Biostatistics
Biostatistics tests selection and interpretation of statistical tests in pharmaceutical and clinical research. The exam focuses on assumptions, error rates, and effect estimates, not hand calculation alone.
Core rules
- Descriptive statistics summarize location and spread; standard deviation describes variability of observations, while standard error of the mean describes precision of a mean.
- Hypothesis tests: type I error is rejecting a true null; type II error is failing to reject a false null. Power equals 1 minus type II error.
- Test choice: independent t-test for two independent normal means; paired t-test for before–after; ANOVA for three or more means; chi-square for categorical association; nonparametric tests when assumptions fail.
- Confidence intervals convey effect size and precision. A 95% CI excluding the null corresponds to p<0.05 in a two-sided test.
Worked example
A trial reports a mean difference of 2.1 mmHg with 95% CI −0.4 to 4.6. Reasoning: the CI includes 0, so the result is not statistically significant at alpha 0.05. Do not conclude no effect; the estimate is imprecise and a clinically important benefit is not excluded. Report the CI and the sample size.
Traps
Interpreting p as the probability the null is true; using standard deviation instead of standard error in error bars; ignoring multiple comparisons; treating a small p as a large effect.