such as pseudo polls. Contact information for two of these organizations is contained in appendix A.
Interviewers who work for high-quality polling organizations should be thoroughly trained to explain to respondents that their cooperation is entirely voluntary. They should also explain that their responses will remain totally confidential and that the poll data will be reported only in aggregated statistical form. These pledges mean that your answers will not be seen by anyone other than the interviewer and the people at the polling organization who process the data for computer analysis. Furthermore, employees of the polling organization should have signed their employer's "confidentiality statement," in which they pledge never to violate the confidentiality of the information that respondents provide.
On occasion, a polling organization may want to deviate from these common ethical practices, but the legitimate poll will always seek and receive the explicit approval of the respondent. One exception is a media- sponsored poll in which the news organization often will be interested in having a reporter call back some respondents for in-depth interviews to gather quotes to use along with some of the respondents' answers to the poll questions as reported in a news story. The interviewer should ask whether or not you would be willing to speak to a reporter, and the reporter should always get your explicit permission to interview you and your agreement that the conversation will be "on the record."
CRESPI, IRVING. 1989. Public Opinion, Polls, and Democracy. Boulder, Colo.: Westview.
This short volume contains a concise review of the potential problems with polls, linking technical and philosophical issues. Its strength is the discussion of the relationship between public opinion and policy and the intermediary role of polls in the process. It presents issues of polling methodology in a cogent fashion that requires no statistical background on the part of the reader.
The last two chapters on the relationship among news, polls, and democracy summarize all the main issues that are continually under review and dispute. Crespi is concerned about the possibility that the expanded use of polls by news organizations will undermine democracy, rather than strengthen it. He sees this as a function of two problems they have to overcome. One is the nature of the topics covered