Population size

[Ivan Oreskovic]

I am doing statistical analysis on the data from the survey whose methodology can be found at http://www.researchictafrica.net/doc...%202011:12.pdf. The analysis is at individual-level.

However, when I declare survey design and apply weights, I obtain population size that is 50% of the actual population for given country.

How to interprete this? Does this has anything to do with the fact that they set population proportion P to 0.5?

[Ivan Oreskovic]

UPDATE: I get the same number if I use subpop() or if statement. In the case of former I get the subpopulation size and in the case of later population size. Number of observations I get is correct.

Furthermore, dataset description states "The weights are based on the inverse selection probabilities and gross up the data to national level when applied.".

What does it mean "to national level" and is it possible that weights gross up the data only to the certain amount of population?

[Ivan Oreskovic]

UPDATE 2: I found the main culprit to my problem. Stata was giving population size which referst to the eligibile population for the survey (age 15 or older) and I was comparing this to the whole population which is obviously much larger.

I can only say to myself: well done Sherlock.

[Jesse Wursten]

Haha, this is a beautiful thread to read as an outsider :D