Communicative Language Testing

3376 Words14 Pages
75 82 2000 Communicative language testing - an attainable goal ? Nick Miyata-Boddy* and Clive S. Langham * The British Council, Tokyo Introduction. In this paper we will first attempt to define the term ‘communicative language testing’. We will then go on to examine ways in which communicative testing differs from other forms of language testing, both in the theoretical basis and what is tested. In the next part we will identify some of the problems communicative language testing faces, and look at how these problems have been addressed. What is communicative language testing? Communicative language testing is intended to provide the tester with information about the testee’s ability to perform in the target language in certain context-specific tasks. It has to be recognised that given the constraints of time and practicality, only a small sample of the testee’s language can be collected, and that however realistic the tasks may be intended to be, the testee’s performance will inevitably reflect the fact that s/he was performing under test conditions. Differences between communicative language testing and other forms of testing We will address this by first briefly identifying other testing methods in the ‘eras’ preceding the emergence of communicative language testing, looking at what they were intended to measure and their theoretical basis. We will then turn to communicative testing and examine two of the communicative models on which it is based, and the characteristics which set it apart from other testing techniques. Spolsky (1975) identified three periods of language testing: the pre-scientific, the psychometric -structuralist and the psycholinguistic-sociolinguistic. Although he has since (Spolsky 1995) offered an alternative view, we will use his original phases in this paper. Spolsky first identifies the pre-scientific era. He recognises it as
Open Document