非英語專業學生英語成績測試的信度和效度分析

王玉粉　張立杰

【摘 要】語言測試，作為檢驗學生語言水平的一種有效方法，與語言教學之間相輔相成，并對語言教學有著明顯的指導作用。對于英語教學來說，成績測試是一個必不可少的環節，并在大學英語教學中起著舉足輕重的作用。因此，本文從效度和信度著手，通過分析某校非英語專業學生的英語期末試卷，從而為試卷設計以及大學英語教學給以一定的參考。

【關鍵詞】非英語專業學生；成績測試；信度；效度；英語期末試卷

Analysis of Reliability and Validity in Achievement Tests for Non-English Majors

--A Case Study on an English Final Test of a College

WANG Yu-fen ZHANG Li-jie

（College of Foreign Language Northeast Forestry University， Harbin Heilongjiang 150040， China）

【Abstract】As an effective way to test students language proficiency， language test has a close relationship with language teaching and also plays a vital role in guiding language teaching. As for English teaching， the achievement test， as an indispensable part， is crucial in college English teaching. Therefore， this thesis aims to analyze the reliability and validity of non-English majors in the achievement test by taking an English final test as an example so as to provide some suggestions for teachers to design an objective achievement test and to improve college English teaching.

【Key words】Non-English majors； Achievement test； Reliability； Validity； English final test

0 Introduction

As is known， we are faced with many tests， such as a quiz and final exams， which aims to test students language proficiency and reinforce teaching and learning. A test refers to a measurement instrument designed to elicit a specific sample of an individual behavior（Bachman， 1997）. It is categorized into diverse types following different criteria. Among those tests， the achievement test is of great significance and is an indispensable part in English teaching. It is designed for students to know how successful individual students， groups of students， or the course themselves have been in achieving objectives（Hughes， 2003）. Previous studies have shown that positive backwash effect of achievement tests is useful for English teaching and can also help students get correct learning methods to improve English skills（Zhao Liang， 2002）. Thus， this thesis aims to analyze the final test from the perspective of reliability and validity to ensure that this test succeeds in testing students achievement and proficiency.

1 Analysis of Reliability and Validity in an English Final Test for Non-English Majors

1.1 Test Reliability of the English Final Test

Reliability is the consistency of test scores and stability of the test results（Heaton， 2000）. But there are many influencing factors. One is the test itself whose reliability depends on the test range and the amount of items. Another is the scoring whose reliability rests with the objectivity and accuracy of scoring.

The author selects the English final test of the sophomore. The test is based on the New Standard College English（the 4th volume）， including unit 6 to unit 10. It consists of four parts.

The final test measures some aspects. As for the close， its a short passage selected from the exercises in the textbook， measuring students abilities to comprehend the context and grasp the use of words and key collocations occurring in the textbook. The second part is reading comprehension with 20 items， which are not from the textbook. The first three passages arent very long， each within about 800 words. The questions arent very hard except those from the third passage. Though the last passage is a little longer， about 1200 words， the questions are not difficult and students can make a correct choice based on the passage. The four passages are different and related with the articles of unit 6 to unit 10. As for the third part， there are two parts： passage translation and sentence translation. The former part selected from the articles in the textbook， is about translation from English to Chinese. The latter part selected from the exercises， is about translation from Chinese to English， including five sentences. The last part is writing. The topic is about “Better to Give”. Students can write a composition within 150 words and should be content-related and grammatically correct.

From the point of the amount of items， there are four parts within two hours. This is rational for the sophomore， even for those students with a low level. From the scoring point of view， 50% of items are objective and the rest is subjective. It indicates that the test is of high reliability.

1.2 Test Validity of the English Final Test

1.2.1 Face Validity

The validity of a test is the extent to which it measures accurately what it is supposed to measure and nothing else. If a test item looks right to other testers， teachers， moderators and testees， it can be described as having at least face validity（Heaton， 2000）. From the above table， it can be seen that the whole test covers four main areas： vocabulary， reading passage， translation and writing. The first and third parts are from the textbook， measuring students mastery of vocabulary and fundamental sentences. The second and the last parts are out of the textbook， measuring students reading and writing skills developed from the textbook. Especially， in the last part， students can put what they have learned， like key words and sentences， into practice. Hence， from the perspective of face validity， the English final test is of high face validity， measuring the knowledge or abilities it claims to measure.

1.2.2 Content Validity

A test has the content validity if it is constructed so as to contain a representative sample of the language skills， structures， etc. with which its meant to be concerned（Heaton， 2000）.

As is seen， the final test includes four parts. Thefirst part is close， including a short passage selected from the text exercises， but the teachers modify some choices based on the changed blanks to avoid that students just make a choice according to their memory. The modified close tests the use and distinction of words and collocations lectured from the textbook.

The second part is reading comprehension with four passages out of the textbook. The genres are narration and argumentation， concerning coral reefs， a Dutch Sinologist， criminality and securitization. Questions are about the theme， structure， vocabulary and details. Among the passages， the third one is a little difficult， because the choices are similar and ambiguous， which seems beyond the students abilities. But the four passages measure students reading skills developed from what they have learned.

The third part is translating the passage and sentences. There is only one passage selected from unit 8 to measure students abilities of understanding crucial words and expressions and the sentence structure. The next part， selected from text exercises， includes five sentences to measure students mastery of necessary expressions and sentence patterns that have been lectured. Although the first part is a little harder than the second part， both parts measure the knowledge the test claims to measure and are within students abilities.

The final part is writing. Given the topic， students are required to write a composition within about 150 words. The writing topic is familiar and students dont need to write some complex sentences. Thus， the last part measures students writing skills learned from the textbook and can be regarded as having a high content validity.

To sum up， this test meets the requirements and syllabus of achievement tests and is of high reliability and validity. Although only 30% are from the textbook， 70% selected from other materials also measure students abilities and skills developed from the textbook. Besides， 50% are objective and 50% are subjective. Thus， the scoring is objective and accurate to some extent.

2 Conclusion

This thesis has conducted an analysis of reliability and validity of an English final test for non-English majors. According to the analysis， this test is of high reliability and validity， but there are still some drawbacks. Hence， in order to be reliable and valid， a test should be designed based on teaching contents， purposes and others to assist teachers in adjusting syllabus and help learners be aware of their strengths and weaknesses and make progress.

【Bibliography】

[1]Bachman， Lyle F. Fundamental Considerations in Language Testing[M].Shanghai： Shanghai Foreign Language Education Press， 1997.

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

[3]Hughes， A. Testing for Language Teachers[M].Cambridge： Cambridge University Press， 2003.

[4]黃曉文.分析評價廣東外語外貿大學公開學院《商務英語閱讀》期末考試試卷的效度和信度[J].語文學刊，2013（10）：133-137.

[5]趙亮.以考促學—項強化、延伸期末考試反撥作用的試驗研究[J].外語研究， 2002（4）：51-54.

[責任編輯：王楠]