Logan Hanks
Methods Glossary
A plain-language reference for the epidemiology and population-science methods behind cancer-prevention research. Each entry gives you a direct answer you can act on — plus deeper context when you need it. New terms added as the work demands.
C
- Causal InferenceCausal inference is the set of methods and assumptions used to estimate the effect that changing an exposure would have on an outcome, as distinct from mere statistical association.
- Collider BiasCollider bias is a spurious association induced between two variables when an analysis conditions on (stratifies, adjusts for, or selects on) a common effect of both.
- Community-Based Participatory Research (CBPR)Community-based participatory research (CBPR) is an approach that engages community members as equal partners with researchers across all phases of a study, sharing decision-making, data, and benefits.
- Confidence IntervalA confidence interval is a range of plausible values for a population parameter computed so that, under repeated sampling, a stated percentage (e.g. 95%) of such intervals would contain the true value.
- Confounder AdjustmentConfounder adjustment is the use of statistical or design methods — such as stratification, regression, matching, or weighting — to remove the distortion a confounder introduces into an exposure–outcome estimate.
- ConfoundingConfounding is a distortion of an exposure–outcome association caused by a third variable that influences both the exposure and the outcome.
D
E
H
I
L
N
O
P
R
- Relative RiskRelative risk (RR), also called the risk ratio, is the ratio of the probability of an outcome in an exposed group to the probability in an unexposed group.
- Risk DifferenceThe risk difference (RD), or absolute risk reduction, is the arithmetic difference in the probability of an outcome between an exposed and an unexposed group.
S
- Screening vs. Diagnostic TestingScreening tests identify asymptomatic people who may have a condition so they can be evaluated further, whereas diagnostic tests confirm or rule out disease in people with symptoms or a positive screen.
- Selection BiasSelection bias is a systematic error that arises when the people included in (or retained by) a study differ from the target population in ways related to both exposure and outcome.
- Sensitivity and SpecificitySensitivity is the proportion of people with the disease who test positive; specificity is the proportion of people without the disease who test negative.
- Social Determinants of HealthSocial determinants of health are the non-medical conditions in which people are born, grow, live, work, and age — such as income, education, housing, and access to care — that shape health outcomes.