To main content

Party Preference Survey in May 2020 - Election today

Political Party Preference Survey, May 2020

Statistical news from Statistics Sweden 2020-06-08 9.30

The Political Party Preference Survey presents election results if a parliamentary (Riksdag) election were held in May 2020, as well as changes since November 2019 and since the 2018 parliamentary election.

In an election in May 2020, votes would be cast as follows.

Social Democrats (S): 33.7 ± 1.1 percent
Moderate Party (M): 20.1 ± 0.9 percent
Sweden Democrats (SD):17.1 ± 0.8 percent
Left Party (V): 8.2 ± 0.6 percent
Christian Democrats (KD):6.4 ± 0.6 percent
Centre Party (C): 6.0 ± 0.5 percent
Green Party (MP): 4.1 ± 0.5 percent
Liberal Party (L): 3.3 ± 0.5 percent
Other parties: 1.1 ± 0.3 percent

In May 2020, about 12 percent of the electorate is uncertain.

Estimate of election results “if an election had been held today”. May 2020, and the difference compared with the 2018 parliamentary election

Diagram

The change refers to a comparison with the 2018 parliamentary election in percentage points. Statistically significant changes are marked with an asterisk (*).

In May 2020, an adjusted method for producing estimates in the survey was introduced. For this reason, Statistics Sweden updated the Statistical Database from November 2010 and onwards with estimates calculated using the new method. To see how the adjusted method affected estimates for each political party from November 2010 to November 2019, go to the figures and tables here. For further information, see Definitions and explanations at the end of this item of statistical news.

The table below presents an estimate of the parties in May 2020, November 2019, and the latest election results. The table also shows changes from the November 2019 survey results and the 2018 parliamentary election. Statistically significant changes are marked with an asterisk (*).

Estimate of election results “if an election had been held today”. May 2020

Type of question: Which party would you vote for if an election were to be held in the next few days?

 Estimate May 2020Estimate November 2019**   
PartyPercentMargin of errorPercentMargin of errorChange since November 2019Election 2018Change since election 2018
C 6.0 ±0.5 7.4 ±0.6 ‑1.4* 8.6 ‑2.6*
L 3.3 ±0.5 3.8 ±0.6 ‑0.5* 5.5 ‑2.2*
M 20.1 ±0.9 18.4 ±1.0 1.7* 19.8 0.3
KD 6.4 ±0.6 6.8 ±0.6 ‑0.4 6.3 0.1
S 33.7 ±1.1 25.0 ±1.0 8.7* 28.3 5.4*
V 8.2 ±0.6 8.6 ±0.6 ‑0.4 8.0 0.2
MP 4.1 ±0.5 5.2 ±0.6 ‑1.1* 4.4 ‑0.3
SD 17.1 ±0.8 22.7 ±1.0 ‑5.6* 17.5 ‑0.4
other 1.1 ±0.3 2.0 ±0.5 ‑0.9* 1.5 ‑0.4*

* The change is statistically significant. ** The estimates for November 2019 were produced using the method introduced in May 2020, which is why they differ from previously published estimates for November 2019.

A table of the survey results from the current and the previous electoral period is available in Statistics Sweden’s Statistical Database. Statistics Sweden’s Statistical Database.

The political parties in May 2020

Changes for all political parties compared with the November 2019 survey and the 2018 parliamentary election are described below. The flow of voters between the parties is also described below as a proportion of the electorate. Only statistically significant changes in the flow of voters are presented.

Centre Party

If an election were held in May 2020, the Centre Party would receive 6.0 ± 0.5 percent of the votes. This is a statistically significant decrease, both compared with November 2019 and with the 2018 parliamentary election.

In relation to November 2019, the Centre Party has a statistically significant net loss of about 0.6 percent of the electorate to the Social Democrats.

Compared with the 2018 parliamentary election, statistically significant net losses are noted of about 1.2 percent to the Social Democrats, about 0.7 percent to the Moderate Party and about 0.2 percent to the Green Party.

Liberal Party

The Liberal Party would receive 3.3 ± 0.5 percent of the votes in an election in May 2020. This is a statistically significant decrease compared with both November 2019 and the 2018 parliamentary election.

Compared with the 2018 parliamentary election, statistically significant net losses are noted of about 1.0 percent to the Social Democrats, about 0.7 percent to the Moderate Party and about 0.2 percent to the Sweden Democrats.

