tag:blogger.com,1999:blog-930120922627919768.post5437842322709983541..comments2024-03-29T05:53:21.060+00:00Comments on SCOT goes POP!: Quite possibly the most bizarre opinion poll finding in historyJames Kellyhttp://www.blogger.com/profile/01516007141763230886noreply@blogger.comBlogger4125tag:blogger.com,1999:blog-930120922627919768.post-17585548420550343042010-04-18T15:58:34.212+01:002010-04-18T15:58:34.212+01:00Jeanne - yes, I know that YouGov don't weight ...Jeanne - yes, I know that YouGov don't weight their subsamples properly, although that's the case for most pollsters. As far as I'm aware, only Angus Reid weight their Scottish subsamples separately.<br /><br />You're quite right that the YouGov figures I've given above aren't untypical of recent weeks, and therefore we shouldn't (yet) jump to the conclusion that the rigged TV debate has changed anything in Scotland.James Kellyhttps://www.blogger.com/profile/01516007141763230886noreply@blogger.comtag:blogger.com,1999:blog-930120922627919768.post-38782920584043167192010-04-18T14:03:57.569+01:002010-04-18T14:03:57.569+01:00Anon. We're not important enough I guess. No o...Anon. We're not important enough I guess. No one is prepared to pay for it.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-930120922627919768.post-71678448047223706002010-04-18T11:57:11.820+01:002010-04-18T11:57:11.820+01:00Why can't they do a full Scotland only poll??!...Why can't they do a full Scotland only poll??!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-930120922627919768.post-88803193870247872010-04-18T05:59:23.216+01:002010-04-18T05:59:23.216+01:00James, YouGov weightings for Scottish subsamples i...James, YouGov weightings for Scottish subsamples in UK-wide polls is totally off--and they admit it. They stated (you can find this on their website) that it wasn't worth bothering correcting the weightings. It is at least 4 percentage points off. <br /><br />That doesn't keep them from adding these totally flawed polls together and publishing them as though they were meaningful. When you put that together with the fact that a sample the size of the Scottish subsample has a margin of error of about 10% anyway--You end up with SNP and Labour a heck of a lot closer than the propaganda would suggest. <br /><br />In fact, those numbers aren't necessarily bad for the SNP at all.J. R. Tomlinhttps://www.blogger.com/profile/01109874615059334200noreply@blogger.com