Confounding Variables Statistics Example 2

Follow the full solution, then compare it with the other examples linked below.

Example 2

medium
Hospitals A and B both perform a surgery. Hospital A has a 90% survival rate, Hospital B has 95%. However, Hospital A takes more high-risk patients. When only high-risk patients are compared, Hospital A has a higher survival rate. Explain this paradox.

Solution

  1. 1
    Step 1: This is Simpson's Paradox โ€” the overall rates reverse when looking at subgroups because of a confounding variable (patient risk level).
  2. 2
    Step 2: Hospital A takes more high-risk patients, which lowers its overall survival rate. But within each risk category, Hospital A actually performs better.
  3. 3
    Step 3: The confounding variable (patient risk level) is unevenly distributed between hospitals, distorting the overall comparison.

Answer

This is Simpson's Paradox. Patient risk level is a confounding variable โ€” Hospital A treats more high-risk patients, lowering its overall rate. Within each risk category, Hospital A actually performs better.
Simpson's Paradox occurs when a trend that appears in overall data reverses when the data is split into subgroups. It arises because a confounding variable is unequally distributed across groups. This paradox demonstrates why controlling for confounders is essential for valid comparisons.

About Confounding Variables

A confounding variable is a third variable that influences both the independent variable and the dependent variable simultaneously, creating a spurious association between them that can be mistaken for a direct causal relationship. Confounders are a major threat to the internal validity of observational studies.

Learn more about Confounding Variables โ†’

More Confounding Variables Examples