The Finnish Voter Barometer surveys are commissioned by the major parties in the Parliament. In election years, other sponsors have also been involved. Up till 1988, the surveys were planned by Gallup Finland (TNS Gallup Ltd, Kantar) and the Department of Political and Economic Studies at the University of Helsinki in collaboration with the political parties. After 1988, the planning has been in the hands of Gallup Finland and the parties.
From 1973 to 1990, the annual voter barometers were collected mainly with face-to-face interviews using structured questionnaires, 1000-2000 interviews/barometer. Since 1991, the data have been collected with GallupChannel (PC's installed in respondents' homes). No surveys were conducted in 1985 and 1989.
Survey themes include voting behaviour, opinions on the government and the parties, and attitudes towards national political issues. In addition, respondents' economic expectations and political participation have often been studied, as well as their opinions on local politics, and the significance of the party leader to a party. Background variables have often included the social class of the respondent and the household head, economic activity of the household head, respondent's education, age, mother tongue, gender, party membership, political views, trade union membership, and family's annual income. There are also regional variables. Even though the variables are not always fully comparable from one survey to another, all kinds of comparisons over time are possible. The FSD collections cover data from 1973-2005.
The variables describe the following: expectations concerning the economical situation, the most urgent reforms needed, changes in the political life during the last 12 months, satisfaction in the government's actions, political alienation, grading the present government's actions, interest in politics, left/right dimension in politics, various coalition combinations of the cabinet and the prime minister's party, image of the parties, constancy of voting pattern in the electoral college election, the parties wished to participate in the government, opinions on economical misdemeanours, and party support. Some variables are connected with the interview and the social structure of the sample.
At first, the target population was stratified according to the number of inhabitants aged 18 or over in each province and in the second stage, by municipality group within each province. Municipalities were classified into five groups on the basis of their level of industrialisation. The stratifying frame achieved in this manner was used to count how many interviews should be done in the municipalities of each municipality group in order to have a geographically and demographically representative sample. The municipalities in the sample have been selected with the Deming zone selection method. This method is based on the use of random numbers and gives each municipality an equal chance to be selected, although municipalities with large population have a better chance to be selected. A municipality may be sampled more than once - in which case more than one starting point is drawn for that municipality. The interviews were conducted in clusters of four subjects using the starting point method. The starting points were drawn from the population register, selecting individuals born on the 5th of any month, and then selecting starting addresses among them from the sample municipalities at fixed intervals. The first interview was conducted at the starting point, after which the interviewer proceeded to the next three households. Within the household, the person whose birth date was closest to the beginning of any month was chosen for the interview. Interviews were conducted in 58 cities and 83 rural municipalities. The number of respondents was 1,303.