Hello,

I am working with the Population Assessment of Tobacco and Health data. It is a panel survey with three waves of data. I have applied balanced replication weights using svyset and then merged all three waves in a wide data.frame. However, I am finding that after merging, I get different results for svy: tab for wave 2 then I did when wave 2 was in a separate dataset. However, the first wave is exactly the same. Did something go wrong when I merged the second wave to the first wave?

Thank you for your help. I tried my best to explain everything, but please let me know if you would like any more information.