Surveys reliable? – After the EU elections: fuss about the pollsters

Date:

The FPÖ obtained “only” 25.36 percent. Before the elections that was even 30 percent. The criticism of fallibility is becoming louder again. The researchers are fighting back.

What percentage may that be? 30 or 25, 23 or 19? After election night, pollsters are once again the focus of criticism. The FPÖ has been far ahead in all investigations for months. For the National Council and EU elections. In the end they won just ahead of the ÖVP. The SPÖ in third place.

What’s happening there? Not much for pollster Peter Hajek (Unique). Before the elections he had an FPÖ of 30 percent. “We overestimated them a bit,” he says. You have to take the fluctuation range into account. That was 26 to 34 percent. The FPÖ came in at 25.36. “That was admittedly a little bit different. We were absolutely right about the other parties.’ The explanation: the DNA was not on the radar. She basically came out of nowhere. And discusses an FPÖ problem with Corona skepticism. Second, many potential Blue voters did not vote.

“Overall well located”
The fluctuation margin is not made transparent enough when published, says Christoph Haselmayer (IFDD), because the media want concrete figures.

“It is not made clear enough here that this is an estimate. “In addition, there is research when it is clearly outside the fluctuation range,” explains communications scientist Matthias Karmasin. Overall, we are in a good position going into these elections. In addition to the 30 percent for the FPÖ, there were two more outliers at another institute: 19 percent for the ÖVP and once 15 percent for the NEOS (ultimately 10.14).

Exciting polls on election day
Trend research based on election surveys by Foresight, ARGE Elections and Peter Hajek before election day also caused surprise. At 5 p.m., 27 percent registered for the FPÖ. ÖVP 23.5, SPÖ at 23. With a fluctuation range of 2.5 in the green area, the researchers argue.

Peter Hajek gives very specific figures here: “95 out of 100 studies fall within the fluctuation margin. So correct. In general you have to say. The research consists of 40 percent craftsmanship, 40 percent experience and 20 percent luck.” It is an approximation of reality, not precision landings. The best way to achieve this is through so-called exit polls, says Haselmayer. This means: national surveys immediately after the votes have been cast on the spot. “But that is far too expensive.”

Source: Krone

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Share post:

Subscribe

Popular

More like this
Related