Validity And Washback Effects Of College English Achievement Tests

Since1960s, researches regarding language testing and evaluation have been centering around validity, one of the two most important criteria for measuring a given test. With the introduction of Hughes’washback validity and Messick’s consequential validity, validity theory stepped into a new era. Researches are conducted mostly under Messick’s framework of validity theory. Many researches on validity and washback effects have been done, both abroad and in China. Most researches in China are related to large-scale high-stake tests like CET, TEM, etc. Not many studies have been done on validity and washback effect in the college English tests.Based on the validity theories mentioned above and the washback hypotheses proposed by Alderson and Wall, this thesis plans to explore the validity of College English Achievement Test in HUSTWB (Huazhong University of Science and Technology Wuchang Branch), and then examine the washback effects the test has on teaching and learning activities. As an independent institute, HUSTWB has a similar College English teacher crew and almost the same students’ academic level to other independent institutes. The research on its language tests may serve as a reference for other independent institutes.This thesis will mainly answer the following2questions:1) Is College English Achievement Test in HUSTWB valid?2) What kind of washback effects, if any, does College English Achievement Test have on English teaching and learning in HUSTWB?In order to answer the first question, two aspects of validity are examined:content validity and construct validity. The author first got access to the final test paper and compared it with the syllabus to measure content validity. Then he collected data of College English Achievement Test in HUSTWB, and employed the software SPSS13.0for further data analysis. Two approaches were adopted:correlation matrix and factor analysis.The second question regarding washback effects was examined by means of questionnaires for teachers and students, and classroom observations. Date collected from the questionnaires is also mainly processed by the software SPSS13.0.Major findings of the research are as follows:1) College English Achievement Test in HUSTWB is to some extent valid. Its content validity in subtests shows different results. While the listening subtest indicates relatively high content validity, the reading subtest shows a modestly poor content validity. With regards to construct validity, correlation matrix suggests that the subtests don’t have a high correlation coefficient with each other; therefore the construct validity is not as high as expected. However, the factor analysis by SPSS13.0indicates relatively high validity as far as the2factors extracted are concerned, one for subjective questions and the other for objective questions.2) Evident washback effects were found on both teachers and students. Questionnaires and classroom observations both show evidence that College English Achievement Test has strong washback on teachers in terms of teaching content, teaching materials, teaching methods, classroom activity design, etc. However, the washback effects on students are not as strong as on teachers, partly due to students’lack of interest in English learning.At the end of the thesis, the author gives his suggestions towards College English Achievement Test and language teaching/learning in HUSTWB.