The Constructs Behind the Clicks
A new ACT research study finds that social and emotional skills, as measured by ACT® Tessera®, directly relate to how students interact with online learning activities and materials, which can be used to predict and improve student educational outcomes.
The study, conducted in partnership with Blackboard, the University of Maryland Baltimore County (UMBC) and VitalSource, sought to understand how social and emotional skills are related to students’ online behaviors and course outcomes within a learning management system (LMS)—an interactive online learning environment—in order to identify ways to help improve student outcomes.
The research found that social and emotional skills have systematic relationships with students’ online learning behaviors within an LMS.
The study involved ACT Tessera, ACT’s assessment system that measures social and emotional learning skills, being administered to 527 UMBC students enrolled in an introduction to chemistry course hosted on Blackboard Learn, the institution’s LMS –prior to the start of the course. Researchers then collected data from Blackboard Data to understand how students performed a variety of learning activities throughout the semester-long course.
“For many years, we’ve been able to predict student grades from LMS data,” said Dr. John Whitmer, lead researcher and senior director of data science and analytics at ACT. “And we also know that social and emotional skills have significant relationships with course grades, but we haven’t understood how the two might be related. This study helps us to understand the deeper constructs that underlie student online behavior, which in turn will allow us to more effectively improve the learning outcomes of those students.”
ACT researchers looked at what kinds of activities students completed within the LMS (e.g. posting on discussion boards, looking at assignments ahead of time, etc.) and used these data to create predictive models to determine course grades. Combining social and emotional skills with these models more accurately predicted course grades than baseline models created from student demographic information and academic experience including current college GPA.