Comparing Objective and Essay-Type Tests in Measuring Student Achievement: A Psychometric Meta-Analysis and Empirical Evaluation

Abstract

The optimization of educational assessment tools is essential for accuratelymeasuring student achievement and defining structural educational standards. Educational researchers have long debated the competing advantages of objective tests (such as multiple-choice and matching questions) and essay-type tests (such as extended response and prompt-driven essays). This paper presents a comprehensive mixed-methods psychometric evaluation comparing objective and essay-type modalities across a sample size of 420 secondary and undergraduate students over a full academic semester. Using classical test theory (CTT) and item response theory (IRT), we analyze the construct validity, content validity, grading reliability, and cognitive domain depth of both formats based on Bloom's Revised Taxonomy. Quantitative metrics indicate that while objective assessments exhibit superior internal consistency (Cronbach's alpha = 0.88) and high breadth of content coverage, essay-type tests demonstrate significantly higher measurement efficiency for higher-order cognitive skills, specifically evaluation and creative synthesis (p < 0.001).Longitudinal predictive modeling shows that scores on essay-type tests arebetter indicators of long-term real-world problem-solving abilities thantraditional objective test scores (beta = 0.52 vs beta = 0.22). Qualitative text mining of grader responses shows that essay scoring remains vulnerable to inter-rater variance and grader fatigue, even when structured rubrics are used. This paper outlines an integrated, balanced testing framework that helps academic institutions combine both formats to optimize psychometric reliability and cognitive depth. KEYWORDS: Objective Assessments; Essay-Type Examinations; Psychometric Reliability; Bloom’s Taxonomy; Classical Test Theory; StudentAchievement Measurement.