Purchase this article with an account.
Cho Kin Cheng, Lisa-Marie Collimore, Dwayne E. Paré, Shakinaz Desa, Steve Joordens; Look at the choices: An examination of looking behaviours in a multiple choice test. Journal of Vision 2009;9(8):413. doi: 10.1167/9.8.413.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Research has shown that the perception of test difficulty can alter performance on multiple choice tests (Weber & Bizer, 2006). The current study explores whether or not perceived test difficulty also has an effect on looking behaviours during testing. Specifically, we were interested in students' fixation and pupil size when participating in a multiple choice test where they were informed that the test would either be easy or difficult. It was hypothesized that when students were informed that the test would be easy, they would generally show less looking behaviours overall. In particular, that these students would look less at the choices following the correct answer.
A sample (N = 24) of undergraduate students participated in this study. Students were given a short article to read and were told that they would be tested on its contents. The test consisted of 12 multiple choice questions with four possible answer choices, and was administered using a desktop mounted eye tracker system (EyeLink 1000). We measured fixation duration, fixation count, and pupil size in four areas of interest; when students looked at the question, the correct answer, the choices preceding the correct answer, and the choices following the correct answer.
Our general hypothesis was supported; students receiving the easy instruction showed lower levels of fixation duration, fixation count and smaller pupil size in general. In contrast to our second prediction, they tended to have longer fixation durations for the choices following the correct answer.
In light of these initial findings, further analysis based on students' actual responses to the questions rather than the correct answers will be examined. Additionally, the results from a questionnaire administered after testing will be discussed.
This PDF is available to Subscribers Only