Parliamentary elections 2023, background analysis of voters

The review examines persons entitled to vote and those who voted in the Parliamentary elections 2023 according to various background factors. Data on voting are data on votes cast, i.e. an entry on voting made in the votes cast register. The data of the review are based on the Ministry of Justice's voting register and votes cast register and Statistics Finland's individual-level data. The review examines only persons entitled to vote and those who voted living in Finland. Individual-level background factors affecting votes cast are also presented for the first time in the StatFin database.

Analysed data

The review examines persons entitled to vote in the Parliamentary elections 2023 and advance voters in the whole country. The second examined group are persons entitled to vote and all who voted in such municipalities and voting districts where an electronic voting register was used. The voter groups examined in the review are separate but partly overlapping. The following terms are used for these voter groups in the text, tables and figures of the review:

Group 1. Advance voters:

• All persons who voted in advance in the elections in the whole country.

Group 2. All who voted in the areas:

• Advance voters and persons who voted on the election day in the voting districts that used an electronic voting register.

There were 4,277,487 persons entitled to vote resident in Finland in the Parliamentary elections 2023. Of these persons, 1,730,789, i.e. 40.5 per cent of all persons entitled to vote, voted in advance. The voting districts where data on voting on the actual election day are available had 2,098,135 persons entitled to vote. Thus, data containing voting on the election day were received about 49.1 per cent of persons entitled to vote.

An electronic voting register is necessary for examining voting turnout on the election day. Data on voting on the actual election day were available from 182 municipalities, from 134 municipalities for all voting districts and from 48 municipalities for some of voting districts.

The voting districts and municipalities where an electronic register was used are not divided evenly on the level of the whole country. The most exhaustive data from constituencies were available from the constituency of Central Finland, on 89.5 per cent of persons entitled to vote. The coverage was weakest in the constituency of Helsinki, where data were available from one voting district that covered 1.9 per cent of persons entitled to vote in the constituency.

The persons entitled to vote included in the data obtained from the areas covered by the electronic voting register correspond well to all persons entitled to vote. In the examined areas, persons entitled to vote do not much differ from all persons entitled to vote in terms of the key socio-economic background factors. Differences in the age, sex, main type of activity and language of persons entitled to vote are marginal between the areas and the whole country. The difference is biggest in the educational structure of the groups, as there are slightly fewer people (1.9 percentage points) with at least master's or equivalent level education in the areas than in the whole country.

Interpreting the data

It is important to note that the number of voters based on data on votes cast is not exactly the same as in the result data of the elections. There may be small differences by area in the numbers of voters extracted from different sources. This is because the number of those who voted in the election result data is based on ballots, while the data on votes cast are based on entries made in the electronic voting register on those who voted.

Differences may arise between these sources, for example, for the following reasons: 1) a vote cast in advance voting is not taken into account on the basis of the Election Act, 2) an advance vote is not received before the deadline, 3) possible procedural errors on the day of the election, e.g. a voter has not left a ballot or two ballots clinging together were given to a voter, or 4) different recording errors. Because the above-mentioned random factors are not corrected in the data on votes cast, they may cause differences in the numbers of voters in different data.

In practice, this means that the exact number of persons who voted cannot in all cases be selected from the database tables related to the review if election results based on ballots are wanted. The numbers of persons who voted presented in the tables describe the entries in the voting register. It is impossible to correct possible erroneous entries afterwards, for example.

Young men are passive voters

In the voting districts from which all voting data were available, a total of 70.9 per cent of persons entitled to vote cast their vote. Women voted in advance more actively than men both in the whole country and in the areas, apart from the over 70 age group. In the whole country, 42.2 per cent of women entitled to vote voted in advance, which is 4.0 percentage points more than men.

Persons who voted in the areas and advance voters by sex and age group in the Parliamentary elections 2023 (incl. data on votes cast in the Parliamentary elections: coverage, %)

Content is being loaded

The voting percentage that takes into account all persons who voted is also higher for women. In the areas 71.9 per cent of women voted, which was 2.1 percentage points higher than for men. Examined by age group, women voted more actively than men in all age groups except for persons aged at least 70. The difference in voting turnout between genders was biggest in the younger end of the age distribution. Women's voting percentage was over 10 percentage points higher in the 19-year age group and nearly 10 per cent higher in the 20 to 24 age group than that for men.

