An Investigation into the Effects of the Number of Options in a Multiple-choice English Test

Abstract

What is to be discussed in this presentation is about the number of options in a multiple-choice test. Multiple-choice tests are undoubtedly one of the most commonly used types of tests, as they can measure a variety of content and learning objectives. Particularly, for testing listening and reading, multiple-choice tests are considered both effective and efficient. What is interesting, however, is that people usually think that multiple-choice tests are easy and simple to construct, although the opposite is true in reality. It is quite challenging to construct a set of attractive distractors, an essential component of a well-designed multiple-choice test. The test development becomes an even more daunting task, when it comes to writing more options. In general, a large number of options are perceived to reduce the element of chance. For this reason, five-option items have recently been adopted as the norm in many English language tests, although it may be difficult and often impossible to develop as many as five good alternatives. Now that five options are popular in use in the field of language testing, it is timely to test for the effects of the number of options in a multiple-choice test. Is it better to have more options in English test items? And what is the optimal number of options in a multiple-choice test of English and why? These are our key research questions. For this, we administered a multiple-choice English test comprised of 3, 4, 5 option items to 567 first-year university students, and performed an item analysis based on the IRT model.