A Content Validity And Washback Validity Study Of Cet Writing

College English Test (CET) is a nationwide exam which is administered by the National Ministry of Education. Based on the College English Curriculum Requirements, CET evaluates the college students’English ability objectively and impartially, promotes the College English teaching and learning in China, and provides reference of selecting talents for the personnel departments at all levels in our country. Of course, no test can be perfect, with CET being of no exception.Reliability, validity and feasibility are the three basic issues of language tests, among which validity is the most important one (Yang Huizhong,1998:55). Validity refers to the extent that can prove the interpretation and use of a test is reasonable. Currently, the researches of validity mainly lie in construct validity, content validity, concurrent validity, predictive validity, face validity and so on(Yang Huizhong,2003). If a test measures the language skills that a tester intends to test, then it has content validity. The higher the content validity a test has, the more accurately it reflects the content it is intended to measure. If the intended content cannot be reflected in a test, the scores of the test will not be accurate, that is to say, test scores may not reflect students’ability. Such a test can easily have negative washback on language teaching and learning. As one of the important parts in CET, to what extent the content and washback validity does to CET writing has become a core issue.The study systematically reviews the nature and the classification of testing validity, and the current researches of writing validity both at home and abroad. It attempts to explore two issues:(1) to what extent is CET writing valid in terms of content and washback validity? And (2) in what way can the content and washback validity be improved?Based on Bachman and Palmer’s task characteristics of theory and the model of Alderson and Wall’s Washback Hypotheses, this article employs the quantitative and qualitative research methods and the research instruments of questionnaires and interviews. The investigation is issued to the540college students who have attended CET-4or CET-6from2009to2011in six universities or institute in Guangxi. The data are collected and analyzed by SPSS.The results show that:(1)Most students consider that CET writing can reflect their writing ability and it has high content validity, but the choice of topics and genres, the arrangement of time and the length of the tasks are still required improvement.(2) CET writing has both positive and negative washback on students’learning attitude, objectives, time allotment, content and methods, but the positive washback is higher than the negative one.The ultimate aim of testing and teaching is to promote students’ learning. In order to improve CET writing’s content and positive washback validity, and make it serve students’ English learning better, the author also offers some suggestions. Lastly, the author indicates the research drawbacks and points out the direction for future studies in this field.