This is a report about the use of polls on television at CBS News. A brief history traces the acceptance of polls during the early days of television in political and public affairs broadcasts. Polls began as an important source of information for documentary broadcasts. However, it was modern election coverage that led to establishing an in-house polling capability. After several years of polling only for CBS News broadcasts, a partnership was formed with The New York Times. This was the first of the joint media polls in the United States, and continues today as a successful venture.
Pre-elections polls have never before been so disqualified in Spain. It will need quite a few years of impeccable work and new favourable electoral conditions for the professional activity of public opinion researchers to recover the image and the respect that without any doubt they enjoy in other Western countries. The big national debate in Spain on Spain's staying in or leaving NATO became one of the biggest political storms imaginable, even to the point that many commentators remembered the old dichotomy of "the two Spains" (which has caused us Spaniards so many headaches throughout our history). Let us look at the development of Spanish public opinion on NATO, from the start of the post-Franco democratic period until the moment just before Spain's joining this organisation (1975-81).
The votes of slightly over 9,1 million Dutchmen for 2 out of 27 parties turned out different from what the samples of all pre-election surveys had told the researchers. All interest had been focussed on whether the governing coalition would keep or loose its small majority. The nearest any Dutch institute approached the actual vote was 75 out of 150 parliamentary seats. The voters gave the coalition partners 81 seats. Although there had been indications of "hidden strength" of the Christian Democrat coalition partner, only an "electronic" NIPO-survey late on the eve of election day detected clear indications of a last minute-swing toward that party. (It caused joy and rising share prices on the Amsterdam Stock Exchange immediately after its publication very early on May 21st. Several of the "hidden strength"-data will be presented. Finally we discuss some of the problems caused by the increasing effects of television electioneering.
My topic is a relatively new type of opinion polling - the pro bono survey. Pro bono surveys are increasingly common in the U.S. and they represent an important form of publicly disseminated polling. I will discuss the general topic and then tell you about one example of such a study: "The People and The Press" which Gallup conducted tor the Times Mirror Corporation.
This paper outlines the role and impact of opinion polls during the two years preceding the recent Irish Constitutional Referendum on divorce, and also in the final stages of the actual campaign. In doing so, it emphasises the relevance of professional interpretation of published polls and highlights the accuracy of the pre-referendum polls in detecting initial trends, and the subsequent swing towards a rejection of the amendment. Two primary factors which contributed to the final result - which are based exclusively on opinion poll findings - are also identified, and commented upon. Finally the paper aims to provide a response to the post-referendum charge of some Irish commentators who claimed that, again, the "polls got it wrong".
This contribution has three aims: 1. To determine the existing assumptions about possible negative influences of opinion polls. 2. To determine what can be said about these assumptions from the point of view of empirical social science. 3. To outline how other phenomena which have not been investigated yet might be dealt with in the future.
Over the past few years, the author has frequently been featured on radio and television, presenting and commenting on the results of political opinion polls. The polls of May 16th and 17th as conducted by Inter/View, as well as those organized by the Nipo and the NSS indicated that the CDA and the VVD stood a considerable chance of not being able to maintain the majority (less than 75 seats in parliament). In the election of May 21st, however, the governing coalition ended up with no less than 81 seats. This discrepancy could be attributed in its entirety to the results of the CDA, the party of the popular Prime-Minister Lubbers. For 11 of the 12 parties, the election results were virtually identical to Inter/View's minimum predictions. Only the CDA scored considerably higher than the predicted maximum. It can be demonstrated that these changes took place within the last few days before the election. Voters switched to the CDA from all other parties. The polls, published before the election, rank as one of the main factors that influenced these major changes. In particular the threat of the coalition losing their majority and Lubbers consequently being unable to "finish the job" (CDA's election slogan), produced these changes.
Past studies have attempted the effect of various factors on the accuracy of opinion polls. In the area of pre-election poll alone researchers have studied the effects of question order (Crespi and Morris, 1984), late trend in voter preference (Day and Becker, 1984), disproportionate voter turn-out (Day and Becker, 1984), biasing pressures (Hennessy and Hennessy, 1961) and structural constraints (Shamir, 1986) on the accuracy of polls. This article attempts to study the effect of 'Rural' component on pre-election polls. In a country like India where nearly 80% of the population live in rural villages, it is very important to study the predictive ability of pre-election polls in rural areas as compared to that in urban areas.
The American press, with a long tradition of running its own polls, treats them as a key part of its news reporting. The key elements for success in this enterprise are commitments to time, space and staff, and a recognition of what polls can and cannot do in the news report. This article examines, according to a list of concerns presented to the author by Robert Worcester for ESOMAR, the key roles to be played by the different participants in the process.
The paper presents a model for investigating the don't know phenomenon in the elections. The model distinguishes four different types of don't know voters according to the way people are stopped or remain stopped in the elections. The testing of the model proves that different cognitive modes exist singly and in combinations among respondents and that there are differences over time in the way people orient themselves to the campaign. Even a part of those who are supposed to be "the decidedâ prove not to be that certain. The results also show that different types of donât know voters differ in the way they follow the campaign in the media.
I would like to suggest that the polling industry should not agonise too much about getting elections wrong. Error may be good for you; an occasional disaster can be good for the industry and good for the public. If polls were to have too long a run of always getting it right, it would raise FALSE expectations for your trade. It would discourage voters and politicians from fighting elections to the end and it would make pollsters too un-self-critical - and then they would be riding for a fall.
A somewhat general survey of the condition of opinion polling round the world is reported. In Western Europe and the English speaking countries in particular they are firmly embedded, recognised and understood by politicians and journalists, together with a few restrictions imposed from outside that would appear trivial by world standards. Elsewhere the development of the profession is related strongly to the nature of the regime and to the prosperity of the country. Good professional work is able to be carried out in most countries in the world, but transport, communications, statistical sources and the society itself lead to different forms of implementation of the basic principles. There is growing an acceptance of the usefulness and informativeness of opinion poll surveys in most countries, whether open societies or authoritarian regimes. The profession is strengthened by the existence of a code of ethics, but in some countries it is threatened by competition from outside. Finally there is growing evidence of the value of opinion polls as a challenge to the proper conduct of electoral procedures, and as an indication of their fairness.