Open vs Closed Book Exams in STEM Education
Author Information
Author(s): Rasi Mizori, Sadiq Muhayman, Ahmad Malik Takreem, Siu Anthony, Ahmad Reubeen Rashid, Yang Zijing, Oram Helen, Galloway James
Primary Institution: King's College London
Hypothesis
How do OBEs and CBEs differ in terms of student exam performance in STEM subjects?
Conclusion
OBEs generally result in better scores than CBEs, but the influence of large language models poses challenges to their effectiveness.
Supporting Evidence
- Meta-analysis showed that OBEs generally resulted in better scores than CBEs.
- Significant heterogeneity was observed in the results with an I2 value of 97%.
- Three studies reported no significant difference in exam scores between OBEs and CBEs.
Takeaway
Open-book exams let students use resources while testing, and they usually score better than in closed-book exams, but new AI tools make it tricky.
Methodology
This systematic review analyzed peer-reviewed articles published from 2013, focusing on the impact of OBEs and CBEs on university STEM students using a random effects model.
Potential Biases
Insufficient control for confounders in several studies and variability in exam difficulty and invigilation practices.
Limitations
The evidence base is small and primarily from three countries, and all studies were conducted before the introduction of major LLMs.
Participant Demographics
Studies were predominantly conducted in the USA (5), followed by the UK (2) and Saudi Arabia (1), covering various STEM disciplines.
Statistical Information
P-Value
<0.01
Confidence Interval
95% CI: 2.79 to 9.04
Statistical Significance
p<0.01
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website