Alternative Question Forms for Attitude Scale Questions in Telephone Interviews |
| |
| Authors: | MILLER PETER V |
| |
| Institution: | Peter V. Miller is an Associate Professor in the Department of Communication Studies at Northwestern University. This is a revised version of a paper presented at the annual meeting of the American Statistical Association, August 1981. Data for the study were collected when the author was a Study Director at the Survey Research Center, University of Michigan. The author wishes to thank Robert Groves, Duane Alwin, Howard Schuman, and Charles Cannell for help and advice. The author alone is responsible for any errors. The project was funded by the National Center for Health Statistics (contract 233-78-2034). The contents of this paper do not necessarily represent the views of the National Center for Health Statistics. |
| |
| Abstract: | The absence of visual aids in telephone interviews has led investigatorsto seek methods for asking questions which are commonly accompaniedby aids in personal interviews. This study experimentally comparedtwo approaches to asking seven-point scale attitude questionsin a national (N = 4,300) telephone health survey. A randomhalf-sample received the attitude questions as single-step numericalselection tasks. The questions were administered to the otherhalf-sample in two stages, the first asking for a general verbalstatement of attitude and the second asking for a more detailedspecification. The two question forms produced comparable resultsin univariate distributions. Items administered with the one-stepform showed slightly higher intercorrelations. This study representsone step in the process of empirically testing data collectionprocedures which have been deemed necessary in the rich folkloreof survey research. Overall, the results show more comparabilitythan might be anticipated, given the multiple differences betweenthe forms. |
| |
| Keywords: | |
| 本文献已被 Oxford 等数据库收录！ |
|
