Block 3: Two-sample tests
Week 12 · Uncertainty II · Imai §7.2.4 · about 3 min
NoteHow this works
Run each cell with Run Code, look at the output, then answer the question at the end of the block. Every cell is editable; Start Over resets it. Errors are information: read them.
The resume experiment (Weeks 3 and 11): callbacks for white vs. black names. \(H_0\): equal callback probabilities.
1. prop.test()
Careful: prop.test counts the first column (call = 0, no callback) as the “success”.
2. By hand
Conclusion: How do the by-hand p-value and the one from prop.test compare?
By hand: difference 0.032, s.e. 0.0078, z = 4.11, one-sided p = 2.0e-5. prop.test reports 2.5e-5: it applies a continuity correction by default (correct = FALSE gives our value). Strong evidence against equal callback probabilities; with randomized names the natural conclusion is discrimination.