Washback Effects on the Vocabulary and Grammar Section of NMET: A Case Study of

Abstract：This paper specifically discusses item quality of 15 Vocabulary and Grammar items of 2010 Shaanxi NMET. Based on the current language testing theories， this paper studies the answer key and intended purpose of each item to analyze the washback of the 2010 Shannxi NMET. Through this analysis， this study indicates that designing successful vocabulary and grammar items is not an easy task. Suggestions on how to design these items in order to achieve intended beneficial washback are also provided to exam writers.

Key Words： Language Testing； NMET； Washback； Vocabulary and Grammar

Introduction

Shaanxi Province started to write NMET on its own since 2006. According to the statistic statement from the groups of management of NMET paper， until the year of 2010， five tests written by Shaanxi Province have been taken by candidates with the number of about 2 million in Shaanxi Province. Apart from that， another approximately 2 million parents together with about 50，000 high school teachers are also involved （Cheng）. Since so many people participated， it is quite necessary for us to analyze and evaluate the quality of this test to make sure the test items have successfully achieved their functions. In other words， we should make sure NEMT affects students and teachers as well as the social groups in a correct way so that students are tested on what they are supposed to be tested on while teachers direct their future teaching according to NMET items.

Testing Principles

Principle 1： The correct answer should be unique and the three distracters are not supposed to be the possible correct answer （Alderson 47）.

No. 18 It never occurred to me _____ you could succeed in persuading him to change his mind.

A. which B. what C. that D. if

The test focus of this item is the Noun Clause， and the test writers look forward to testing the difference among the four options and how they apply to the language communication. The most important requirement of a multiple-choice item is that the “correct” answer must be genuinely correct （Liu 48）. As for this item， the only possible answer key is C， because option A and B are used as subject， object or predicative in the clause in terms of the ingredient. Moreover， option D is excluded as it doesn’t fit to this sentence meaning. From the analysis above， it can be concluded that this item has a beneficial washback， from which students will understand they have to hold the ability to distinguish the difference among relatives when reviewing.

Principle 2： Each option should fit equally well into the stem （Heening 121）.

No. 17 John opened the door. There____ he had never seen before.

A. girl did stand B. a girl stand

C. did a girl stand D. stood a girl

Option A is a normal sentence but it emphasizes the verb “stand” by using the help verb “did”. Option B is a simple one， nothing special. Option C is partial inversion while option D is a complete inversion. If candidates are not aware what this item is intended to test， it is difficult to make a choice since the four options can all fit into the stem properly， which is what a good multiple choice question looks. The test focus is inversion. When beginning with the adverbials like “here”， “there” and so forth， the sentence should be completely inversion.

Conclusion

In conclusion， NMET， as one of the assessment tools， plays a significant role in curriculum design and instruction， mainly because the test results are used in policy making. After grading the NMET papers， the local educational department will promulgate the lowest criterion examination and students will see whether they are qualified to enter college. The larger a test scale is， the greater impact it will produce. Moreover， the impacts of NMET upon teaching and learning even ripple the educational system and the society as a whole. Although it is very complicated to write successful items in the real practice， as long as item designers work for it， the NMET will achieve its beneficial washback.

Works Cited

Alderson， J. Charles， Caroline Clapham and Dianne Wall. Language Test Construction and Evaluation. Cambridge： Cambridge University Press， 2000.

Bachman， Lyle F. “Modern Language testing at the turn of the century： assuring that what we count counts.” Language Testing 17（2000）：1-22.

Cheng， Hongmei. “Validity Study on Multiple-choice Question of National Matriculation English Test.” Journal of Chinese Education Research 12（2009）

Feng， Xiaoyuan. “What is Language Testing？” School of English Studies of Xi’an International Studies University， Xi’an. 20 March， 2010

Heaton， J. B. Writing English Language Tests. Beijing： Foreign Language Teaching and Research Press， 2000.