TOG Review (RCOG Journal)
MRCOG Part 2
Teaching and Research
A practical introduction to medical statistics
Key Statistics
- P<0.05 — Conventional significance threshold
- ±1.96×SE — 95% CI = effect size ± this
- 80% — Typical target power (1−β)
Don't-miss facts
-
A statistically flawed study is an ethically flawed study
-
95% CI = effect size ± 1.96 × SE; SE = SD/√n (uncertainty in estimate, not spread of data)
-
P<0.05 conventional threshold – P-value is the probability of the observed (or more extreme) result assuming H0 true, NOT the probability H0 is true
-
Type I error = α (reject true H0); Type II error = β (accept false H0); Power = 1−β, usually target 80%
-
Statistical significance ≠ clinical importance – always ask about effect size, not just P-value
-
OR ≈ RR only when event prevalence <10%; OR overestimates RR at higher prevalence
-
Fisher's exact test for small samples/single-digit cell counts; chi-squared otherwise
-
Worked example: smoking & low birthweight – RR 1.61 (95% CI 1.06–2.44), OR 2.02 (95% CI 1.08–3.77)
Parametric vs non-parametric (2 independent groups)
Parametric: t test (ratio/interval, normal) → effect size + CI + P
Non-parametric: Mann–Whitney U/Wilcoxon rank-sum (ordinal/non-normal) → P-value only
Abbreviation Key
- SD — standard deviation
- ANCOVA — analysis of covariance
- SE/SEM — standard error (of the mean)
- MWU/WRS — Mann–Whitney U / Wilcoxon rank-sum test
- CI — confidence interval
- IQR — interquartile range
- RCT — randomised controlled trial
- α — probability of a type I error
- OR — odds ratio
- β — probability of a type II error
- RR — risk ratio (relative risk)
Source: Scally AJ. A practical introduction to medical statistics. The Obstetrician & Gynaecologist 2014;16:121–8. doi:10.1111/tog.12081
MRCOG AI is an independent educational tool. It is not affiliated with, endorsed by, or connected to the Royal College of Obstetricians and Gynaecologists (RCOG).