I’m including only those in the Sample Adult subset and have restricted the dataset by using ASTATFLG.
-
I’m exploring risk factors at person level. Is it appropriate to use the perweight while considering the sample adult dataset alone
-
I will be pooling the data for a 5 year period. Should I use the same sampling adjustment (perwieght/5) when using only this sample adult dataset.
Thanks for your time.
The sample weight SAMPWEIGHTis designed based on the sample adult and sample child sampling framework. Therefore, this weight should be used when analyzing variables from the sample adult or sample child supplements of the IPUMS NHIS data. Regarding adjusting the weight for pooling samples together, your method is certainly reasonable. Technically, you should multiply the weight by the sample size of sample i divided by the total pooled sample size. In NHIS data, this ratio will be quite close to 1/n if you are pooling n samples together.