Comparing a rate in a subpopulation to the rate in the full population: How it may be done when using survey data, and available software tools - ARCHIVED

Articles and reports: 12-002-X20050018030


People often wish to use survey micro-data to study whether the rate of occurrence of a particular condition in a subpopulation is the same as the rate of occurrence in the full population. This paper describes some alternatives for making inferences about such a rate difference and shows whether and how these alternatives may be implemented in three different survey software packages. The software packages illustrated - SUDAAN, WesVar and Bootvar - all can make use of bootstrap weights provided by the analyst to carry out variance estimation.

Issue Number: 2005001
Author(s): Lafortune, Yves; Roberts, Georgia
FormatRelease dateMore information
HTMLJune 23, 2005
PDFJune 23, 2005