Chapter

Nonresponse Error at the Item Level

in The Total Survey Error Approach

Published by University of Chicago Press

Published in print October 2005 | ISBN: 9780226891279
Published online March 2013 | e-ISBN: 9780226891293 | DOI: http://dx.doi.org/10.7208/chicago/9780226891293.003.0007
Nonresponse Error at the Item Level

Show Summary Details

Preview

The type of nonresponse discussed in this chapter involves respondents' failure to answer every appropriate question, known as item nonresponse. Nonresponse is a problem to the extent that it biases results, so that sample statistics are not on-target estimates of the corresponding population parameters. The first part of this chapter discusses the different ways in which item nonresponse occurs, along with ways to minimize item nonresponse. The second section considers some of the effects of item-missing data. The third section presents procedures for handling missing data. Because item nonresponse can distort data analysis results, this is a statistical topic, though many survey practitioners do not handle missing data in the ways that statisticians usually advise. There are three basic types of item nonresponse in surveys. The most important type for the purposes of this chapter is the “don't know” answer, because it is, unfortunately, a common response. Item nonresponse also occurs when respondents refuse to answer some questions, which is a much rarer situation.

Keywords: nonresponse error; data; statistics; surveys; don't know answer

Chapter.  10360 words. 

Subjects: US Politics

Full text: subscription required

How to subscribe Recommend to my Librarian

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.