Sign-ups rise with ad spending, but both peak every December. What should you suspect?
Why D is the answer
Season is a confounding factor
A third factor can drive both variables, so correlation alone does not prove cause.
In detail
- A third factor can drive both variables.
- Correlation alone does not prove cause.
- Correlation alone does not prove cause.
Use case
An illustrative scenario, not a real incident.
Ads and sign-ups both rise in December, but holiday demand may drive both.
See it in different ways
mindmap
root((Data analysis))
Metrics
Loss
R-squared
Charts
Line: over time
Scatter: two variables
Care
Same data
Confounding factorsmindmap
root(("December season"))
n1["Drives"]
n2["Ad spending"]
n3["Sign-ups"]Key point: Look for a confounding factor.
Why the other options are wrong
- A. Ad spending must be the cause of all sign-ups
- That jumps from correlation to cause.
- B. The tokenizer is splitting words incorrectly
- Tokenization is unrelated.
- C. The dataset contains too few December rows
- The peak is in December, so those rows exist.