The biggest, wealthiest, most advanced polling firms in the US couldn’t get it right either, horrendously so in recent memory. I don’t think that’s down to conspiracy, but how to weight your samples when so many people don’t want to answer a phone call from an unusual number.
The biggest, wealthiest, most advanced polling firms in the US couldn’t get it right either, horrendously so in recent memory. I don’t think that’s down to conspiracy, but how to weight your samples when so many people don’t want to answer a phone call from an unusual number.
I know a lot of firms missed on the 2016 election, but the last two (plus the midterms) have all been within the margin of error.
What’s an example of them getting it wrong?