3 Chapter 1 Introduction For centuries, grammar has occupied a central role in language teaching and learning. The study of grammar was not just considered an essential feature of learning, but was thought to be sufficient for learners to actually acquire another language (Rutherford, 1988). The belief is reflected by the dominance of grammar in language pedagogy and testing. In class, teachers mainly focus on presenting and explaining grammatical structures. Language learners are frequently tested on these areas largely by discrete-point measurement such as multiple-choice items. Although the central role of grammar in language acquisition has remained unquestioned, grammar pedagogy and assessment have encountered much debate among language educators and testers. In the field of grammar assessment, some educators and scholars have pointed out that the overuse of multiple-choice items, which are largely based on the perspectives of structural linguistics, cannot keep pace with the way grammar is currently viewed. However, there remains a lack of consensus on what type of assessment tasks might best test grammatical abilities. Due to the above reason, some famous large-scale examinations, such as TOFEL and CET, no longer include a separate section for testing grammar. Instead, grammar assessment merges into all the parts in the test paper. Although the vocabulary and structure part has been eliminated from the new CET, we see the advent of a new test type: Chinese-English translation. In the test specifications provided by the National College English Testing Committee, it is clearly stated that Translation is chiefly to examine students ability to use proper words and grammatical structures and to convey ideas in idiomatic English. Since the Chinese-English translation is a brand new test task for assessing grammar, it will become the focus of the present study. 1.1 Rationale of the study College English test (CET), a large-scale, high-stake national examination, is designed as an objective and accurate evaluation of the English proficiency of the college students in China. Since its first implementation in 1987, CET has motivated 4 teachers and students to constantly improve English teaching and learning. Gradually, CET has received much attention both from institutions of higher learning and society at large. With millions of test-takers every year, CET is undoubtedly the most popular standardized English test in China. The year 2006 witnessed a reform of CET, where enormous changes occurred in the whole test. One of the changes of the new CET is the emergence of a brand-new test type, Chinese-English translation. It accounts for 5% of the total score in CET. According to the test specifications, this new test type is directed at testing students “employment of proper words and grammatical structure and conveyance of ideas in idiomatic English”. As the discrete-point measurement of grammatical ability has been the target of much criticism, it is understood that the translation part is intended to be used as a substitute for the vocabulary and structure part in the old CET. It provides a more integrative way to assess students grammatical ability. Though five years has passed since its first implementation, there is a surprising lack of discussion on the validity of this new test type. Can it validly assess students grammatical ability? What attitudes do students hold toward this new test type? In what way does it influence grammar pedagogy and English pedagogy as a whole? To answer questions such as these, it is necessary to conduct a systematic empirical study which may justify or falsify the use of translation as a test-type for testing grammar and structure in the new CET. 1.2 Purpose of the study Informed by Bachman and other scholars framework of test validation, the author conducts both quantitative and qualitative studies to examine the validity of Chinese-English translation as a test type for grammar and structure. Through the study, the author hopes to get a better understanding of the construct, content, face and response validity of Chinese-English translation in the new CET. Hopefully, some constructive suggestions can be made to improve this new test type. With an aim to address the above issues, the author formulates the following research questions: 1. Can the test type of Chinese-English translation validly assess students 5 grammatical abilities? 2. Would the test type of Chinese-English translation prove to be more valid in comparison with other test types (i.e. Multiple-choice grammar test, error correction, etc.)? 3. What kinds of attitudes do Chinese college English learners hold toward the Chinese-English translation part? 4. What are the possible backwash effects that may result from the use of Chinese-English translation? 1.3 Research methods The present study involves both quantitative and qualit