A Study On The Validation Of C-test On Measuring Learner's Language Proficiency In Chinese ESL Classroom

Cloze procedure was originally developed to assess the readability of text in native language in the 1950s (Taylor, 1953). Later other researchers such as Oiler (1973, 1979) and Anderson (1979) have proposed that cloze procedure is an effective means of measuring second and foreign language proficiency. Due to its variance in deletion start and ratio as well as its scoring method, there are a variety of cloze procedure, the representatives of which are classical cloze test (the fixed-ratio cloze and the rational-deletion cloze), multiple-choice and the C-test. In my thesis, C-test is to be the focus of study to testify its validation of C-test on measuring learners' language proficiency in Chinese ESL classroom.The rationale of the study lies in the following two aspects. First, Klein-Braley and Raatz believe there are some problems with the classical cloze, such as the deletion ratio and the scoring method. Thus they have developed C-test as a modification of cloze tests to to tap learners' language proficiency. Other researchers such as Carroll (1987), however, highly suspected its reliability and validity. Second, there are insufficient validation studies on C-test in China, and it is still quite new to Chinese teachers and testers. In my thesis, I will probe into its reliability, concurrent validity, face validity and its backwash to the Chinese classroom teaching and testing.The research was divided into two parts: the C-test and the questionnaire. The subjects were chosen from two Grade One classes in two senior high schools with their scores of the senior high school entrance exams as the criterion measure. The test papers were scored by hand and the data were then processed through computer software SPSS. As for the results of the questionnaire, both quantitative and qualitative approaches were made use of to find out the face validity of the C-test. The overall results show that C-test bear a satisfactory reliability and concurrent validity but its face validity needs to be improved. Thus it can be concluded that C-test can measure second-language learners' language proficiency but needs more empirical studies and researches to prove its future in China's ESL teaching and testing.