.) data analysis platform independently. On Day 5, they collected evidence from CODAP graphs to debate whether animal age affects adoption readiness. This poster reports findings from pre/post-tests with a focus on one open-ended item, “what is data and data science”, teacher reflection, and classroom observation.

Results
Sixteen students completed both the pre- and post-tests. Their responses to both tests were open-coded to identify data-related understanding (e.g., data ubiquity, representations). The number of codes applied to each response was counted for a paired t-test to check conceptual understanding. Analysis showed that students demonstrated a more nuanced understanding of data and data science from the pre-test (M=1.25, SD=1.29) to post-test (M=2.62, SD=1.31), t (15) = 4.57, p < .001. In the post-test, students showed an understanding that data exists everywhere in their lives, can be represented in graphs, and can be used for life decisions. The teacher appreciated the physical activities on Day 2 and 3 as accessible, especially for students with reading/writing needs. The teacher noted that physical experiences supported students’engagement with learning. Classroom observation supported this as students appeared eager to use data to advocate for the animals they role-played and requested an extension beyond the planned time on these days.

Significance
This study contributes to the understudied areas of developing data literacy for elementary students (Herro et al., 2025; Podworny et al., 2025). By integrating physical activities, this study challenges the view of math as disembodied, detached, and unemotional, and to create a more humanizing, inclusive, and accessible approach to data work (Matuk et al., 2022).

Authors

• Mengxi Zhou, Indiana University

• Joshua Adam Danish, University of Illinois at Urbana-Champaign