Multiple-choice test development competencies of junior high school mathematics teachers in Ghana: a triangulation methodology

Background Previous studies on test development competencies have used three distinct approaches, but none have combined these methods, leaving a gap in understanding teachers’ test development skills comprehensively. Purpose Using a triangulation methodology, this study examined the multiple-choice test (MCT) development competencies of junior high school (JHS) mathematics teachers in the Sekondi-Takoradi Metropolis, Ghana. Methods A survey was first conducted with 218 mathematics teachers, followed by a documentary analysis of tests developed by purposefully sampled teachers. Finally, item analysis was performed using data from the teachers’ tests. The collected data were analysed through confirmatory factor analysis, structured qualitative content analysis, item response theory (IRT), and Kendal’s Tau-b. Results The study found that teachers generally reported multiple-choice test competencies in areas of test item assembling, content validity, and test option handling. An evaluation of sample items revealed that most of the developed MCTs lacked clear directions, contained ambiguous or irrelevant content, and had generally poor quality. The IRT analysis revealed generally poor psychometric properties of the test developed by the teachers. A moderate, statistically significant positive correlation was observed between adherence to recommended MCT development principles and the quality of items produced. Conclusion JHS mathematics teachers primarily developed low-quality MCT items, which is concerning because poor-quality tests undermine accurate assessments and sound educational decisions in Ghana. Recommendations The Ghana Education Service, NaCCA, and District Directors should prioritise sustained, hands-on professional development workshops in test construction and item analysis to strengthen in-service teachers’ assessment literacy and improve classroom testing practices.