You are reading a single comment by @Oliver Schick and its replies. Click here to read the full conversation.
  • Remember where you were on 7 May 2015? Remember how the final polls had the two main parties neck and neck, and then came that 10pm exit poll predicting Labour’s Scottish wipeout and putting the Tories just a few seats short of an overall majority?

    On polls:

    I mentioned this before, but worth repeating: weighting of (most) polls has been changed since 2015 (I think survation is still using its old model). Labour supporters are taken to be less likely to vote, and this is taken into consideration. This can have three impacts, ranked in what I think are probably the most likely scenarios:

    1) Polls will be more accurate.
    2) Polls will under estimate Labour support (youth does turn out).
    3) Polls will once again over estimate Labour support (turn out is once again even lower than estimated).

    What is important is how national support ends up being shared within individual constituencies.

  • Yes, she only alludes to 'lax turnout models' without making going into them in detail, it's a useful reminder. I still have no idea what or whom to trust on this, though.

About