NotesFAQContact Us
Search Tips
ERIC Number: ED311512
Record Type: Non-Journal
Publication Date: 1989-Aug
Pages: 27
Abstractor: N/A
Reference Count: N/A
Determining Nonresponse in Survey of Newspaper Editors.
Chang, Tsan-Kuo; And Others
In survey research, nonresponse has been one of the major concerns about survey accuracy and validity. A case in point is the increasing use of surveys in mass communications research; much previous research has dealt with nonresponse among the general public, but little is known about nonresponse in special subgroups in the population where any bias caused by nonresponse may have significant effects on the conclusions derived from the data. As part of a larger project, a study used a multivariate approach to determine nonresponse bias, if any, in a national survey of newspaper editors. A six-page closed-ended questionnaire was sent to 540 newspaper editors around the United States and received a 52% response rate after three mailings. Discriminant analysis examined the differences between respondents and non-respondents with respect to several intra- and extra-media variables simultaneously. Results showed that nonresponse does not necessarily cause a severe, significant bias in terms of some market and organizational characteristics of the newspapers selected in the sample. The major reason for nonresponse appeared to be related to newspaper editors' lack of time in the newsroom. These findings sugget that a well-designed and carefully executed mail survey is still the best and least expensive alternative for researching American journalists. (Four tables of data are included, and 29 references are attached.) (SR)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A