Subgroup and Longitudinal Measurement Invariance of the Computational Thinking Test - Short (CTt-S)

• In Event: AERA Roundtable Session Wednesday 7:45 am - Gold 2
  In Roundtable Session: Computational Thinking Assessment and Learning Across K-12

Abstract

Validated assessments of computational thinking (CT) are critical for educators and researchers. Thus, we developed and validated a shortened version of the Computational Thinking Test (CTt), named the CTt-S. We administered the 8-item CTt-S to middle (n=173) and high school (n=144) students across three U.S school districts. We assessed item-response theory (IRT)-based differential item functioning (DIF) by student gender, race/ethnicity, home language, school type, and measurement occasion. The one-factor, 2-parameter IRT model demonstrated excellent fit to the data. DIF analyses indicated only one item showed evidence of DIF (by home language spoken). The CTt-S functions similarly across subgroups and time. It could be used to generate insights and valid inferences about CT alongside other measures within a single survey administration.

Authors

• Joseph C. Tise, Institute for Advancing Computing Education

• Monica McGill, Institute for Advancing Computing Education