liveBook · Manning

Correlation

As explained in Hidden Influences

A statistical association in which two variables vary together without by itself establishing that one causes the other.

When people see more positive messages online, they tend to post more positive messages themselves; the same pattern appears with negative messages. This relationship is correlation: two variables vary together without proving that one causes the other. section 7.1

Correlation alone cannot show why the pattern exists. People may genuinely feel different, imitate the style they encounter, or be affected by another factor. The study measured sentiment through words, so negations such as “not joyful” could be misclassified. section 7.1

A careful interpretation keeps several limits in view:

  • Treat observed exposure and later behavior as a relationship, not proof that exposure caused the behavior. section 7.1
  • Check effect size: reducing negative posts lowered negative-word usage by only B=−0.07%, possibly a negligible amount. section 7.1
  • Question how variables were measured, especially when sentiment analysis ignores negation. section 7.1

How Hidden Influences covers it

How deep it goes
Defines correlation, distinguishes it from causation, and examines effect size and sentiment-measurement limitations.
Examples
Uses positive and negative social-media posts, a B=−0.07% effect, and the phrase “not a joyful day.”
What you can do after
Evaluate observed associations by separating correlation from causation and checking effect size and measurement quality.
How it connects
Connects correlation with causation, social contagion, sentiment analysis, and A/B testing.

Also explained in

Compare how 22 books explain correlation →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Hidden Influences. The section links point to the parts of the book it draws on.