Causation, or causality, describes a relationship in which one event, factor, or variable produces a change in another. Statistical analysis must distinguish genuine influence from patterns created by randomness, association, or regression to the mean. section 16.3 section 2.5.4
Use these checks before treating an observed relationship as causal:
- Do not infer influence merely from a strong correlation; correlated continuous variables may move together without one affecting the other. section 2.5.4
- Compare an apparent pattern with independent random trials, because randomness can create sequences that look orderly or predictive. section 16.3
- Expect nonlinear effects when causes and outcomes do not follow a simple proportional relationship; Pareto charts can help prioritize corrective action. section 15.5
NBA variables and apparent influence
A correlation matrix can reveal that games played and win shares are strongly related, such as MP and WS48 correlating at about 0.66. That coefficient describes linear association, not proof that one variable causes the other. Select continuous variables, compute pairwise coefficients with cor(), and visualize them in a ggplot2 heat map. section 2.5.4
sample(x = c(0, 1), size = 10, prob = c(0.5, 0.5), replace = TRUE) ## [1] 0 1 1 1 1 1 1 0 1 0
Simulate fair, independent coin flips with sample() to see how random sequences can resemble meaningful streaks without establishing causation. section 16.3
