Student Engagement and Data Science Skills during Authentic Collaborative Inquiry: Insights from Contrasting Group Analysis

Abstract

We have previously developed an authentic, collaborative
inquiry model to simulate students’ interests using
real-world context and data, engage students in data
science practice, improve their data science practice
through external feedback and finally, trigger them to
reflect on the learning experience. This study further
delves into how student groups with different data science
performances demonstrate data science skills and how they
engage in the data science workflow. We analyzed the video
recordings of a productive and unproductive group and found
that the productive group scoped their investigative
questions early, followed and was mostly engaged in the
data science flow. In contrast, the unproductive group
focused on non-investigative questions and did not realize
issues with their inquiry until receiving external
feedback, which triggered their revisions and
disengagement. This study provides insights into challenges
K-12 students could encounter during collaborative data
science inquiry and how to support their learning.