CHALLENGES IN LANGUAGE ASSESSMENT

بِسْمِ اللّهِ الرَّحْمَنِ الرَّحِیمِ

CHALLENGES IN LANGUAGE ASSESSMENT

چالش در ارزیابی و سنجش

 Anyone involved in language education today recognizes the importance of testing and assessment. At the same time, however, politicians, parents, learners, and teachers all have different perspectives on why tests are so important and what they can be used for. As a consequence, an important part of English language teachers’ expertise includes a more thorough understanding of principles of assessment. However, in articulating those principles, teachers and researchers alike encounter three thorny issues.

1)      Defining the purpose of a test

Terms such as “placement,” “proficiency,” and “achievement” might be used what a test is supposed to be used for. These terms may work for the public, but teachers need a more systematic, thorough, and accurate way of considering the purpose of an English Language assessment because the purpose of the assessment is critical for choosing or developing a good test. We think of test purpose as consisting of three interrelated concepts:

I. One way to look at test purpose is to consider the inference that test users want to make on the basis of test scores. For example, we designed Longman English Assessment test users to make an inference about examinees’ English Language reading comprehension, listening comprehension, and structures. Making an inference about these three

aspects of English ability was one way of looking at the test purpose.

II. A second way is to consider the use of the test results. Typically, a test is given because a decision   has to be made about examinees, for example, their readiness to study at a university, their eligibility for a job, or the grade that they should receive in a course. Longman English Assessment is not intended for any of these relatively high-stakes test uses; instead, it is intended to give learners an idea of their level of English ability and to offer level-appropriate advice for improving their English.

III. The third aspect of test purpose is any other effect that the test is intended to have, such as serving as a motivating, interesting, and informative experience for learners.

2)      Understanding what the test measures

If we say that a test is a measure of reading comprehension, for example, as in the discussion of test inference above, two important questions remain:

A.  What do we mean by reading comprehension?

B. What does a specific score on the reading comprehension test indicate about a student’s reading comprehension?

The first question has to do with how the test inference is defined. Of course, from a psycholinguist’s perspective, many aspects of a reader’s linguistic knowledge and processes are called into play during reading comprehension. From the perspective of language assessment, we try to simplify this somewhat to include the idea that linguistic processing of morph-o-syntax and the lexicon are required as well as comprehension of rhetorical (metaphorical) organization and cohesion (ارتباط منطقی) across different test types and functions.

The second question has to do with how test performance on the specific test questions relates to what the test taker is actually capable of when reading, for example, an English language course book. Since we were interested in what the test performance would mean relative to English language courses, we developed most of Longman English Assessment through the use of materials from English language textbooks. We defined the types of questions we would ask on the basis of a detailed — but not too detailed — description of what it would take to understand the texts at various levels.

3)     Checking on validity

Everyone seems to agree that they want VALID ENGLISH LANGUAGE TESTS, but as teachers we need to have a clear idea about what this means. In fact, assessment specialists see the expression “VALID TEST” as misleading, because it implies that validity is an inherent quality of a test. A test is not in and of itself valid or invalid; instead, validity refers to the use of a test. For example, a test that we develop to assess ability to use complex sentences in an essay might be valid as part of an achievement test in an English language writing course. In other words, that test use would be valid. However, if the same test were used to place students into a particular level of reading comprehension that test use would be recognized as invalid. The test was the same in both cases, but the validity depended on its use.

The idea that validity refers to test use is critical because it means that English language teachers are often best able to comment on validity relative to their own test programs and classes. However, to do so, it is important that they understand the characteristics of tests that help to increase the validity of their use — characteristics such as reliability and authenticity. In developing Longman English Assessment, we were concerned about these test characteristics, but the final analysis of the validity of its use needs to be based on the teacher’s own understanding of these characteristics in view of its use.

These three testing challenges, as well as other questions about what and why we teach the way we do, are magnified when instruction and assessment take place through technology. Because of the commitment of time and resources involved in constructing a computer-based assessment, developers take great care to define the test purpose, interpret the score meaning, and consider evidence concerning the validity of the test. As a consequence, the use of carefully designed computer-based materials should help teachers to reflect on these three issues in all of the assessments they use.