This article proposes using a statistical hypothesis test to check for the nonresponse bias that results when there is a corrrelation between strength of opinion and reply on issue preference. This idea is developed through an example of two mail polls conducted by a Columbus, Ohio newspaper to ascertain voter preference prior to the November 1983 election.
All Science Journal Classification (ASJC) codes
- Sociology and Political Science
- Social Sciences(all)
- History and Philosophy of Science