I am using Mexico 2010 census data to do an individual-level analysis.
Upon reading the available files about survey design for Mexico 2010, I found the following information.
- It was a one-stage stratified cluster sample by the municipality.
- Enumeration areas (blocks of dwelling within a locality) selected by simple random sampling with strata.
- A sample unit is specified as dwellings(household).
- Clustering at both the municipality and household level.
Code:
svyset serial [pweight=person_weight], strata(municipality) vce(linearized) singleunit(missing)
0 Response to Census data use of weights and survey-design
Post a Comment