Chapter 4 · Section 4 practice

Causal Questions and Research Designs

0 of 4 answered

These questions review the examples in the companion slides. Choose an answer, then check the explanation.

Question 1

“Traders who use a risk dashboard have higher risk-control scores.” Which action needs a causal answer?

  • Use adoption as a predictor of a later risk-control score
  • Report the adoption–score association
  • Require every trader to adopt the dashboard, expecting the observed improvement
  • Report the correlation in the annual review

Question 2

Add \(U\) as a second regressor. The coefficient on \(T\) will:

  • Stay near 2.17 — \(T\) is still strongly correlated with \(Y\)
  • Collapse toward 0 — the true effect
  • Flip to about −2
  • Become undefined because \(T\) and \(U\) are correlated

Question 3

The identifying assumption of DiD is:

  • The two groups have the same level of \(Y\) before treatment
  • Treatment was randomly assigned
  • Without treatment, both groups would have followed parallel trends
  • The control group is larger than the treated group

Question 4

Which condition can never be verified from the data alone?

  • Relevance — \(Z\) predicts \(T\)
  • Exclusion — \(Z\) affects \(Y\) only through \(T\)
  • That \(n\) is large enough
  • That \(T\) is binary