Validating Parallel Language Tests

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.