Associations Between Student Performance and Characteristics of Teacher Feedback Within an Online Math Assessment Platform (Poster 17)

• In Event: Friday Poster Session 9:35 am
  In Poster Session: Trends and Issues in Technology-Based Environments

Abstract

This study explores if and how characteristics of personalized and elaborated teacher-feedback (i.e., feedback length and uniqueness) are associated with students’ overall math performance on an online math assessment platform. In analyzing teachers’ tendencies to provide longer or more unique feedback to students with varying levels of academic performance, we aggregate our data at the student- and teacher- level and use regression analysis. We find that lower student performance is independently associated with both teacher-feedback length and uniqueness. However, modeled jointly, feedback length (but not uniqueness) is significantly associated with lower student performance. Our results, consistent across both levels of data, provide foundation for further research examining the connections between elaborated feedback and student learning in online contexts.

Authors

• Darin Johnson, University of Pennsylvania

• Teresa M. Ober, Educational Testing Service

• Michael Flor, Educational Testing Service

• Blair Lehman, Brighter Research

• Reginald Marshall Gooch, Educational Testing Service