Abstract:
Provided that English language testing at micro and macro level has been subjected to test construct- underrepresentation and construct-irrelevant, the study aimed at analyzing language test competencies and its usefulness: A case of National Examination Council of Tanzania (NECTA). Both Integrative Communicative Competence Model by Celce-Murcia et. al. (1995) and Bachman and Palmer (1996) model of test usefulness were used as the theoretical lens. To achieve the objectives above, a case study design of NECTA-Certificate Secondary Education Examination (CSEE) for English language were systematically selected whereby 2009, 2011 and 2013 national English Examinations were chosen for analysis. The researchers used two themes for the study namely: tester variability and construct- underrepresentation. Documentary review and content analysis were used as a tool for data collection and analysis respectively. The researchers found that tester variability is both condition and a matter of degree, provided that other language competencies such as linguistic competencies and strategic competencies were easy to be tested and better achieved compared to other areas such as discourse competencies and socio-cultural competencies which found to be difficult in achieving using paper-pencil tests unlike performative test. In case of test usefulness and test-under construction, the findings also showed that the test items remain uncertain and matter of degree regarding the coverage-ability, inclusions, authenticity and intergrativeness as well as predicative-ability. Other language competencies were easily assessed using paper-pencil or written production while those required oral production, interaction and meditation were easily ignored, for instance: speaking elements, listening elements and pragmatic competencies. However, what seem to be achieved in terms of communicative approach remains superficial drills of communicative approaches. It is recommended that since language testing is different from testing social science subjects, it is likely that both performantive and pen-pencil test should be designed by NECTA to achieve the all competencies of the language as claimed in the national secondary English language curriculum.
