Using baseline studies in the investigation of test impact

Abstract

The purpose of this article is to discuss the role of ‘baseline studies’ in investigations of test impact and to illustrate the type of thinking underlying the design and implementation of such studies by reference to a recent study relating to a high-stakes test of English language proficiency. Baseline studies are used to describe an educational context before the introduction of an innovation intended to cause change. Data gathered at the baseline stage can serve as a point of comparison when attempts are made to determine whether change has indeed occurred. Though of obvious relevance to test impact studies, baseline investigations are rarely discussed in the language testing literature. This article presents the research objectives, instruments and procedures used in the baseline stage of a study investigating the impact of the revised Test of English as a Foreign Language (TOEFL). While this example shows the type of work that has proved effective in a specific setting, its key features may also prove useful in other contexts. The article concludes with a summary of these features and a reminder of their function within baseline investigations.