skip to main content skip to footer

The Effects of Different Levels of Performance Feedback on TOEFL iBT Reading Practice Test Performance CALL TOEFL iBT ELL

Sawaki, Yasuyo
Publication Year:
Report Number:
ETS Research Report
Document Type:
Page Count:
Subject/Key Words:
TOEFL iBT, Test Performance, Reading Fluency, Reading Assessment, Japanese Students, Second Language Testing, Computer-Assisted Language Learning (CALL), Test Validity, Test Preparation, Feedback, Multilevel Modeling, English Language Learners (ELL), Instructional Video


The purpose of the present study is to examine whether performance on the TOEFL iBT Reading practice test is affected by 3 different levels of feedback provided to learners upon completion of reading exercises: (a) correctness of learner response (the knowledge of correct results [KCR] feedback), (b) KCR feedback and rationales for correct/incorrect answers, and (c) KCR feedback and the rationales along with on-demand video lectures. Japanese learners of English completed 18 reading exercises and reviewed the results by using the type of feedback provided to the group to which each learner had been assigned. They also completed 2 TOEFL iBT Reading test forms: 1 as a pretest and 1 as a posttest. A multilevel modeling analysis of data from 193 participants who completed the study showed that despite a gain of over 3 scaled score points on the TOEFL iBT Reading test from the pretest to the posttest, the level of feedback was not found to affect the test score. A learner background variable, previous knowledge about the TOEFL iBT test, was identified as a predictor positively related to the test score. Moreover, a series of ANOVAs conducted on learner survey data suggested that the level of feedback affected the degree to which study participants perceived the usefulness of the study materials in understanding different item types that appear on the TOEFL iBT Reading section and in developing strategies for understanding the gist of the text within a short time.

Read More