Biostatistics
Core concepts
Know the difference between descriptive and inferential statistics, the meaning of a confidence interval (a range of plausible values for the parameter, reflecting precision, not probability that the true value lies inside), and the logic of hypothesis testing (p-value = probability of data at least as extreme assuming the null is true). Understand type I and type II error, power, and how sample size, effect size and variance drive power.
Choosing a test
Match the test to the data: t-test for comparing two means, chi-square for categorical associations, ANOVA for three or more means, non-parametric alternatives when assumptions fail, correlation and regression for relationships. For survival data, know Kaplan-Meier curves and the log-rank test. For diagnostic accuracy, be able to build and interpret a 2x2 table and calculate sensitivity, specificity, positive and negative predictive values, and likelihood ratios - remembering that predictive values depend on prevalence while sensitivity and specificity do not.
Traps
Confusing statistical significance with clinical importance; interpreting a p-value as the probability the null is true; assuming a non-significant result proves no effect (absence of evidence is not evidence of absence); and misreading confidence intervals that cross the null. Watch for multiple comparisons inflating type I error.
Worked example
Sensitivity = TP/(TP+FN); specificity = TN/(TN+FP). If a test has sensitivity 90% and specificity 95% and prevalence is 1%, the positive predictive value is only about 15% - most positives are false positives. Always recompute PPV when prevalence changes.