An Evaluation Paradox: The Issues of Test Validity in the Realm of Writing Test as the Final School Examination in the Indonesian Senior High School Milieu

Abstract

Even though there are four English language skills in the Indonesia’s national curriculum at upper secondary schools, each of these skills is given an unequal emphasis since only reading and listening skills are formally tested in the national examination. Although writing competence possesses a particular stake as the determinant of students’ achievement after students undergo a three-year education at the upper secondary school level, it appears that the existing writing tests are low in terms of test validity, as demonstrated by a preliminary study. A further study is carried out to probe the issues of test validity by deploying the framework of test validity, which pertains to theory-based validity, context validity, scoring validity, criterion-related validity, and consequential validity in the scrutiny of the existing writing tests. It is revealed that the current writing tests are fraught with validity problems in all of these facets of test validity. This is corroborated by interview data in the preliminary study and the analysis of the existing writing tests. These particular issues obviously evoke an ambivalence between the exalted educational objectives in the national curricula and the edifice of English assessment. Based on the findings, several implications and directions rise for future praxis of writing assessment.