[ad_1]
The Civey Voting Institute works with a fully automated multi-stage process. All representative real-time surveys are reproduced on a German network of more than 20,000 websites (“river sampling”), so not only SPIEGEL users are surveyed. Anyone can participate in online surveys and their responses will be included in the representative result, provided they have registered. From these users, Civey extracts a cited sample that ensures that it corresponds to the population in terms of age, sex and population density, for example. In a third step, the results are weighted according to other sociodemographic factors and voter values in order to correct for distortions and avoid manipulation. More information can also be found in the Civey FAQ.
Why is it necessary to register?
The register helps to weight the responses and thus enables a representative survey result of the voting population in Germany. Each participant is asked about their gender, year of birth, and place of residence. Then everyone can give their opinion in more polls on different topics.
How are the representative results?
The response of each participant is weighted so that the result of a survey is representative of the population. For Sunday’s question and the government monitor, this population includes the eligible population in Germany. Weighting is done fully automatically based on personal information provided during registration and a user’s past response history. More details on the methodology can be found in the Civey white paper.
Can you reach enough participants online?
Opinion polls are generally conducted over the phone or online. For the results to be meaningful, it is critical to know how many people can be contacted and how many actually participate in a survey when spoken to. Internet connections and connections to landlines are currently equally widespread in Germany: in around 90 percent of households each, and mobile phones by up to 95 percent. Willingness to participate is in the single-digit percentage range for all methods, and experts rate them particularly low for telephone surveys.
With both methods, there is a group of people who cannot be contacted because they have no connection to the respective network or do not want to participate in the survey. That is why you always have to address a large number of people to get a meaningful result. In addition to SPIEGEL, Civey’s surveys are currently embedded in more than 20,000 websites, including various media. This ensures that as many population groups as possible can be easily reached.
How do I know the quality of a result?
Before a survey result is representative, a sufficient number of different people must participate. Civey clarifies whether this has already been achieved by specifying a statistical error probability for each survey result. The number of participants and the survey time are also published for each survey.
What does it mean when the colored areas in the graphics overlap?
In our graphs, the statistical error is shown as a color range. This interval shows the uncertainty associated with a survey value. For example, with the Sunday question, you cannot say exactly what percentage a party would get in an election, but you can specify an interval in which the outcome is most likely. If the intervals of two survey values overlap, then, strictly speaking, no statements can be made about the difference. For Sunday’s question, this means: if the polls of two parties are so close that their error intervals overlap, it cannot be deduced which of the two would do better in the election.
What about my data?
Users’ personal data is encrypted and stored on German servers and remains secret. Civey employees only work with user IDs for reviews and cannot connect users with their feedback. Users’ personal information is used primarily to weight responses and ensure that surveys are not tampered with. To avoid this, Civey uses both statistical and technical methods. In addition, Civey works with external partners who create target groups for advertisers. Your responses can only be used by the partner to model these target groups if users have accepted the data protection declaration of both Civey and an external partner. However, a member does not receive any information about his political and religious attitudes or those with which he can identify. Civey users also do not receive advertising based on their responses. As a logged in user, you can object to this being passed on to partners at any time. You can find more information on data protection at Civey here.
Who is behind the Civey polls?
At this point, mobile / landline website and app readers have the opportunity to participate in a representative Civey survey. Civey is a Berlin-based online opinion research institute. To collect its representative surveys, the company’s software, founded in 2015, connects websites to form a survey network across Germany. In addition to SPIEGEL, the “Tagesspiegel”, “Welt”, “Wirtschaftswoche” and “Rheinische Post” are also included. Civey was funded by Investitionsbank Berlin’s ProFit funding program and the European Regional Development Fund.
