VALIDITY AND RELIABILITY OF CLASSROOM ASSESSMENTS IN RIVERS STATE SECONDARY SCHOOLS

Abstract

This study examined the validity and reliability of classroom assessments in Rivers State secondary schools. The study was anchored on Classical Test Theory and Assessment for Learning theory and adopted a descriptive survey design complemented with document and psychometric analysis. The population comprised public and private secondary school teachers and Senior Secondary II and III students in Rivers State. A multi-stage sampling procedure was used to select 10 secondary schools, 100 teachers, and 300 students. Data were collected using the Teachers’ Assessment Practicesm Questionnaire (TAPQ), sampled classroom test question papers and marking schemes, and students’ continuous assessment and examination score records. Content analysis, descriptive statistics, Cronbach’s alpha, Pearson correlation, and multiple regression were used for data analysis at the 0.05 level of significance. Findings revealed that most classroom tests showed limited content coverage and over-emphasis on lower-order cognitive skills, indicating weak content and cognitive validity. The overall mean reliability coefficient of sampled tests was 0.70, with several tests falling below acceptable levels. Teachers demonstrated moderate assessment practices, but critical technical activities such as preparation of table of specifications and item analysis were weak. Training in measurement and evaluation significantly predicted reliability of classroom assessments, explaining 36% of the variance. The study concluded that while classroom assessments in Rivers State secondary schools demonstrate moderate reliability, their validity and technical quality remain inconsistent and largely influenced by teachers’ assessment literacy. The study recommends systematic professional development, institutionalization of test moderation procedures, and strengthened quality assurance mechanisms to enhance the psychometric soundness of classroom assessments.