4.11 Precision of causal statements

  • Precision is everything!
  • Three examples of informal “because” statements (Imbens and Rubin 2015, 3, 4–5)15
    1. “My headache went away because I took an aspirin”.
    2. “She got a good job last year because she went to college.”
    3. “She has long hair because she is a girl.”
  • Q: Go through the examples. How well are treatment/control defined (and the outcome)? What is the (non-)action that the treatment corresponds to? Who is acting?
    • Lesson: Try to formulate the treatment in terms of an intervention/action!
  • Q: What actions/interventions could we think of for the treatment variable race (e.g., black vs. white)? (Imagination!)16

References

Imbens, Guido W, and Donald B Rubin. 2015. Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction. Cambridge University Press.

  1. 2. Less clear what treatment and alternative are. What is the alternative value to “she went to college”? 3. Alternative to “girl” not at all clear. What is the acion associated? To define a causal question properly, try to frame the treatment as an intervention that happens at a particular moment in time! ↩

  2. e.g., controversial film by Günther Wallraff.↩