The construct validation of ELI listening placement tests

Abstract

This study presents the validation process for the listening placement tests administered by the English Language Institute (ELI) at the University of Hawai‘i at Mānoa. The research questions are: (a) How does the ELI define the listening comprehension construct validity? (b) How well does the ELI Listening Placement Test (ELI LPT) measure the listening comprehension construct? And, (c) how differently do test-takers perform on the dictation test and the multiple-choice test according to language group? Participants in the research included international and immigrant students enrolled over three semesters in spring and fall 2010 and spring 2011. The study was conducted using a quantitative approach including test score analysis, test item analysis, and a survey, as well as a qualitative approach including curriculum analysis and interviews with administrators and instructors. The findings from the evaluation process addresses the three research questions, the ELI listening comprehension construct, positive and negative evidence concerning construct validity, and the different test performances of language groups. Some constructive suggestions based on these results are suggested for the ELI as well as follow-up research topics.