Moderate Party

If an election were held in May 2020, the Moderate Party would receive 20.1 ± 0.9 percent of the votes. This is a statistically significant increase compared with November 2019, while it is not a statistically significant change compared with the 2018 parliamentary election.

Compared with November 2019, the Moderate Party has a statistically significant net gain of about 1.8 percent from the Sweden Democrats. At the same time, the Moderate Party has a statistically significant net loss of about 1.0 percent to the Social Democrats.

In relation to the 2018 parliamentary election, the Moderate Party has statistically significant net gains of about 0.7 percent from the Centre Party and about 0.7 percent from the Liberal Party. At the same time, the Moderate Party has a statistically significant net loss of about 0.9 percent to the Social Democrats.

Christian Democrats

The Christian Democrats would receive 6.4 ± 0.6 percent of the votes in an election in May 2020. This is not a statistically significant change compared with November 2019 and with the 2018 parliamentary election.

In relation to November 2019, the Christian Democrats have a statistically significant net loss of about 0.2 percent to the Social Democrats.

Compared with the 2018 parliamentary election, a statistically significant net loss of about 0.4 percent is noted to the Social Democrats.

Social Democratic Party

The Social Democrats would receive 33.7 ± 1.1 percent of the votes in an election in May 2020. This is a statistically significant increase compared with both November 2019 and the 2018 parliamentary election.

In relation to November 2019, the Social Democrats have statistically significant net gains of about 2.6 percent from the group “Do not know”, about 1.8 percent from the Sweden Democrats, about 1.0 percent from the Moderate Party, about 0.9 percent from the Left Party, about 0.7 percent from the Green Party, about 0.6 percent from the Centre Party, and about 0.2 percent from the Christian Democrats.

Compared with the 2018 parliamentary election, statistically significant net gains are noted of about 1.2 percent from the Centre Party, about 1.0 percent from the Liberal Party, about 0.9 percent from the Green Party, about 0.9 percent from the Moderate Party, and about 0.4 percent from the Christian Democrats.

Left Party

If an election were held in May 2020, the Left Party would receive 8.2 ± 0.6 percent of the votes. This is not a statistically significant change compared with November 2019 and with the 2018 parliamentary election.

In relation to November 2019, the Left Party has a statistically significant net gain of about 0.3 percent from the “Do not know” group. At the same time, the Left Party has a statistically significant net loss of about 0.9 percent to the Social Democrats.

Compared with the 2018 parliamentary election, the Left Party has a statistically significant net gain of about 0.3 percent from the Green Party.

Green Party

If an election were held in May 2020, the Green Party would receive 4.1 ± 0.5 percent of the votes. This is a statistically significant decrease compared with November 2019, while it is not a statistically significant change compared with the 2018 parliamentary election.

In relation to November 2019, the Green Party has a statistically significant net gain of about 0.2 percent from “other parties”. At the same time, the Green Party has a statistically significant net loss of about 0.7 percent to the Social Democrats.

Compared with the 2018 parliamentary election, the Green Party has statistically significant net gains of about 0.2 percent from the Centre Party and about 0.2 percent from the “Other parties” group. At the same time, the Green Party has net losses of about 0.9 percent to the Social Democrats and about 0.3 percent to the Left Party.

Sweden Democrats

The Sweden Democrats would receive 17.1 ± 0.8 percent of the votes in an election in May 2020. This is a statistically significant decrease compared with November 2019, while it is not a statistically significant change compared with the 2018 parliamentary election.

In relation to November 2019, the Sweden Democrats has statistically significant net losses of about 1.8 percent to the Moderate Party and about 1.8 percent to the Social Democrats.

Compared with the 2018 parliamentary election, a statistically significant net gain of about 0.2 percent from the Liberal Party was noted.

Other parties

The group “Other parties” would receive 1.1 ± 0.3 percent of the votes in an election in May 2020. This is a statistically significant decrease, both compared with November 2019 and with the 2018 parliamentary election.

In relation to November 2019, the “other parties” group has a statistically significant net loss of about 0.2 percent to the Green Party.

Compared with the 2018 parliamentary election, the “other parties” group has a statistically significant net loss of about 0.2 percent to the Green Party.

