Tests of instruction and education are supposed to assess the effectiveness, efficiency, and adequacy of teaching/learning resources (material and human) in relation to the goals and objectives of the instructional or educational system. Language is a highly abstract and complex communication system and does not lend itself to easy analysis. As a result, language theories have been unable to propose universal pedagogical methodologies. This article aims at identifying some of the theoretical and operational problems of TESOL and, to the extent possible, it makes recommendations for teachers and test-designers. It confronts the two major streams of approaches to testing, viz. the analytical and the integrative. The integrative--or synthetic--approach stresses communicative skills as opposed to discrete-point fragmentation of the communicative event. Sociolinguistically and learner-centered, the communicative approach, though laudable in its intent, has failed to develop testing instruments with acceptable reliability. Today, many researchers combine the two approaches, because proficiency is hard to achieve without competency. One of the problems facing the test-designer, in this perspective, is what weight to ascribe to each approach. The apparent antagonism between quantification and qualification renders the predicting of language learning output on the basis of instructional input well-nigh impossible to assess. The communicative approach sees in contextualization the key to improved language perception and processing.
Tests in instruction and education have been said to be used with a view to assessing "the effectiveness, efficiency, and suitability of the materials in relation to the instructional objectives" (Romiszowski, 1986:401). As simple as this statement seems, it demands a lot of defining of terms and explaining of the very understandings and goals of instruction and education. One of the difficulties facing all players...