Modify Survey Weights
See Also
This menu can be used to modify survey weights for particular observations, adjusting the other weights in the sampling unit or stratum to ensure that the overall sum of the weights remains unchanged.

Two common problems in survey data are item non-response (i.e. a missing value for one question although valid responses are present for others) and outliers. For an item response that occurs entirely at random, one method of dealing with it is to analyse the question with a modified set of weights with the weight for the missing observation redistributed over the rest of the units in the stratum or sampling unit.

Outliers

Specifies the units whose weights are to be adjusted. These should be supplied as a list separated by commas.

Input Weights

This is the variate containing the weights that are to be modified.

Save new weights

Provides a space to supply the name of an identifier to save the modified weights within.

Stratification factor

For a stratified random survey, this is the factor indicating the stratum to which each unit belongs. For unstratified designs this should be left blank.

Sampling units

For single stage designs this should be left blank.

Labels

This is an optional variate, factor or text which is used to label influential points. If it is left blank points are labelled by the numbers 1...n.

Available Data

This lists data structures appropriate to the current input field. The contents will change as you move from one field to the next. Double-click on a name to copy it to the current input field; alternatively, you can type the name directly into the input field.

See Also