Subgroup Mining of Learner Behaviors With Interactive Diagrams in Research Quest

• In Event: Poster Session 12
  In Poster Session: Explorations at the Intersection of Technology and Learning

Abstract

This study applies a subgroup discovery mining method to discover usage patterns with an interactive map diagram that predicted gains in student understanding with Research Quests. Research Quests are online, inquiry-based investigations that promote critical thinking skills via authentic resources. The current research focused on student interactions with a fossil map of the Cleveland-Lloyd Quarry. A total of 143 students in 6 classes completed a 2-hour learning session. The rules induced from the subgroup mining task were found to accurately predict 61% to 69% of student performance on formative assessments of the least and most useful observations drawn from the map and their broader implications. We discuss these results in terms of implementing detectors of usage patterns to individualize instruction.

Authors

• Eric G. Poitras, University of Utah

• Kirsten R. Butcher, University of Utah

• Michelle Hudson, Brigham Young University