Manfred Güllner, the head of the opinion research institute Forsa, explains what was wrong in the polls before the state elections in Saxony-Anhalt. Who do the wrong prognoses fall on their feet?
The opinion polls for the state elections in Saxony-Anhalt partly foresaw a head-to-head race between the CDU and AfD. It turned out quite differently. The CDU landed at 37.1 percent, the AfD at 20.8. How is that explained?
You have to see that this supposed head-to-head race was not announced by all institutes that conducted surveys before the election, but only by two institutes. The research group Elections, which, as in many previous elections, showed the smallest deviation, had a lead of the CDU over the AfD of seven percentage points and communicated this on ZDF. That there is such a clear deviation from the final result will happen again and again. That has happened in the past. Just think of the federal election of 1965, in which polls before the election reported a neck-and-neck race between Willy Brandt and Ludwig Erhard, but Erhard won the election by eight percentage points. What is not allowed is that here in Saxony-Anhalt a head-to-head race was predicted that was completely unrealistic. 28 percent for the AfD – that would have been an election sensation, but there was absolutely no basis for blowing such a number into the world.