Abstract

As a high-stakes standardized test, IELTS is expected to have comparable forms of test papers so that test takers from different test administration on different dates receive comparable test scores. Therefore, this study examined the text difficulty and task characteristics of four parallel academic IELTS reading tests to reveal to what extent the four tests were comparable in terms of text difficulty, construct coverage, response format, item scope, and task scope. The Coh-Metrix-TEA software was used for the text difficulty analyses and expert judgments were used for task characteristics analyses. The results show that the four reading tests were partly comparable in text difficulty, comparable in terms of construct coverage and item scope, but not comparable in terms of response format and task scope. Based on the findings, implications were discussed on test development and future research.

This work is licensed under a Creative Commons Attribution 4.0 License.