Testing is an indispensable component of language teaching and learning, used to examine the effects of teaching and learning. With communicative language testing as a mainstream in the development of language testing, authenticity as one of the most important qualities which determine the overall usefulness of a test, is emphasized more than ever. Listening ability constitutes an important part of a language learner's language competence, however, there are few study conducted on the authenticity of listening test. The Test for English Majors Band 4 (TEM4) is a well-established large-scale standardized test in China, in which a listening sub-test takes up 30 percent of the total score. Therefore, it is worth exploring the authenticity of the TEM4 listening sub-test.Under the theories of Bachman's communicative language ability and communicative language testing, this paper describes the input characteristics of TEM4 listening sub-test tasks and TLU tasks on the basis of the framework of task characteristics proposed by Bachman and Palmer, in order to find out to what extent the TEM4 listening sub-test has its authenticity. This paper attempts to explore the input authenticity of TEM4 listening sub-test, aiming to answer the following two questions: 1) To what extent does the TEM4 listening sub-test task have its authenticity, in terms of the input characteristics? 2) Is there any room for improvement on the authenticity of the TEM4 listening sub-test?To answer the above questions, firstly, a literature review of communicative language testing, authenticity and listening test is presented. Then, an analysis of the TEM4 listening sub-test type and content is conducted, meanwhile, the authenticity of the TEM4 listening sub-test in terms of its input characteristics is analyzed by comparing the characteristics of the TLU task with the corresponding features of TEM4 listening sub-test. Finally, a questionnaire is conducted, aiming to provide some suggestions on improving the authenticity of TEM4 listening sub-test.The findings from the research are drawn as follows: 1. Generally speaking, the TEM4 listening sub-test has comparatively high authenticity, in terms of the input characteristics. 2. The vehicle, the degree of contextualization, the input variety and naturalness of the input of TEM4 listening sub-test don't replicate well with those in the domain of TLU listening. Thus, improvements should be made in those aspects 3. Improvements can also be made from the records, topics of the listening material, and the item types.This study analyzes the input authenticity of TEM 4 listening sub-test, points out some problems and makes some conducive suggestions about further improvement of this test. However, there are still a lot of limitations, like the small size of the samples, which needs to be further improved. |