Comparing Student Performance on Paper-and-Pencil and Computer-Based-Tests

• In Event: 58.037 - Issues and Challenges in School-Based Assessment Design

Abstract

Can results from computer-based tests (CBT) and paper-and-pencil tests (PPT) be considered equivalent measures of student knowledge? Studies addressing this question are growing, but additional research is needed before best practices for CBTs can be developed. We report results obtained using PPT and two different CBT systems. We used propensity score matching to create equivalent demographic groups for each testing mode and Rasch analysis to model student performance. Our results indicate student performance for the different test modes varies with grade level, students’ primary language, and the specific CBT system used. We discuss these results in terms of the current literature and differences between the CBT systems.

Authors

• Joseph Hardcastle, American Association for the Advancement of Science

• Cari F. Herrmann-Abell, American Association for the Advancement of Science

• George E. DeBoer, American Association for the Advancement of Science