Dartmouth College Library Bulletin
POLLING THE PUBLICVIRGINIA L. CLOSE
PUBLIC OPINION as a topic and public opinion polls have been dealt with three times in issues of the Library Bulletin. 1 This article will not repeat generally what was said earlier but will concentrate on newer, contemporary records of polling data. Questions from students concerning polls have become more common, making an updating worthwhile. Not only are new sources of polling information available but there are also topics of historical interest that have not previously been studied fully. For example, in a 1988 issue of the Public Opinion Quarterly there is an article examining why the Literary Digest's 1936 presidential election poll failed. 2 That event lingers in perception as a failure that led to the demise of the journal. The author of the article states that '. . . no analysis has been conducted to determine why the Literary Digest poll was wrong. Consequently, we have some ideas -- really competing hypotheses -- as to why the poll failed but no empirical research by which to determine the source of the error.' 3
New bibliographical guides are being published to aid the researcher. In 1990 Graham R. Walden's Public Opinion Polls and Survey Research: A Selective Annotated Bibliography of U.S. Guides and Studies from the 1980's appeared. It describes polling research as a '. . . bibliographically neglected field.' 4 Walden's book does not contain a record of polling sources and lists of polls but it is an annotated listing of studies about polling. However, in 1988 this same author contributed an article, 'Public Opinion Polls: A Guide to Accessing the Literature,' to R.S.R.: Reference Services Review. This article complements his volume on polling research. 5 In 1990, also, the Public Opinion Quarterly included in its regular feature, 'The Polls,' an article by Tom W. Smith and Frederick D. Weil entitled 'The Polls-A Report. Finding Public Opinion Data: A Guide to Sources.' 6 The sources are gathered under five major categories (Archives, Survey Organizations, Publications, Scales and Indices, and Secondary Analysis) .
Older guides to the literature of public opinion and polling can be identified through guides in the Reference Room, and through the online system, as well as by checking the April 1968 issue of the Library Bulletin (p. 88).
The publications that carry polling results are varied. Some started up, appeared for a period of time, and ceased; others changed title and continued on in altered form. Still others began solidly and went on that way. There are also names, both of polling publications as well as of archival repositories of polling data, that have been long established. Mentioned below are several of the more prominent sources of data.
The Gallup Organization and Louis Harris and Associates are both represented in the reference collection. (Perhaps the article by Walden in the R.S.R: Reference Services Review, located in the Reference Room, is as convenient as any in giving a brief description of the organizations and their publications.) For Gallup there is The Gallup Poll. Public Opinion, which first appeared in 1972 with a volume covering 1935-1971. With the volume for 1978, the present pattern of coverage of one year per volume began. Each of the volumes contains helpful introductory pages, indexing, and a chronology of events for the year. To keep the annual volumes up to date, readers can consult The Gallup Poll Monthly (December 1983+). Earlier titles for this journal are The Gallup Opinion Index (1965-January 1981) and The Gallup Report (February 198I-November 1989).
Louis Harris and Associates issues a weekly, generally two-page, publication called The Harris Poll. In addition, Harris also distributes at the present time, and not so frequently, a usually longer publication called the Business Week/Harris Poll. Both The Harris Poll and this are indexed in one annual index with the Business Week/Harris Poll identified by a (BW) after the index entry.
Results of the Gallup and Harris polls show up in a variety of places, most commonly in the various newspapers. But in addition to these polling organizations there is other polling information available, provided that one knows where to look for it. For example, the National Journal subtitled The Weekly on Politics and Government, carries in each issue a one-page feature called 'Opinion Outlook: Views on the American Scene.' A variety of polling sources are cited as sources for this feature: among them Peter D. Hart Research Associates, Inc. for NBC News, the Los Angeles Times, the Gallup Organization, Louis Harris and Associates, and Hart and Teeter for NBC-Journal. The American Enterprise, a bimonthly published by the American Enterprise Institute, includes in each issue a section of approximately twenty pages titled 'Public Opinion and Demographic Report' and includes a bottom-of-page note that `The Roper Center for Public Opinion Research at the University of Connecticut helps compile the data used for this section. A nonprofit research facility, the Center maintains the largest and most comprehensive collection of survey data in the United States and seeks to promote their informed use.' The data, charts, and tables are all identified as to specific sources, which include those which have been mentioned above. The most recent issue of the Dialog Chronolog(January 1992) reports the availability now of a new data base for online searching.
