Sample size

Sample size estimation was based on the minimum desirable change in knowledge score. To detect an effect size difference of 0.67 in a simple random sample with a power of 80% and significance level of 95%, 74 subjects were required. This figure was inflated to 148 to allow for randomisation by cluster (rho = 0.05) and 20% attrition.

Flow of participants

Access jobs at BMJ Careers
Whats new online at Student BMJ