Validating the performance standards in the 2005 and 2006 national primary school achievement tests in Mathematics and Language Arts
De Lisle, J. (2008). Validating the performance standards in the 2005 and 2006 national primary school achievement tests in Mathematics and Language Arts. In L. Quamina-Aiyejina (Ed.), Reconceptualising the agenda for education in the Caribbean: Proceedings of the 2007 Biennial Cross-Campus Conference in Education, April 23-26, 2007, School of Education, UWI, St. Augustine, Trinidad and Tobago (pp. 563-579). St. Augustine, Trinidad: School of Education, UWI.
De Lisle, Jerome
Performance standards are informed expectations of student achievement levels in a population. These expectations are based on knowledge or skills in the curriculum (content standards) or on the content and demands of test items. Arguably, performance standards for national assessments of educational achievement are required to evaluate quality and equity within an education system. In 2005, performance standards were introduced into the reporting system for the Trinidad and Tobago primary school national assessments. The critical question was the meaningfulness and usefulness of these performance standards; an issue that relates to validity. Validation considers the legitimacy and defensibility of procedures. Evidence for the validity of standards may come from three different sources: procedural, internal, and external. This paper analysed these three sources of validity evidence for the performance standards established in the 2005 and 2006 national assessments. Following recent trends, the study makes significant use of qualitative data obtained from the judges in evaluating the cognitive processes and perceptions associated with judging and setting standards in situ. This type of data was obtained from journals and responses to open-ended questions in a questionnaire administered to participants during the standard-setting process. The 2005 data suggest that while some evidence was strongly supportive of procedural validity, problems in logistics and management of the process proved noteworthy limitations. Targets of improvement for future standard-setting exercises should focus on addressing these deficiencies. Additionally, future evaluations of the process should include additional quantitative information necessary to judge the internal and external validity