Dynamics of item reading and answer changing in two hearings in a computerized while-listening performance test: an eye-tracking study. Issue 5 (3rd July 2020)

• Record Type: Journal Article
• Authors: Aryadoust, Vahid
• Is Part Of: Computer assisted language learning. Volume 33:Issue 5/6(2020)
• Page Start: 510
• Page End: 537
• Publication Date: 2020-07-03
• Subjects: Answer changing -- two hearings -- eye-tracking -- item format -- item reading -- while-listening performance tests; Language and languages -- Computer-assisted instruction -- Periodicals; 418.00285
• DOI: 10.1080/09588221.2019.1574267
• Languages: English
• ISSNs: 0958-8221

Abstract

Abstract: The aim of the present study is two-fold. First, it uses eye-tracking to investigate the dynamics of item reading, both in multiple choice and matching items, before and during two hearings of listening passages in a computerized while-listening performance (WLP) test. Second, it investigates answer changing during the two hearings, which include four rounds of item reading taking place during: pre-listening in hearing 1, while-listening in hearing 1, pre-listening in hearing 2, and while-listening in hearing 2. The listening test was completed by 28 secondary school students in different sessions. Using time series, cross-correlation functions, and multivariate data analyses, we found that listeners tended to quickly skim the test items, distractors, and answers during pre-listening in hearing 1 and pre-listening in hearing 2. By contrast, during while-listening in hearing 1 and while-listening in hearing 2, significantly more attention was paid to the written stems, distractors, and options. The increment in attention to the written stems, distractors, and options was greater for the matching items and interactions between item format and item reading were also detected. Additionally, we observed a mixed answer changing pattern (i.e., incorrect-to-correct and correct-to-incorrect), although the dominant pattern for both item formats (67%) was wrong-to-correct. Implications of the findings for language research are discussed.