Show Less

Testing a Nation

The Social and Educational Impact of the College English Test in China


Mark Garner and Dayong Huang

Many countries have national policies in relation to English language teaching that are monitored through standardized tests, and students’ performance in these tests may have a significant impact on their career prospects. When such high stakes become attached to a language test, it begins to play a far greater role than originally intended.
A preeminent example is the College English Test (CET), taken biannually by upwards of ten million students in China, which makes it the world’s largest national English test. Its impact is evident in many areas of Chinese society. Specified grades on the CET are requirements for graduation from universities, many job applications and even some residence permits. Consolidated CET results are widely used for rating teachers for promotion and for competitively grading institutions, hence influencing strategic planning by universities, government departments and companies, particularly those engaged in publishing or bookselling. The CET has, furthermore, given rise to a highly organized cheating ‘industry’, which is the subject of frequent governmental disclaimers and warnings.
This book reports on an extensive study of the impact of the CET in China, both on the lives of students and teachers and on educational and governmental institutions. The authors also draw theoretical and practical implications from their study for educational planners in other countries.


Show Summary Details
Restricted access

Chapter 2 Previous studies of test impact


This chapter reviews discussions and empirical studies concerning test impact in relation to both education in general and language teaching in particular. It consists of three parts. The first part covers general education and focuses on theoretical and empirical studies of the ef fects of testing. The second part deals with language testing, with an emphasis on the most widely studied aspect of test impact, namely washback. The third part synthesizes the main conclusions from the review and indicates gaps in the current research literature, some of which are partially filled by the findings reported in Chapters 4–8. Test impact studies in education Although the impact of testing was discussed as early as 1802 (Wall 1997:291), systematic scholarly interest began about the middle of the last century (Wall 2005:33). More recently, Popham (1987:679) asserted that ‘measurement-driven instruction (MDI)’ is the most ‘cost-ef fective way’ to improve education. His justification was that high-stakes tests change the nature of instructional programmes and make teachers focus their instructional activities on the knowledge and skills assessed by the tests. He outlined five conditions for the tests to work (Popham 1987:680). Tests must be criterion referenced rather than norm referenced. They must assess defensible content and proficiency. They must have a manageable number (5–10) of assessment targets. They must function as vehicles to improve teaching. Finally, educators must receive adequate teaching support. In 30 Chapter 2 support of his views, Popham cites studies showing a correlation between the...

You are not authenticated to view the full text of this chapter or article.

This site requires a subscription or purchase to access the full text of books or journals.

Do you have any questions? Contact us.

Or login to access all content.