ABSTRACT
Validity in testing confirms that a test is relevant and well grounded. Nevertheless, it remains a contentious area because it is believed that many language tests do not always accurately measure what they claim to measure, especially when a discrepancy is subsequently observed between the results of a test and the actual language performance ability of test-takers. An example of this would be a grammar test which claims to measure a learner’s writing ability, namely the ability to use correct syntax and usage in writing a composition. Any experienced ESL teacher knows that there is a mismatch between grammatical proficiency, as demonstrated on grammar tests, and the concrete ability to produce plain, competent English prose with a minimum of syntactic and usage errors. Plainly, such grammar tests are not measuring the learner’s grammatical ability at a sufficiently deep level. Therefore, the validity of such testing instruments is at best suspect, and possibly irrelevant, possessing at best a low level of validity.