This is 'Public Opinion Online (Poll)' produced by the Roper Center. The announcement in Chronolog says: 'Did you know that a great amount of the opinion polling done in the U.S. is never reported publicly? Most surveys contain a large number of questions (it is not at all unusual for a poll to contain 50 or more items), and broadcast and printed accounts of this research report only a fraction of them.' (p. 16) `Public Opinion Online' will open these up.
Since reference has been made to the Roper Center collection, the Inter-University Consortium for Political and Social Research should also be mentioned. Its Guide to Resources and Services is in the Reference Room. Page xxviii of the current guide (for 1991-1992) has an 'Outline of Archival Holdings.' In the guide Section XIV, Mass Political Behavior and Attitudes, includes a subsection on Public Opinion on Political Matters. The sources of the consortium's data on polls are many and varied, and are, it might be added, only one area of its resources. The College is a member of the consortium and questions about access to its resources can be addressed to the Reference Department.
There are two other major continuing sets in the Reference Room. One is the American Public Opinion Index, published since 1981 by the Opinion Research Service. The first section of the guide identifies the questions and by a symbol identifies the agency asking the questions. The other, second, section indicates what organization the symbols represent. An accompanying microfiche collection, American Public Opinion Data [microform] provides responses to some ninety-five percent of the questions. For those questions for which there are not responses in the microfiche collection, Section II of the guide supplies information on those agencies that have to be contacted to obtain them. The microfiche are located in the Jones Microtext Center.
For countries beyond the United States there is the Index to International Public Opinion, prepared by the Survey Research Consultants International, Inc., and published since 1978. The volumes are arranged, first of all, in three categories: single nation surveys, multinational surveys, and world surveys, and each of these is divided topically. Indexes offer approaches by subject categories, by countries and regions where the surveys were conducted, and by countries and areas referenced in the surveys. Presently some 165 countries are represented by the public opinion polls. The editors note, in the preface to the 1983-1990 volume, that the statistics '. . . are relevant to many perennial topics such as labor-management relations, how nations perceive one another, the role of organized religion in human society, the status of women and minorities, the functioning of the judicial system, and violence and terrorism.' These same topics, plus many others, are equally applicable to the polling documentation for this country.
Also to be kept in mind is that many of the ordinary indexes and bibliographies that are used daily in the Reference Room will supply good references to articles and other publications on polling. One final publication that could be of potential aid is the International Encyclopedia of Communications, published in four volumes in 1983 by the Oxford University Press. A check of the index reveals entries under Public Opinion, Opinion Measurement, and Poll.
The Literary Digest is not the only journal to have fallen victim to presidential election forecasting. Walden's article, referred to earlier, says that the Gallup Organization found the year 1948 to be 'perhaps the darkest hour for the organization. The incorrect forecast [Gallup predicted Thomas Dewey over Harry Truman] resulted in newspaper editors canceling subscriptions to The Gallup Poll and the Illinois Legislature standing for one minute of silence for The Gallup Poll before beginning its session.'7 There are probably many more intriguing circumstances like these that an enterprising student can discover and pursue.
1. Virginia L. Close, "'Focalizing" the News,' Dartmouth College Library Bulletin n.s. 6 (1966): 39-45; -- 'Opinion Polls,' Dartmouth College Library Bulletin n.s. 8 (I968): 86-88; Theresa Blake, 'Journey's End,' Dartmouth College Library Bulletin n.s. 14 (1974): 106-107.
2. Peverill Squire, 'Why the 1936 Literary Digest Poll Failed,' Public Opinion Quarterly 52 (1988): [I25]-I33.
4. (New York: Garland Publishing, Inc., 1990), xiii.
7. R.S.R.: Reference Services Review 16 (1988): 66.