VALU 1991 - SVT exit poll survey parliamentary election 1991

Swedish electorate

With the help of statistics concerning completed postal voting in the 1988 election (Nils Johnson and Johan Weine), 10 post offices were selected to be included in VALU-91. At the selected post offices, those who voted by mail during a morning and an afternoon shift on the Thursday and Friday before election day were asked to fill in the survey. The morning session was three hours and the afternoon session one hour. The starting point for each session was chosen randomly. On election day, VALU-91 was carried out at 26 polling stations. This corresponded to 31 electoral districts. The electoral districts were distributed among the various regions in accordance with the size of the regions in terms of the number of eligible voters. This means, for example, that fewer electoral districts were selected in the two Norrland regions than in the other regions. Within each region, the constituencies that were to be included in VALU-91 were chosen by random stratified sampling (stratified by size and geographical location within the region). In this way, 26 electoral districts were selected. For each electoral district, a morning session of 2 hours and an afternoon session of 1 hour were then selected. The starting point of the session was chosen randomly.