Ad-Hoc studio in a building, seen from above

ARD Election day studio 2009, Photo: Michael (CC-BY-NC-SA 2.0:

Die Wahl ist vorbei, schwierige Koalitionsverhandlungen stehen bevor. Wer gibt wo nach? Welche rote Linie kann man vielleicht doch überschreiten? Die ein oder andere Frage, um die es da gehen wird, wurde ja bereits in unserem Wahltest gestellt, mit dem man die konkreten Ausssagen der Wahlprogramme mit der eigenen Position vergleichen konnte. Da wäre doch interessant zu wissen, wie die rund 400.000 Nutzer*innen des Wahltest zu diesen Fragen stehen. 

How well did the programmes perform?

First, the most important question that the election test can answer: How well do the election programmes meet the taste of voters in need of orientation (for whom they were written). In this chart, the sum of the agreements is higher than 100%, since it is possible to have an agreement with several parties per question. Values below 25% are unusual here - there is always a match somewhere. A party with significantly higher values can obviously appeal to many people with its programme, while a party with values barely above that has a rather controversial programme. It is difficult to compare top scores because they depend strongly on the number of parties and answer options - the more questions are asked, the smaller the differences. In essence, the higher the score, the more popular the party. Against this background, the comparatively poor performance of the CDU/CSU programme in this election is interesting.

Environment and Climate Issues

In the area of environmental protection and climate, the results of our voting test can be summarised as follows: The majority of our users are in favour of climate protection measures - but reject them if they directly restrict their own way of life. While an overwhelming majority of 85% of our users want to bring forward the German coal phase-out to 2038 or 2030 (57%) (Germany's current target is 2050) and 55% are in favour of switching Germany's electricity supply completely to renewables by 2040 at the latest, only 28% are in favour of a ban on new registrations of cars powered by internal combustion engines. The majority of users (59%) also reject a speed limit (currently, ARD reports a majority in favour of a speed limit, but this is certainly also due to the benevolent questioning of whether a speed limit as a measure for climate protection is going in the right direction). The positions on a ban on new registrations and the exclusion of a speed limit, which were formulated rather vaguely in the exploratory paper of the traffic light coalition, would at any rate coincide with the opinion in the election test.

Economy and Employment Issues

The topic of economy and work brings more surprises:
While it was to be expected that many workers would acquire a taste for working at home during the various lockdowns (78% are in favour of a right to a home office), the majority of our users (59%) are against an entitlement to unemployment benefits without a prior twelve-month deposit. A result that is surprising given that the parties in favour of this demand, the SPD, the Left and the Greens, had the highest agreement in our election test - apparently not with this position.

Migration and Integration Issues

Auch dieser Bereich bietet Ergebnisse, die mit den Erwartungen brechen: Das zuletzt deutlich gestiegene Niveau der deutschen Entwicklungshilfeausgaben wollen ganze 77% beibehalten, 23% davon sind sogar für eine Erhöhung der Ausgaben. Ganze 59% der Nutzer*innen lehnen dagegen eine Aufhebung der seit 2015 geltenden Grundrechts-Einschränkung beim Familiennachzug ab. Hier drängen sich Parallelen zum Themenbereich Klima auf: Humanitäre Maßnahmen finden nur dann großen Anklang, wenn die Maßnahmen nicht vermeintlich ins eigene Leben einschneiden – wenn etwa mehr Geflüchtete in die Nachbarschaft nachziehen könnten. 

Political Participation Issues

The trend towards astonishing results continues: 62% of our users reject lowering the voting age to 16 - although this is a declared goal of the SPD as well as the Greens, Left and FDP, i.e. the four strongest parties in our election test. It is possible that this result gives a small indication of the demographic distribution of our users (which we do not record). The responses on the level of participation are also interesting. With 51%, the majority would like to see more referendums, but only 7% are in favour of citizens' councils as a means of state decision-making. How effective citizens' councils can be is shown not only by projects in Ireland, but also by a pilot project initiated by the association More Democracy (Mehr Demokratie e.V.), which we were allowed to support.

