Bloom-Blueprinted Multiple-Choice Assessment in First-Year Chemistry: Item Performance, Attainment, and Student Perceptions

Abstract Chemistry departments increasingly require summative assessments that are scalable, reliable, and capable of evaluating more than factual recall. This study examines an assessment transition from traditional Short-Answer Questions (SAQs) to Bloom-blueprinted Multiple-Choice Questions (MCQs) across a large-cohort first-year undergraduate chemistry sequence tracked over multiple academic cycles. MCQs were designed using a deliberate 4:3:3 blueprint mapped to Bloom’s taxonomy, with 40% lower-order cognitive skills (LOCS), 30% intermediate application and analysis, and 30% higher-order cognitive skills (HOCS)-targeted items. Across successive cohorts, item-level analyses showed a consistent decrease in attainment as intended cognitive demand increased, providing evidence consistent with the intended tiered design rather than direct evidence of students’ reasoning pathways. Overall module-level performance remained broadly within historical ranges, although selected-response formats were associated with narrower score distributions and an initial increase in mean performance in the transition year. Student survey data added an important qualification: students viewed SAQs as better for demonstrating conceptual understanding and deeper reasoning, even when they did not reject MCQs overall. The findings support the conditional claim that carefully designed MCQs can produce reliable, discriminating, cognitively tiered assessment tasks at scale, but their validity depends on careful blueprinting, distractor design, linguistic accessibility, and post-examination review.