redsquirrel
This Machine Kills Progressives
Depends on polling company I think. Pretty much all made changes after 2015 but IIRC different companies made slightly different changes. Not sure if Britain Elects has made some changes to account for the 2015 inaccuracies, probably not as you'd then be double counting.Actually, while I am aware that the methods of compiling data have changed, I can't find any information on how exactly - does anyone know this?
If it has been weighted to account for the possibility that Labour turnout tends to be lower than polls suggest, it could be that the polls are wrong in the other direction this time. Labour voters seem more highly motivated this time, so changes made after 2015 could be based on an assumption that no longer holds true.
EDIT: recap here.
Since then polling companies have made changes to try and address that problem. Different companies have taken different approaches. The most significant though are a mix of adding new controls on samples by education and interest in politics and changes to turnout models. We obviously won’t know until the election has finished whether these have worked or not.