Study title

The first Swedish exit poll survey (VALU) was carried out in connection with the elections in 1991, and it was followed by a similar survey in 1994. Voters leaving the polling stations were asked to fill in a questionnaire including questions on: party voted for in the general and local elections; time for decision how to vote; party voted for in the general election 1991; vote intention at the referendum on Swedish EU-membership; placement on a political left-right scale; confidence in Swedish politicians; gender, age, occupational group and union membership. The respondents also had to indicate how important a number of items were for the decision how to vote, and to give an opinion on the policy of the different political parties. In VALU 1994, Sweden (constituencies) was divided into geographical regions grouped around six major university cities (Lund, Göteborg, Växjö, Stockholm, Sundsvall och Luleå). In each of the regions, a university teacher was engaged as regional survey leader. The survey leaders acted as supervisors for the field agents who carried out the survey in the polling stations and early voting stations respectively. All survey leaders had knowledge of statistical methodology.