Sunday question

from Wikipedia, the free encyclopedia
This article or paragraph presents the situation in Germany . Help us to describe the situation in other countries.

As Sunday question (also voting intention question ) is in the German and Austrian opinion and election research is the question of the current voting intention. The question has established itself as a standard tool in empirical research and is used by academic and commercial survey institutes. It is used as the basis for calculating the current mood and for projections .

Periodic

The Sunday question has been regularly surveyed in Germany since 1949, so that trends can be determined and temporal relationships shown. Publications on the Sunday issue are piling up before elections, but even when there is no election campaign , data is regularly collected and published - often on a monthly basis. The Sunday question only represents a current opinion of the population and is only suitable to a limited extent for forecasting the actual election result, even if it is often referred to as a forecast by the media and parties. For the most part, the research institutes only publish projections that modify the data collected on the Sunday question, taking into account other factors such as long-term party identification . Which factors the institutes use is part of their trade secret.

methodology

In Germany, the wording “If there were (actually / really) the Bundestag election next Sunday , which of the following parties would you vote?” Has prevailed and is adapted to the respective political level ( state , federal, European level ). Depending on the study design or the tradition of the uplifting institute, the question is sometimes asked with a given party name, sometimes openly. In its press code , the German Press Council recommends that the print media publish surveys always linked to the number of respondents, the time of the survey, the client and the exact question, and to state whether the survey can be methodologically considered representative.

Data collection

The polling institutes use different interview methods to collect the primary data on the Sunday question.

Personal interview

With this method (also called "face-to-face interviews") a randomly selected target person is visited at home and asked to answer a given list of questions. This method is as old as opinion polling itself and for many years was the only technically recognized method for collecting data. In many opinion research institutes, it was gradually replaced by telephone interviews between the 1980s and 1990s for reasons of cost.

Telephone interviews

A list of randomly selected telephone numbers is worked through in telephone interviews. The questioning can be done on the phone by a person, or the data is collected by an automatic message and option on the part of the called party. Telephone interviews as a method for collecting data are scientifically controversial.

  • Emnid asks about the party preference by phone. The caller reads out a list of parties. The list currently (as of mid-April 2017) consists of the following parties: AfD, CDU, CSU, FDP, GRÜNE, LINKE and SPD. If the preference of the person called does not exist (e.g. NPD or Free Voters ), he is asked to name the party of his preference.
  • Infratest dimap asks about the party preference by telephone without reading out a list. The called party is asked to name the party of his preference.
  • Forsa asks party preference over the phone without reading a list. The called party is asked to name the party of his preference.

Political mood and projection

The raw data from the survey reflect the political mood at the time of the survey. For a better assessment of the result, together with the Sunday question, assessments on current issues (e.g. reform programs, tax policy, danger of war and other things) are asked and put in relation to the determined election result. It is also common and informative to ask for opinions on certain topics (economic competence, social policy, charisma of the top candidate , etc.) and thus present the strengths and weaknesses of a party in a more differentiated manner than would be possible in an actual election. The results of the Sunday question are subject to an election cycle . These data flow into the projections , which are intended to correct the political mood into a statement about the actually expected voting behavior. In addition to the political mood, experiences about medium and long-term loyalty among voters, tactical voting behavior (e.g. different awarding of the first and second votes) and a delay effect (“time lag”) also flow into this .

criticism

In the aftermath of the state elections in Baden-Württemberg in 1992 and 1996, polls, in which the Allensbach Institute set the polls of the party The Republicans far too low , caused public criticism . Allensbach forecast 4.5 or 4 percent share of the vote. The actual Republican election results, however, were 10.9 and 9.1 percent. In an interview on the incidents, the managing director Renate Köcher admitted that she did not want to offer the party a platform and therefore had not published the higher polling figures she was already aware of before the election. This approach met with criticism in journalism and science.

In the 2005 Bundestag election , the predictions of all opinion research institutes were far removed from the actual election result. The Union was predicted up to eight percentage points too much. In the run-up to the 2013 federal election , the chairman of the newly founded AfD, Bernd Lucke , publicly accused several institutes of manipulating the polls to the detriment of the AfD, even though, according to his information in the raw data, the party was well above the five percent hurdle . The Forsa Institute then obtained a temporary injunction from the Cologne Regional Court with a declaration of cease and desist against the AfD chairman. The election result was 4.7%.

