Language Testing- Introduction

Language Testing

Introduction
In a broad sense, a lot of tests can be classified as language tests, ranging from
exams at school (e.g. vocabulary tests, grammar tests etc.) or university to
certificates aiming to provide the holder with some sort of standardised
qualification. In this paper, I will only be dealing with certificates which claim toprovide a fair, standardised assessment as the basis for wide recognition. We will not be concerned with the testing of language within a restricted classroom environment, as clearly, this has to follow different rules, meet different necessities and is much more likely to employ alternative forms of assessment.
In general, we can distinguish two kinds of tests: proficiency tests assess theamount to which the testee has reached ‘proficiency’, i.e. a certain predefinedlevel, while achievement tests usually follow the principle of test as you teach .Therefore, it seems much easier to prepare for as well as to
interpret the results of achievement tests. However, to be of any use for
successful testees of language tests, it should be possible for, say, potential employers to draw inferences from the obtained certificates or level in a given test to the candidates’ actual language skills. This is what proficiency tests do. Whereas achievement tests look backwards in that they assess what should already have been learnt, proficiency tests tend to look forward in that they assess a person’s language skills and allow for interpretations of their future performance to be made6. This is why many, if not all, ‘official’ language tests or certificates are proficiency tests. Usually, the successful candidate is supplied with an interpretation grid accompanying the actual certificate to show which tasks typical candidates obtaining a particular level are able to fulfil. Apart from
statistical errors, however, problems can also arise due to the basic framework the tests refer to and the way in which this link is established. In the case of the TOEIC-test as well as for the KMK certificate, this framework is the Common European Framework of Reference (CEF).
However, it is not enough for language tests to refer to abstracted outside
descriptors of language competence, they also have to prove their reliability as well as their validity. In language testing, reliability means that the test really functions consistently ,whereas validity indicates the amount to which a testees result is true, i.e. whether it correctly reflects the testees actual language ability.