5.8 Associational vs. causal inference

  • Joint distribution is basis for any quantitative analysis (Holland 1986, 948; Pearl 2009)
  • Associational inference:
    • Summarize joint distribution with statistical model (e.g. regression model)
    • Does not tell us anything about causality, e.g. coefficient represents effect in both directions (Trust ↔ Victim)
  • Causal inference:
    • Summarize joint distribution with statistical model
    • Add assumptions
    • Causal interpretation!

References

Holland, Paul W. 1986. “Statistics and Causal Inference.” J. Am. Stat. Assoc. 81 (396). Taylor & Francis: 945–60.

Pearl, Judea. 2009. “Causal Inference in Statistics: An Overview.” Stat. Surv. 3. The author, under a Creative Commons Attribution License: 96–146.