Construction And Practice Of SETCEE Model

Middle school English teaching takes as a goal the development of students' communicative competence. Whether or not the goal is reached should be measured through testing. Spoken English Test of College Entrance Examination (henceforth, SETCEE) is an important means of measuring students' oral communicative competence. However, the test items of the current SETCEE, such as reading aloud, asking and answering questions and oral composition, fail to measure students' oral communicative competence. There are serious problems in terms of reliability and validity. In retrospect of the theory of communicative competence model of language testing experts at home and abroad, this paper, based on the difficulties of spoken English test, constructs a SETCEE model in an effort to test students' oral communicative competence, which is characterized by authenticity, interactiveness, situationality, functionality and flexibility. Also, from the perspective of validity, reliability, difficulty, practicality and washback effect, its scientificness is proved.To begin with, this thesis introduces the importance of communication in the information times, the unique role of English in international communication and the urgent demand for English talents. Then the language learning and testing issues are brought out. In Part 2, with the accomplishment of the term 'communicative competence', it looks back respectively the three different communicative competence models put forward by Canale & Swain, Bachman and Li Xiaoju. Part 3 analyzes the difficulties of oral English test and validity and reliability problems existing in the SETCEE of Jiangxi Province: unequal physical conditions; improper scoring approach; test items failing to measure students' real oral communicative competence. In Part 4, the SETCEE model constructed in this research and each of the procedures are presented: guiding ideology, testing objectives, item type, item design, training of administrators and testers in test implementation, and scoring approach and scoring criteria. In Part 5, tentative test with the new model isconducted and recorded. In an attempt to learn its social effect, questionnaires are also made on both the testers and the testees who experience the real SETCEE and our tentative test. The result shows that the new model is superior to the current SETCEE and enjoys a high validity. Through remarking by individual testers themselves and remarking among testers, two groups of the results are achieved. The correlation coefficient of both results is high, which shows that our SETCEE model contains high reliability. Finally, from the perspective of difficulty, practicality and washback to middle school English teaching, appropriateness, scientificness and feasibility of our model are proved. Part 6, based on the arguments and demonstrations above, draws the conclusion that our model contains high scientificness and feasibility, and due to the items and scoring approach and criteria being improved, it can measure students' oral communicative competence more objectively and more fairly than the present-day SETCEE.