Figure 1 Net flows from November 2019 to May 2020. Percent of the electorate*

Diagram

* Figure 1 illustrates statistically significant net flows from November 2019 to May 2020 between the different parties. The figures indicate significant net gains and net losses in percent of the electorate. The flows are based on all those who submitted party answers or answered “do not know” when asked which party they would vote for in an election in each survey round. The flows cannot be used to calculate total changes in parties’ proportion of voters compared with November 2019, because only statistically significant flows are presented and also because there are flows between parties and the “do not know” group. (Unweighted information is available in table 22 in the table publication published on 11 June.)

Figure 2 Net flows between the 2018 parliamentary election and May 2020. Percent of the electorate*

Diagram

* Figure 2 illustrates statistically significant net flows in the electorate from the 2018 parliamentary election to May 2020 for the different parties. The figures indicate significant net gains and net losses in percent of the electorate. The flows are based on the answers of all the respondents in the survey who stated which party they voted for in the 2018 parliamentary election and which party they would vote for in an election in May 2020. The flows cannot be used to calculate total changes in parties’ proportion of voters compared with the 2018 election, because only statistically significant flows are presented and also because there are flows between parties and the “do not know” group. (Unweighted information is available in table 21 in the table publication published on 11 June).

Estimate of election results “if an election had been held today”, women (W) and men (M). May 2020

Diagram

 

Estimate of election results “if an election had been held today”, women and men. May 2020
PartyWomenMenTotal
 PercentMargin of errorPercentMargin of errorPercentMargin of error
C 6.7 ±0.9 5.3 ±0.8 6.0 ±0.5
L 3.3 ±0.8 3.3 ±0.7 3.3 ±0.5
M 18.3 ±1.6 21.9 ±1.5 20.1 ±0.9
KD 6.2 ±1.0 6.6 ±1.0 6.4 ±0.6
S 38.8 ±2.0 28.6 ±1.7 33.7 ±1.1
V 8.6 ±1.0 7.7 ±1.1 8.2 ±0.6
MP 5.9 ±0.9 2.3 ±0.6 4.1 ±0.5
SD 11.1 ±1.4 23.1 ±1.6 17.1 ±0.8
others 1.0 ±0.5 1.2 ±0.5 1.1 ±0.3

Estimate of election results “if an election had been held today”, by region . May 2020. Percent
Region C L M KD S V MP SD Other
  % ± % ± % ± % ± % ± % ± % ± % ± % ±
Sydsverige 4.1 ±1.5 3.3 ±1.6 19.8 ±3.0 5.7 ±1.9 31.0 ±3.7 6.6 ±1.9 5.4 ±2.0 22.8 ±3.4 1.2 ±0.9
Småland med öarna 6.0 ±2.3 3.1 ±2.0 17.0 ±3.6 6.9 ±2.7 35.3 ±4.8 6.5 ±2.6 4.1 ±2.2 20.0 ±4.2 1.2 ±1.2
Västsverige 7.1 ±1.6 3.7 ±1.4 20.4 ±2.7 7.3 ±1.7 34.2 ±3.2 7.6 ±1.6 3.5 ±1.2 15.0 ±2.4 1.1 ±0.7
Stockholms Län. excluding Stockholm municipality 5.4 ±1.8 3.6 ±1.4 26.3 ±3.7 7.8 ±2.4 30.5 ±4.0 5.7 ±2.0 4.6 ±2.0 14.9 ±3.1 1.2 ±1.2
Stockholm municipality 7.3 ±2.5 4.3 ±1.8 21.5 ±3.9 4.0 ±1.8 32.6 ±4.5 14.9 ±3.3 5.9 ±2.0 8.4 ±2.7 1.2 ±1.0
Östra Mellansverige 5.8 ±1.6 3.3 ±1.3 23.9 ±3.1 6.0 ±1.6 30.3 ±3.3 9.5 ±2.2 4.1 ±1.5 17.0 ±3.0 .. ..
Norra Mellansverige 5.7 ±2.3 3.0 ±1.6 14.7 ±3.6 7.3 ±2.9 39.3 ±5.2 7.1 ±2.6 0.9 ±0.9 20.1 ±4.7 .. ..
Mellersta och övre Norrland 7.1 ±2.4 1.3 ±1.0 10.4 ±3.0 5.9 ±2.4 42.7 ±4.9 8.4 ±2.7 3.6 ±1.8 19.0 ±4.1 1.6 ±1.3
Total 6.0 ±0.5 3.3 ±0.5 20.1 ±0.9 6.4 ±0.6 33.7 ±1.1 8.2 ±0.6 4.1 ±0.5 17.1 ±0.8 1.1 ±0.3

