Validating Parallel Language Tests: A Study of Two 'Parallel' National Reading Tests in Nepal

Validating Parallel Language Tests investigates the parallelness of the school level two equivalent national reading tests. The study is based on the test performance of the test-takers and their questionnaire responses with regard to the tests. It also draws on the responses of the test-takers’ teachers and the ‘experts’’ judgments on the two tests. The study uses the classical test theory in order to analyse the tests scores. The ‘experts’ judgments are analysed to examine the construct (content) validity of the tests. Also, the test task characteristics (using Bachman and Palmer’s [1996] framework) are examined comparing the two tests, followed by an analysis of the test-takers and their teachers’ responses. The findings suggest that the two tests do not appear to be parallel which is caused by the discrepancy between them with regard to reading skills and test task characteristics. This study also recommends certain ways that can help the examining body and the test developers to improve the parallel tests. This book will be useful for postgraduate students, their tutors and researchers interested in validating language tests and assessment