Research Methods
Core concepts
Understand study designs, including randomized controlled trials, cohort studies, case-control studies, cross-sectional surveys, and qualitative methods. Know the concepts of validity, reliability, bias, confounding, randomization, blinding, and control groups. Be able to identify sampling methods and the difference between probability and nonprobability sampling. Familiarity with ethical principles such as informed consent and institutional review is expected.
Traps
A frequent error is confusing internal and external validity: internal validity concerns whether the study correctly establishes cause and effect, while external validity concerns generalizability. Another trap is assuming that randomization eliminates all bias; it addresses confounding but not measurement bias. Questions may also test whether a design can establish temporality; cross-sectional studies cannot establish that exposure preceded outcome.
Worked example
A study finds that people who drink coffee have lower rates of a disease. If the study measured both at one time, it is cross-sectional and cannot establish causation. Rule of thumb: temporality requires a cohort or trial design.
Study tip
For each design, ask three questions: can it establish temporality, can it estimate incidence, and is it prone to recall bias?
Biostatistics and Data Literacy
Core concepts
Understand descriptive statistics, including measures of central tendency and dispersion, and the normal distribution. Know the meaning of p-values, confidence intervals, statistical significance, and the difference between type I and type II errors. Familiarity with sensitivity, specificity, predictive values, and basic tests such as t-tests and chi-square is expected. Be able to interpret a graph or table critically.
Traps
A common error is interpreting a p-value as the probability that the null hypothesis is true; it is the probability of observing data at least as extreme if the null hypothesis is true. Another trap is confusing statistical significance with clinical importance; a large sample can make a trivial difference significant. Questions may also test sensitivity versus specificity: sensitivity is the ability to detect true positives, while specificity is the ability to detect true negatives.
Worked example
A test with 90% sensitivity and 95% specificity applied in a population with low prevalence will have a low positive predictive value, because most positive results are false positives. Rule of thumb: predictive values depend on prevalence, while sensitivity and specificity do not.
Study tip
Before interpreting any statistic, identify the sample size, the comparison group, and the outcome measure.