Predictive validity and assessment stringency of pre-board examinations: a three-year longitudinal study of cbse class 10 students

Abstract: Pre-board examinations play a vital role in Indian secondary schooling by assessing students’ readiness for the Class 10 board examination conducted by the Central Board of Secondary Education (CBSE). Despite their widespread use for academic planning, remediation, and counselling, empirical evidence supporting their predictive validity in the Indian context remains limited. This study examines the extent to which Pre-board examination scores predict final board examination performance across three consecutive academic cohorts (2023–2025) within a CBSE-affiliated school. A quantitative longitudinal design was employed using anonymised average scores from five core subjects for 223 students. Data were analysed using descriptive statistics, Pearson and Spearman correlation coefficients, linear regression models, paired-sample t-tests, and score distribution analyses. Across all cohorts, board examination scores consistently exceeded Pre-board scores by 9–15 percentage points (p < .001), indicating greater stringency in internal assessments. Strong and stable associations were observed between Pre-board and board scores (Pearson r = 0.89–0.93; Spearman ρ = 0.90–0.94). Regression analyses further demonstrated that Pre-board performance accounted for approximately 80–86% of the variance in board examination outcomes. The findings indicate that, although Pre-board examinations systematically underestimate final board performance, they reliably predict relative academic standing and function as effective diagnostic tools at the school level. The study highlights important implications for the interpretation of Pre-board results, school-based assessment practices, and alignment with the assessment reforms emphasised in the National Education Policy (NEP) 2020. Mean differences ranged from 9.07 to 14.95 percentage points (p < .001; Cohen's d = 1.2–1.6). This provides the first longitudinal school-level evidence of Pre-board predictive validity in CBSE Class 10 assessments.