Understanding Correlation vs Causation: Linking and Lurking Variables

This lesson explores the difference between correlation and causation, emphasizing that correlation does not imply causation. It explains the correlation coefficient and common statistical errors, using examples like stress and ulcers, and beach rescues and temperature. The lesson also introduces linking and lurking variables, illustrating how they can affect perceived relationships between variables. Finally, it highlights the importance of experiments in establishing causality.

What is a common mistake people make when interpreting statistical data?

• Assuming correlation implies causation

• Ignoring the correlation coefficient

• Assuming all data is normally distributed

• Believing all variables are independent

In the example of beach rescues and temperature, what is the linking variable?

• The number of lifeguards

• The number of people swimming

• The time of day

• The type of beach

What is a lurking variable in the example of reading scores and shoe size?

• Height

• Weight

• Age

• Gender

Why can't we conclude that chemical exposure causes illness based on correlation alone?

• Because chemicals are always harmful

• Because there might be a linking or lurking variable

• Because correlation always implies causation

• Because illness is not quantifiable

What will you learn in a later lesson regarding establishing causality?

• How to interpret statistical graphs

• How to design experiments

• How to calculate correlation coefficients

• How to identify lurking variables