In most surveys you will need to apply weights before analysis and there are different weights applicable depending on what variables you are using and the units of comparison. The weighting structures will vary depending on the design of the survey, for example if only one person was interviewed per household then people who live in smaller households have a greater probability of being selected than individuals who live in larger households. To understand the weighting structure you need to familiarise yourself with any hierarchies in the data structure - how many levels, are variables carried over from different levels of the hierarchy -for example, aggregates of individual household members income brought across to the household level or vice versa.

The technical documentation for the survey will tell you everything you need to know about how to weight the data. Where there are different weights that need to be applied at the same time these are sometimes presented as compound weights -a single weight variable that adjusts for probability of selection and to adjust for nonresponse or to gross up to population totals all of which may also be available as individual weights.