SAMPLING WEIGHT DHS

[natiesim]

Hello, Please I need your suggestions. I want to use data of DHS (demographic health survey) to analyse the effect of the consommation of a product on the probability to work. Unfortunately, I have only 100 people who consume the product. I thought I could use the sampling weight to allow my 100 people to be representative of the country population. However when I use tab var [aw=hv005] (where hv005 is the household weight) I have about 300 000 000 people who consume the product, that doesn t make sense since the real population of the country is about 11 000 000. when I use the recommandation of the DHS material gen weight=hv005/1 000 000; tab var [aw=weight] I get less than 100 people who consume the product. so I would like to know how I can use sample weights in my descriptives statistics to have a representativeness of the population with my few data. Thanks.

[Steve Samuels]

Unfortunately the sum of sample weights provided by DHS do not sum to population totals. This is a great loss, especially for questions such as yours. I do have one question: If your response is the probability that a person works, why are you using a household weight rather than a person weight?


Even though your subpopulation is very small, you should not exclude others from the data set. Instead, specify the subgroup to survey analyses with a subpop() option.
e.g.

sysuse auto,clear
svyset _n [pw = turn]
svy, subpop(if foreign=0): mean mpg

Note that Statalist protocol is to register with a user name that consists of a first and last name. See FAQ 6 for more and instructions on how to change your user name.