I found that Mexico 1995 has 1,243 cases with zero weights. The same for Venezuela 1971, 1,315 cases. I wonder, what might be the explanation for having zero weights, considering these cases are larger than in Brazil 1980.
Many Thanks,
Alejandro
Seeing weight values of zero in the data is common. Many samples have legitimate weights of zero, which are assigned during the stratification process. This often signals demographic characteristics that are over-represented in the sample data. Note that most of the zero weight cases in the Venezuela 1971 sample are men living in group quarters in the Federal District. Similarly, most zero weight cases in the Mexico 1995 sample are men living in group quarters.
For both Mexico and Venezuela, the weights were created by the national census agency; therefore, we do not have any reason to assume these zero weights are not legitimate. You might try contacting the census agencies of these countries for further information on why zero weights were assigned.
Hope this helps.
Thank you. It is helpful.