The political scientist Wolfgang Gibowski , who once worked for the ZDF - Politbarometer , commented on the reliability of election polls as follows: "It is undisputed that the results of the Sunday question [...] were often very unrealistic." The actual electorate did not correspond to the participants in the poll because some of the lots could not be interviewed by phone because they could not be reached or refused to participate. The head of the Forsa opinion research institute Manfred Güllner concluded: “No institute can ask a 100% representative audience of the actual electorate on a Sunday question. There remain the bottom line is always uncertainty and fuzziness. "Moreover, let Leihstimmen as part of a tactical choice decision, for example, from the CDU, the FDP transferred will be adequately understood. This aspect also includes the fact that most institutes do not ask separately for first and second votes , but only for a party preference. In addition, there is no way of reliably finding out who will actually take part in the election. The respondents can only claim that they intend to vote, since an indication of abstaining from voting is viewed as a political disinterest and stigmatization is feared. Another inaccuracy of the polls results from the fact that voting decisions are increasingly made immediately before the polls, sometimes only spontaneously in the voting booth.

If the number of cases for one party is too low, no evaluation is possible. The actual unreality of the question was also criticized. The SPD parliamentary group leader at the time, Frank-Walter Steinmeier , said: "The Sunday question doesn't say much because all those questioned know very well: Sunday is no election."

Web links

Wiktionary: Sunday question  - explanations of meanings, word origins, synonyms, translations

Individual evidence

  1. ↑ a b Jochen Groß: The prognosis of election results. Approaches and empirical performance . VS Verlag für Sozialwissenschaften, 2010, ISBN 978-3-531-17273-6 , p. 48 ff .
  2. ↑ Lena-Maria Schaffer and Gerald Schneider (2005): The forecasting quality of election markets and opinion polls for the 2005 Bundestag election . Politische Vierteljahresschrift , Vol. 46 (4), pp. 674–681.
  3. ↑ Jochen Groß: The prognosis of election results. Approaches and empirical performance . VS Verlag für Sozialwissenschaften, 2010, ISBN 978-3-531-17273-6 , p. 54 f .
  4. ↑ Günther Lachmann: "Why the AfD has such bad poll numbers." ' In: Die Welt , June 5, 2013.
  5. ^ Don A. Dillman: Navigating the Rapids of Change: Some Observations on Survey Methodology in the Early 21st Century .  ( Page no longer available , search in web archives )  Info: The link was automatically marked as defective. Please check the link according to the instructions and then remove this notice. (PDF) In: Public Opinion Quarterly , Volume 66 (3), 2002, p. 4.@1@ 2Template: Toter Link / igitur-archive.library.uu.nl  
  6. ↑ Sunday question . ifd-allensbach.de; Retrieved April 24, 2017
  7. ↑ Edith De Leeuw, Mario Callegaro, Joop Hox, Elly Korendijk, Gerty Lensvelt-Mulders: "The Influence of Advance Letters on Response in Telephone Surveys."  ( Page no longer available , search in web archives )  Info: The link was automatically defective marked. Please check the link according to the instructions and then remove this notice. (PDF) In: Public Opinion Quarterly , Volume 71 (3), 2007, p. 413.@1@ 2Template: Toter Link / igitur-archive.library.uu.nl  
  8. ↑ Sunday question - Emnid , wahlrecht.de
  9. ↑ a b Glossary for survey research , compiled by the Wahlen research group.
  10. ^ Dieter Roth: Demoskopie and politics. On the relationship and the misunderstandings between two heavily criticized professions . In: Hanna Kaspar et al. (Ed.): Politics - Science - Media. Festschrift for Jürgen W. Walter on his 75th birthday. , VS Verlag für Sozialwissenschaften, 2009, ISBN 3-531-16621-2 , pp. 241-256, here pp. 250 f.
  11. ↑ Solidly determined . In: Der Spiegel . No. 14 , 1996, pp. 27 ( online ).
  12. ^ Opinion research - disaster from the urn . In: Der Spiegel . No. 14 , 1996, pp. 26th f . ( online ).
  13. ↑ Miriam Hollstein, Manuel Bewarder : The oracles of the nation . In: Welt am Sonntag , March 27, 2011.
  14. ↑ AfD boss Bernd Lucke: "There is a problem with the opinion polls" . In: Handelsblatt , August 30, 2013.
  15. ↑ Zoff about survey data: AfD receives slap in legal dispute with Forsa . In: Spiegel Online , September 17, 2013.
  16. ↑ Hans Peter Schütz: Demoskopie: The pitfalls of the election polls . In: Stern , August 20, 2013.
  17. ↑ Martin Lindner at Stuckrad-Barre: The FDP smokes! In: Stern , October 25, 2012.
  18. ^ Konrad Fischer: Election researcher in the crisis . In: Wirtschaftswoche , May 23, 2013.