Data Expedition in a Knowledge-Building Community

• In Event: Exploring Data Science Across the Curriculum and Across Grade Levels

Abstract

The ongoing “data revolution” is transforming knowledge production, governance, and civic engagement. This paper addresses how data science can be integrated into social studies at the elementary level through a pedagogical model named Data Expedition. Data Expedition integrates student inquiry with open data and is distinguished from existing initiatives by its foci on (a) harnessing real-world open data for student usage, (b) integrating data science in the school curriculum through computationally rich inquiry, and (c) scaffolding students’ self-expression and computational participation within and across knowledge communities.

In our pilot study, 22 sixth-grade students were supported to conduct “data expeditions” in a knowledge-building community (Scardamalia & Bereiter, 2014) using the Common Online Data Analysis Platform (CODAP) and Knowledge Forum (KF). Throughout the semester, the teacher incorporated data expeditions in social studies. Students worked in flexible groups to examine world issues such as gender equality and climate change. Our research questions were: (a) To what extent were students able to analyze open data using CODAP? (b) How did data and analysis facilitate students’ problem finding, theory building, and peer interaction?

Our data sources included 21 student CODAP notebooks and students’ written discussion in KF. Content analysis was applied on these data. First, we coded all CODAP notebooks in terms of (a) graph type (e.g., scatterplot) (Angra & Gardner, 2016); (b) structural complexity (Friel, Curcio, & Bright, 2001); and (c) level of graph comprehension—“Level 1–Reading the data,” “Level 2–Reading between the data,” “Level 3–Reading beyond the data” (Curcio, 1987; Friel et al., 2001). Second, we examined KF notes related to CODAP notebooks with a focus on the question–theory dynamics in the data cycle (Gould et al., 2016).

CODAP analysis. Students predominantly relied on scatterplots, the software’s default graph choice. Students demonstrated a command of structural components such as regression lines and variable legends. Content analysis of student interpretations showed their capability in reading between and even beyond the data (M=2.05, SD=0.79; see Figure 1). These graphs also triggered various conceptual and emotional expressions, including explanations, questions, surprises, and discontent (see Figure 2).

KF analysis. Content analysis of student notes revealed that students primarily focused on sharing perspectives and seeking resources to raise awareness about a world issue, using data to support findings and to create a sense of urgency. For example, in one data cycle, a student used CODAP to explore a dataset on which countries pollute the most and was surprised to find that North America had the highest CO2 emission per capita (see Figure 3). This unexpected finding prompted some students to start a second data cycle on why people have not taken action on climate change, while others felt compelled to share this result on social media to inform the public.

This study demonstrates that with proper support, 12-year-olds are capable of engaging in productive data analysis, graph comprehension, and collaborative discourse around data. CODAP as a user-friendly data science tool mitigates technical challenges while collaboration tools like KF can further aid student expression, storytelling, and knowledge building.

Authors

• Bodong Chen, University of Minnesota - Twin Cities

• Valeria Barbaro, University of Minnesota - Twin Cities

• Leanne Ma, OISE/University of Toronto

• Ben Peebles, University of Toronto - OISE