Export to your Reference Manger

Please Copy & Paste



Bookmark and Share

The Impact of Typical Survey Weighting Adjustments on the Design Effect: A Case Study

[journal article]

Chatrchi, Golshid; Duval, Marie-Claude; Brisebois, François; Thomas, Steven

fulltextDownloadDownload full text

(859 KByte)

Citation Suggestion

Please use the following Persistent Identifier (PID) to cite this document:http://nbn-resolving.de/urn:nbn:de:0168-ssoar-427021

Further Details
Abstract In survey sampling, the final sample weight assigned to each sampled unit reflects different steps of weighting adjustments such as frame integration, nonresponse and calibration. The analysis of the design effects for each of these adjustments casts light on their effects on the precision of survey estimates. In this paper, we limit our scope to the Canadian Community Health Survey (CCHS), briefly describe the weighting process of this survey and examine design effects at different steps of the weighting process to quantify how the overall variability in estimates can be attributed to the complex survey design and to each of the individual adjustments in the weighting process. As expected, the results suggest that the use of unequal person-selection probabilities and the nonresponse adjustment have the most negative impact on the design effect of the CCHS while calibration and winsorization decrease the design effect and improve the precision of the estimates.
Keywords survey research; weighting; sample; Canada; response behavior; data capture; random sample; variability; private household; survey; social stratification
Classification Methods and Techniques of Data Collection and Data Analysis, Statistical Methods, Computer Methods
Document language English
Publication Year 2015
Page/Pages 15 p.
Journal Survey Methods: Insights from the Field (2015)
DOI http://dx.doi.org/10.13094/SMIF-2015-00006
ISSN 2296-4754
Status Published Version; peer reviewed
Licence Creative Commons - Attribution-Noncommercial-No Derivative Works