Introduction to Research Methods in Political Science:
In working with sample data, investigators will sometimes include a “weight” variable to correct for known over or under representation in the sample. Sometimes the over or under representation will have been done deliberately, as when a small group is over sampled in order to provide enough cases to permit reliable analysis of that group. Sometimes the over or under representation will be discovered after the sample is taken and compared with some known characteristics of the population from which it was drawn. For example, when a survey of public opinion is carried out, it may turn out that the sample is not representative (in terms of age, race, gender, region, etc.) of the overall population as measured in a recent census.
A weight variable can be created using COMPUTE statements. Often, a dataset you are using will already include one or more weight variables.
To weight data by such a variable, click on “Data” (in the menu bar) and “Weight Cases….” Select “Weight cases by,” select a variable in the left window, click on the right arrow, and click on OK.”
April 28, 2013 .
© 2003---2013 John L. Korey. Licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.