Farid Asadi

Growth, Experiment Driven.

Confounding Variables When You Analyze

Posted on

When you review the data, you discover that higher consumption of ice cream is associated with a higher rate of sunburns. But does this mean eating ice cream causes sunburns?

This is where the confounding factor comes into play. Confounding factors cause spurious associations between dependent (sunburns) and independent variables (ice cream).

In this case, the hot temperature is the confounding factor. In hot weather, both ice cream consumption and sunburns increase. In a spurious association, however, the variable or events (ice cream and sunburns) are correlated and NOT causally connected.

Thus, it is crucial to:

  • Always look for the factors that are not shown.
  • Control the confounding factor.
  • Make use of causality diagrams, and be aware of confounding factors.

Read On

  1. Value–Action Loops in Growth; How Value Creation Generates User Decisions?
  2. Strategy Experimentation: How to Choose a Strategy?
  3. You Lose Your Focus Here!
  4. Price–Information Relationship in CRO
  5. Applying The Loss Aversion Principle on Pricing Pages
  6. Why Google Analytics 4 Can’t Go Far?
  7. Best Practices in Conversion Optimization