A COMMON PROBLEM IN ASSESSING WRITING IN LANGUAGE PROFICIENCY TESTS (Mr Tich)

 Proficiency tests are becoming popular across the world. The number of participants in these exams is significantly increasing. People are more concerned about the accuracy in scoring speaking and writing skill. In the previous essay, I wrote about A COMMON PROBLEM IN ASSESSING SPEAKING IN LANGUAGE PROFICIENCY TESTS. In this article, I will discuss A COMMON PROBLEM IN ASSESSING WRITING IN LANGUAGE PROFICIENCY TESTS. When assessing Writing in language proficiency tests, I frequently find a problem with unclear and overwhelming assessment criteria and descriptors of the tests. This results in consuming time discussing, understanding and agreeing with other raters the criteria being used for scoring.

There are some causes to this common issue. The main cause is that different writing proficiency tests have distinct assessment criteria to understand and apply. The more writing tests I score, the more overwhelmed I feel. Even in one single test, there are a large number of various criteria and descriptors for a wide range of scores. Take IELTS writing test for an example, there are four assessment criteria including Task Achievement, Coherence and Cohesion, Lexical Resource, and Grammatical Range and Accuracy. The band ranges from zero to nine. This means there are different descriptors for separated criteria in distinct bands.

Apart from being overwhelming, some assessment criteria are not really clear. This causes inaccuracy in scoring and finalizing the candidates’ ultimate marks. For instance, regarding the Writing band descriptors (public version for IELTS tests), some phrases such as “a wide range of structures”, “a variety of structures”, “a sufficient range of vocabulary”, “an adequate range of vocabulary”, “a very limited range of words and expressions”, “only basic vocabulary” are used as descriptors for the same criteria but different bands. Some of them are alike (“an adequate range of vocabulary” and “a sufficient range of vocabulary”, or “a very limited range of words and expressions” and “only basic vocabulary”). Besides, a few descriptors are vague or ambiguous. For example, how wide and how various “a wide range of” and “a variety of” should be understood.

To sum up, unclear and overwhelming assessment criteria and descriptors of the tests really challenge me and raters in general. Therefore, examiners should have careful discussion and agreement on assessment criteria and descriptors before scoring to ensure the accuracy, reliability and validity for the tests.