Bibliographic details and abstracts are available to all. Downloads of full-text dissertations are restricted to University of Portsmouth members who must login. MPhils may be accessed by all.

Abstract

Since its inception over 100 years ago, the multiple-choice test item has become ubiquitous in
many professional and educational contexts, including language testing. The debate over
how many options is best for a multiple-choice item has produced a number of studies from
fields outside of language testing that conclude that that three option multiple-choice items
can perform just as well, if not better than their four-option counterparts. Similar studies
have been conducted within the language testing context however they are relatively few in
number and are less conclusive regarding which format is superior. This study, conducted on
university-age Vietnamese learners (n=81), involved comparing the performance of two
multiple choice tests, comprised of items containing four and three options each. The two
tests were analysed for difficulty, item discrimination, distractor appeal and discrimination,
and test reliability. The findings indicate that three-option items are easier and contain more
appealing and discriminating distractors. They also suggest that four-option tests are more
reliable, and four-option items are slightly more discriminating. These finding largely reflect
those found in the research literature and provide further evidence that three-option items
perform at least as well as four-option items. It is recommended to conduct the study again
but with two face-to-face test administrations, rather than relying on a simulated test, as well
as to expand the focus of research to other language skills, different demographics within the
Vietnamese population, and online exam delivery.

Course: Applied Linguistics and TESOL (DL) - MA - C2023

Date Deposited: 2022-09-06

URI/permalink: https://library.port.ac.uk/dissert/dis14012.html