Study title
VALU 2022 - SVT exit poll survey parliamentary election 2022
Creator
Ekengren Oscarsson, Henrik (Department of Political Science, University of Gothenburg)
Näsman, Per (Division of Safety Research, Royal Institute of Technology)
Holmberg, Sören (Department of Political Science, University of Gothenburg)
Pettersson, Mikael (Sveriges Television)
Study number / PID
https://doi.org/10.5878/3akm-m070 (DOI)
Abstract
This exit poll survey (VALU) was carried out at the parliamentary election 2022 by Sveriges Television (SVT) in collaboration with researchers at the Division of Safety Research, Royal Institute of Technology (KTH), Stockholm and the Department of Political Science, University of Gothenburg. The main purpose of VALU 2022 was, as for previous exit polls, to provide analytical support for the election night broadcast on Swedish television during the election night, as well as for other post-election analyses.
In VALU, voters leaving polling stations are asked by public service broadcaster Sveriges Television to fill in a questionnaire and put it anonymously in a sealed box. After collecting and processing the responses, Sveriges Television can present an election forecast and an analysis of the reasons underlying the outcome of the election in its election night broadcast. The surveys are referred to as VALU, an abbreviation for vallokalsundersökning, Swedish for exit poll survey. The first exit poll survey in Sweden was carried out at the parliamentary election in 1991. Since then, similar studies have been carried out in connection with all parliamentary elections, referendums, and elections to the European Parliament.
In VALU, Sweden (constituencies) is divided into geographical regions grouped around major university cities. In each of the regions, a university teacher is engaged as regional survey leader. The survey leaders act as supervisors for the field agents who carry out the survey in the polling stations and early voting stations respectively. All survey leaders have knowledge of statistical methodology.