Voting is more common in higher income brackets

The income level of those who voted and of persons entitled to vote is examined with the help of disposable monetary income. Income data derive from the latest taxation data from 2021. Disposable monetary income refers to monetary income after taxes. This consist of earned income, property income, benefits in kind and transfer income.

The income of persons entitled to vote is divided into quintiles. Quintiles are derived by arranging persons entitled to vote by income and by dividing the group into five equal parts. The groups formed in this way have around 854,000 persons entitled to vote for all persons entitled to vote in the whole country. In the areas from which data were received on voting on the election day as well, the groups have around 418,000 persons entitled to vote.

For the whole country, income data are missing for some 7,300 persons entitled to vote. In the data for the areas containing election day votes, income data were missing for some 3,500 persons entitled to vote. The median for the disposable income of all persons entitled to vote was around EUR 23,500, in the examined areas the corresponding median income was around EUR 23,100.

The level of income did not have much effect on advance voting. In the lowest income quintile, advance voting was relatively lower and in the highest one more common than in the other income brackets.

Content is being loaded

When examining all who voted, the effect of income on voting turnout is evident. As a rule, voting was more common in higher income groups than in lower income groups. Altogether 85.1 per cent of those belonging to the highest income quintile voted. By contrast, the voting turnout was 58.4 per cent in the lowest quintile. Thus, the difference in voting turnout was nearly 27 percentage points between the lowest and highest income quintiles.

High education increased the probability of voting

Education had a clear connection to voting turnout. Persons entitled to vote with lowest tertiary or higher level of education voted more often than average in the Parliamentary elections. The voting percentage of persons with master's or doctoral level education was 19.9 percentage points higher than that of all persons entitled to vote and 35.9 percentage points higher than that of all persons entitled to vote with only basic level education.

In educational groups, men's voting turnout that takes into account all who voted was higher than women’s among those with lowest tertiary level and bachelor's level education. There was in practice no difference in the voting turnout of sexes among those with basic level education and at least master's level education. Women's 2.1 percentage points higher voting turnout in practice comes from the group of those with upper secondary education. This educational group is clearly the largest, it includes 45.6 per cent of persons entitled to vote in the areas and 44.1 per cent of all persons entitled to vote.

Content is being loaded

Lowest tertiary level qualifications decrease in the population and no new qualifications are completed in Finland in this category, so the age structure of the educational group differs clearly from other groups. Lowest tertiary level education covers qualifications above upper secondary level that are not university of applied sciences degrees.

In groups of main type of activity advance voting was most common among pensioners, 51.1 per cent of whom voted in advance. In other groups of main type of activity, the share of advance voters was relatively smaller than among all persons entitled to vote.

Employed persons were the most active voters among all persons entitled to vote, 75.2 per cent of whom voted in the examined areas. Pensioners' voting turnout was on level with all persons entitled to vote (70.7%), in other main type of activity groups voting turnout was below the level of all persons entitled to vote. Those in the inactive population (49.3%) were the most passive voters among those whose main type of activity was known.

Content is being loaded

Voting turnout of persons with foreign background was low

The background of persons entitled to vote and persons who voted is viewed by means of language and origin. Voting turnout was higher for persons speaking national languages as their native language than for all persons entitled to vote. In all, 0.6 percentage points more of Finnish and Sami speakers and 5.6 percentage points more of Swedish speakers voted compared to all persons entitled to vote.

In the Parliamentary elections, the right to vote is tied to nationality, that is, there are fewer persons speaking other than national language as their native language entitled to vote than in the Municipal elections, for example. Of all persons entitled to vote, 91.6 per cent were Finnish-speaking, 5.2 per cent Swedish-speaking and 3.1 per cent other language speakers. In the areas from which data were available, there were slightly more Finnish-speaking persons entitled to vote among all who voted and slightly fewer Swedish and foreign-language speakers than among all persons entitled to vote.

Content is being loaded

The voting turnout of foreign-language speakers was clearly lower than for those speaking national languages. In all, 17.3 per cent of foreign-language speakers voted in advance in the whole country. Correspondingly, 39.1 per cent of all foreign-language speakers entitled to vote in the areas voted, which was nearly 32 percentage points lower than among all persons entitled to vote.  

Content is being loaded

Viewed by origin, voting turnout appears very similar to the language groups. For all who voted in the areas, the difference between voting turnout among persons of Finnish and foreign background was over 32 percentage points.