Weighting

Weighting involves counting a respondent as either more than one person or less than one person. In the 2016 ANES, the weights are used to correct for a higher or lower chance of a respondent being included in the sample than he or she would have by chance alone. In addition, the weighting attempts to increase the sample's representativeness on demographic characteristics. For the most part, this weighting is transparent: you will not notice that it is happening at all. However, there are times that weighting can result in fractional respondents causing some contingency table marginals or frequency distributions not to add up exactly. This is a result of rounding error in the weighting of respondents.