Description of the regions

Sydsverige: Blekinge County and Skåne County
Småland med öarna: Jönköping County, Kronoberg County, Kalmar County and Gotland County
Västsverige: Halland County and Västra Götaland County
Östra Mellansverige: Uppsala County, Södermanland County, Östergötland County, Örebro County and Västmanland County
Norra Mellansverige: Värmland County, Dalarna County and Gävleborg County
Mellersta och övre Norrland: Västernorrland County, Jämtland County, Västerbotten County and Norrbotten County

A table of regional survey results from the current and the previous electoral period is available in Statistics Sweden’s Statistical Database.

Statistics Sweden’s Statistical Database

Definitions and explanations

The Party Preference Survey in May 2020 was conducted with a national random sample consisting of 9 208 persons entitled to vote in a parliamentary election, without an upper age limit. Answers from these persons are collected through telephone interviews and online questionnaires. As in previous surveys, all available telephone numbers, both to mobile phones and landline phones, have been used to contact the selected persons.

Among the persons included in the sample, 27.9 percent could not be reached, 2.7 percent were unable to participate, and 16.3 percent did not wish to be participate. The total non-response rate was 46.9 percent. There were 4 888 respondents in total, which corresponds to 53.1 percent. A more detailed presentation of non-response is made available in connection with the publication on 11 June.

The data was collected during the period from 29 April to 27 May.

The results are presented in the form of point estimates ± margins of error. The interval that is formed by the percent estimate ± margin of error here is a 95 percent uncertainty interval, which is an interval that, with 95 percent probability, includes the actual value of the population, if no systematic errors occur.

Adjustment of estimation method

In May 2020, an adjusted method was introduced into the Political Party Preference Survey as a result of Statistics Sweden’s ongoing quality improvement of the survey.

The objective of the method adjustment was to adapt the estimation method to a larger non-response and to improve the precision of the estimates. In addition, this change improved comparability between the Election today estimate and the Political party preference estimate.

In connection with the current publication, the Statistical Database at Statistics Sweden was updated with recalculated estimates from November 2010 and onward. This enables comparisons over time without having to take into account the method adjustment. The results from November 2019 are also calculated using the new method in this statistical news.

At the total level, this adjusted method only leads to minor differences in estimates compared with previously reported results. The breakdown by sex shows some statistically significant differences, in particular for political parties in which there are great differences between support from women and from men. This is due to the addition of sex in the auxiliary information to the estimation method to improve the precision of the estimates.

The adjustments in brief

  • Auxiliary information was expanded for both the Political party preference estimate and the Election today estimate.
  • A transition was made to a common estimator for both estimates that was previously produced in different ways.
  • The imputation method for the Election today estimate was updated.

Further information on the adjustments is available at Justerad skattningsmetod i Partisympatiundersökningen – En beskrivning av de metodjusteringar som implementerats i maj 2020 (pdf).

Next publishing will be

On 11 June, information about political party preferences in May 2020 will be published, in total and in different demographic groups. Statistics Sweden has opted to split up publication in order to publish “Election today estimate” as quickly as possible after completed data collection, and also to make it clear that it concerns two different measurements. “Election today estimate” is measured using the question “Which party would you vote for if a parliamentary election was held in the next few days?” while the political party preference is measured using the questions “Do you prefer one of the political parties more than the others?” and “Which party do you prefer the most?”.

The publication on 11 June also contains information about EU and euro preferences, in total and in different demographic groups.

Statistical Database

More information is available in the Statistical Database

Feel free to use the facts from this statistical news but remember to state Source: Statistics Sweden.

Statistical agency and producer

Statistics Sweden, Democracy Statistics

Address
Solna strandväg 86
171 54 Solna
fax 08-506 947 72

E-mail
psu@scb.se

Enquiries

Regina Vilkénas

Telephone
+46 10 479 41 18
E-mail
regina.vilkenas@scb.se

Johan Eklund

Telephone
+46 10 479 45 38
E-mail
johan.eklund@scb.se