Block 4: Survey sampling and the list experiment
Week 5 · Measurement I · Imai §3.4 · about 4 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.
1. Are the sampled villages representative?
204 of 1,864 villages were sampled (village.surveyed = 1). Compare them with the others.
2. The list experiment
Control respondents see 3 items; the other groups see the same 3 plus one sensitive item (harmed by ISAF, or by the Taliban). They report only how many items apply. The difference to the control group estimates the share for whom the sensitive item applies.
Conclusion: Compare the list experiment with the direct question (ISAF 37%, Taliban 33%). What do you see?
List experiment: about 5% for ISAF and 22% for the Taliban, far below the direct question. Different measures give different answers; the sampled and non-sampled villages, however, look alike on altitude.