Enhancing exam question quality in medical education through bootstrapping

Area of Science:

• Medical Education
• Quantitative Psychology

Background:

• Medical schools face challenges in curriculum and exam evaluation due to small sample sizes.
• Traditional methods for assessing educational programs require extensive data collection, which is inefficient.

Purpose of the Study:

• To demonstrate the utility of advanced quantitative methods, specifically bootstrapping, for curriculum and assessment evaluation.
• To provide a more efficient and reliable approach to psychometric analysis in medical education.

Main Methods:

• Utilizing bootstrapping, a resampling technique, to generate multiple simulated samples from a single dataset.
• Applying these simulated samples to create confidence intervals for evaluating exam questions and curricula.

Main Results:

• Bootstrapping provides reliable data for assessment evaluation, even with small initial sample sizes.
• This method allows for the creation of confidence intervals, enhancing the accuracy of evaluations.

Conclusions:

• Bootstrapping offers a robust and economical alternative to traditional psychometric evaluation methods.
• This technique improves the psychometric quality of exam questions and contributes to fair, valid assessments in medical education.