Case-based learning improves exam performance¶

Claim · All claims
Evidence · 2 studies · 1 causal r2, 1 review r3 · q2–q3
Currency · newest study 2015 · checked against its sources 2026-10-20 (automated, 2 of 2 entries) · not reviewed by a person

This claim concerns case-based learning — instruction organized around the analysis of realistic, contextualized cases — and its effect on performance on course examinations relative to conventional instruction.

Subclaims¶

q2 i? In one introductory biology cohort at a community college, exam questions on topics taught through case studies scored higher than matched control questions on topics taught by discussion and textbook reading (an average difference of 18 percentage points; no standardized effect size reported). → Bonney 2015

q3 i? A systematic review of 104 papers on case-based learning in health professional education found that students enjoy it and believe it helps, but the empirical data as a whole are inconclusive about its effect on learning compared with other activities. This heavily qualifies the claim. → Thistlethwaite et al. 2012

Evidence¶

Bonney 2015¶

Bonney, K. M. (2015). Case Study Teaching Method Improves Student Performance and Perceptions of Learning Gains. Journal of Microbiology & Biology Education, 16(1), 21–28. doi:10.1128/jmbe.v16i1.846

q2 · quasi-experiment, within-cohort comparison of matched topics, not randomized · i? · no standardized effect size reported; mean case-topic scores 70–79% vs control 52–60% · n=56 students · causal · r2

Fifty-six consenting students in a first-semester general biology course at Kingsborough Community College (CUNY) were taught four topics (chemical bonds, osmosis and diffusion, mitosis and meiosis, DNA structure and replication) through case studies and four other topics through short lectures, textbook reading, worksheets and discussion, with equal class time for each. On the same regular course exams, case-taught topics scored higher than control topics matched for format, difficulty and point value, a statistically significant difference for each of the four cases. Case studies written by the course's instructor did no better than published ones by other instructors. The comparison is between different topics rather than randomized groups, so topic difficulty could still differ despite the matching, and the study comes from one instructor.

Thistlethwaite et al. 2012¶

Thistlethwaite, J. E., Davies, D., Ekeocha, S., Kidd, J. M., MacDougall, C., Matthews, P., Purkis, J., & Clay, D. (2012). The effectiveness of case-based learning in health professional education. A BEME systematic review: BEME Guide No. 23. Medical Teacher, 34(6), e421–e444. doi:10.3109/0142159X.2012.680939

q3 · systematic review, narrative synthesis · i? · no pooled effect size; the data were judged not amenable to meta-analysis · n=104 papers · review · r3

A Best Evidence Medical Education review of case-based learning in prequalification programmes in medicine, dentistry, veterinary science, nursing, social care and the allied health professions, searching six databases from 1965 to September 2010. Of 104 coded papers, 23 were rated higher quality. Most studies used a single cohort (61%) and evaluated outcomes only after the intervention (75%), and the variety of interventions ruled out meta-analysis. The authors conclude that students enjoy case-based learning and think it helps them, but that the evidence is inconclusive on whether it improves learning more than other activities, and that any benefit of small-group case work may come from the group learning rather than from the cases.

Studies bearing on this claim¶

Does teaching through realistic cases, by itself, raise learning? This table gathers the studies behind this claim and the 2 narrower claims that bear on it: 4 distinct studies (3 review, 1 causal), q2–q3, 0 of 4 with an effect size coded. One row per study, codes as each entry records them; no effect is averaged and no verdict is drawn. Built by scripts/synthesize_claim.py from eval/synthesis/; a row that looks wrong is fixed on the claim it comes from.

Discussion¶

Why cases might help exam performance. Cases situate abstract principles in concrete, realistic contexts, which is the core move of situated learning and cognitive apprenticeship. Working through cases requires learners to activate prior knowledge, discriminate relevant from irrelevant detail, and apply concepts to messy situations — processes associated with deeper encoding than passive review. Structured case discussion formats such as the Harvard method add accountability and argumentation, which plausibly strengthens the retrieval routes that exams test.

Boundary conditions to expect. The effect is unlikely to be uniform. Plausible moderators include:

• Assessment alignment. Cases should pay off most on application, analysis, and transfer items; on pure factual-recall exams, case-based instruction may show little advantage over direct instruction. Any evaluation of this claim must therefore match case-based formats to outcome measures that test application, not just recall.
• Learner expertise. Novices may lack the schemas needed to extract principles from an unstructured case, consistent with the expertise reversal effect; scaffolding or worked structure within the case matters more for beginners. For novices, a fully guided alternative such as worked examples may outperform open-ended case work at lower cost.
• Case quality and multiple cases. A single rich case can anchor one concept but may encourage overgeneralization; exposure to multiple varied cases, as argued by cognitive flexibility theory, is the standard remedy for ill-structured domains.
• Process support. Unstructured case discussion can degenerate into opinion exchange; pairing cases with activation prompts, structured roles, or collaborative learning routines is likely necessary for gains.

Open questions. Whether exam gains persist at delay, whether they generalize across professional and non-professional domains, and how case-based formats compare with cheaper alternatives such as worked examples all remain to be established. Two sources are recorded: a non-randomised within-cohort comparison in one community-college biology course (Bonney 2015, n=56), which favoured case-taught topics, and a systematic review of 104 health-professions papers (Thistlethwaite et al. 2012) that judged the evidence on learning inconclusive. No randomised comparison and no standardised effect size are recorded, so the size of any exam benefit, and its direction outside the one course studied, are not established.

Related Claims¶

• Active learning improves exam performance. — case discussion is a structured form of active learning; this claim situates cases within that broader evidence base
• Cognitive flexibility theory: multiple cases support transfer in ill-structured domains. — explains why multiple, varied cases rather than a single exemplar drive transfer
• Collaborative learning improves outcomes. — most case-based formats are collaborative; the group process may mediate exam gains
• Activation improves learning. — effective case analysis depends on learners activating relevant prior knowledge before and during case work
• Discussion promotes more active thinking than lecture, though lecture is no worse for transmitting information — related
• Interactive LRE methods engage students, who value LRE classes as relevant and interesting — related
• Mastery Learning Improves Outcomes — related
• In one NVCC summer bridge cohort, 100 percent of students continued to the next semester, described as much higher than normal retention rates — related
• A flexible hypertext course group averaged 76% on achievement tests against 59% for a direct-design group that studied different software, so the contrast confounds design with content — related
• For introductory biology students with low topic-specific prior knowledge, problem-solving-first (PS-I) outperformed instruction-first (I-PS) on a near-transfer test — related
• Elaborative interrogation improved true/false test performance in an introductory biology course context — related