Regression analysis
Modeling how an outcome changes with one or more predictors.
What it means
Regression analysis is a family of statistical methods that model the relationship between an outcome variable and one or more predictors, estimating how the outcome is expected to change as each predictor varies. Its coefficients summarize associations, optionally adjusted for other variables in the model, and form the backbone of prediction and statistical control across the sciences. A coefficient is causal only under strong assumptions about which variables are included and excluded; absent those, it describes association, not effect. Variants extend the idea to binary outcomes (logistic), counts, hierarchical data, and beyond, but all share the risk of misinterpretation when correlation is read as causation.
Examples
Predicting house price from square footage, location, and age, and reading each coefficient as the price change per unit holding the others fixed.
A hospital models recovery time against age, treatment and other conditions, reading the treatment coefficient as the days saved once the rest are held fixed.
A team regresses weekly sales on advertising spend and finds a strong coefficient — but spend always rises before Christmas, so the model is partly measuring the season, not the ads.
First described in Galton (1886); Pearson; Fisher.