Support Adaptive Testing: The Effects of Scaffolds in Computer-Based Tests

Abstract

This paper presents an investigation of the viability of an innovative approach to computing-based multiple-choice adaptive testing, called Support Adaptive Testing, focusing on the effectiveness of scaffolds in testing. Support Adaptive Tests (SATs) are based on the principles of assessment for learning, scaffolding, and learner autonomy. The SATs under investigation initially appear as traditional computer-based multiple- choice tests of reading, but items answered incorrectly are repeated with scaffolds and explicit feedback is given at the end of the test. The SATs were taken by 276 Thai and 121 Vietnamese undergraduate students, and data concerning perceptions, responses and time collected. Results show that the students were positive about the SATs, Test Highlighting was the most preferred and most used scaffold, the format where a predetermined appropriate scaffold was provided proved most effective especially for Vietnamese students, and the scaffolds differed in effectiveness for different reading micro-skills.