Social Issues

The topic area of social affairs offers little reason for surprise: the majority of users (57%) are in favour of free daycare places, almost all are in favour of stronger regulation of rental costs (only 11% are against) and an increased minimum wage is also supported by 83%, with 45% even in favour of a minimum wage of 13 euros. The test results on basic child security (with 65% the majority is in favour) and the taxation of people with a very high annual income (only 22% are in favour of a reduction, the rest are in favour of an increase) are also in line with the high agreement values of the SPD, the Greens and the Left.

Old-Age Security Issues

Der Themenbereich Rente / Alterssicherung war in den Wählerbefragungen, auf die wir unsere Themenauswahl stützten, zweitweise das am dritthäufigste genannte Thema. Die Ergebnisse des Wahltest in diesem Themenbereich überraschen vor dem Hintergrund nicht und belegen, dass sich auch die Wahltest-Nutzer*innen um die Alterssicherung Sorgen machen. Für alle Änderungsvoschlägen ergeben sich hohe Mehrheiten, die höchste von 92% für eine Festschreibung eines Mindestrentenniveaus. Bei dieser Frage ist bemerkenswert, dass sogar eine deutliche Mehrheit von 64% die weitergehende Forderung nach einer Festschreibung bei 53% unterstützt. Das ist übrigens ein Ergebnis, das bei binären Wahlberatungstools wie dem Wahl-O-Mat nicht ermittelt werden könnte. Durch die künstliche Verknappung der politischen Angebote auf zwei Optionen ("ja/nein") hätte man hier nur allgemein nach einer Festschreibung oder konkret nur nach einem bestimmten Wert fragen können. 

Health Issues

Since health policy is not one of the most dynamic issues beyond Corona, it is striking that the two questions asked are already familiar from previous elections. There was also little change in the majorities. In the election test, 63% were in favour of a citizens' insurance scheme (similar to infratest dimap, where the figure was recently as high as 69%). According to the exploratory paper, this will probably not happen, but the legalisation of cannabis, also supported by the majority, will. It would be nice to be able to devote oneself to new policy projects in the health sector during the election tests.

Foreig Policy Issues

Foreign policy is the only policy area we included in the electoral test, although it was not among the ten most frequently mentioned in the survey on the most important issues in Germany. It replaces the "Corona" issue complex we omitted, which did not appear as such in the election programmes geared to the next 4 years and was partly reflected in other policy areas. In addition, for the first time a party represented in the Bundestag demanded an exit from the EU and put it to the vote. This question seemed too existential to be left out, especially in view of the consequences of Brexit for Great Britain. With this demand, however, the AfD achieved hardly any resonance beyond its supporters in the election test. This was in contrast to the Greens, who were the only ones to demand that the Nord Stream 2 natural gas pipeline not be put into operation, and who were able to rally at least 41% of the voters behind this issue.

What does the votetest recommend?

The electoral test only finds matches between the election programmes and only 25 selected questions. This is not yet a recommendation, because of course other aspects play a role - especially previous actions, top personnel or existing ties - and that is a good thing. Nevertheless, it is interesting to see how often a party most closely matches our users' own convictions. We recorded this for the first time in the 2021 election test. You probably won't be able to read much from it, but this will improve with further evaluations in future election tests due to the possibilities for comparison. What is striking, however, is the high level of agreement for the left - while the real election result dropped. One explanation could be the focus of this year's election test on the top issues such as climate protection and social policy, where the left usually represents the most far-reaching positions.

Is all this representative?

No, because the answers in the election test are stored anonymously, we cannot determine demographic data and therefore cannot ensure that the figures cover a broad population group. In addition, the 450,000 visitors became aware of the electoral test mainly through word-of-mouth, especially by sharing results on social media. This so-called "self-recruitment" can lead to distortions, e.g. if supporters of parties whose convictions are otherwise less reflected in other election consultation tools find their results particularly surprising than others and share them more often, thus mobilising more like-minded supporters (another possible explanation for the good performance of the Left Party). Nevertheless, the data are not worthless: on comparable questions, the results of the election test are not far removed from the figures also collected by opinion research institutes. Perhaps the same applies to the other questions.