To produce estimates appropriately adjusted for survey non-response it is important to check all of the variables in your analysis and select the weight of the smallest analysis subpopulation.
All weights can be found on data files for the respective survey component (e.g. Anthropometry, Medical History, Biochemistry, etc.). In general, it is recommended that you select the weight variables for the 1-65 stands full sample, or the 1-100 stands detailed sample, based upon the variables used for your analyses. If you use variables from the detailed examination, then you MUST use the weights for the detailed sample. However, if you require certain variables which were only available for one of the smaller subsamples then you must choose the appropriate weight for that sample.