renton Posted July 23, 2015 Share Posted July 23, 2015 You absolutely did handpick a poll to support your nonsense claim that Yes 'started out at 20-25%'. You've admitted so yourself. Perhaps it would be useful if you could list the starting position of Yes with all of these BPC posters please. I absolutely did not handpick a poll, I actually merely pointed to several different polls from different pollsters that showed yes below 30 points, including SSAS, Ipsos and Yougov. Define starting position, polling methodologies from 2011 to 2012 varied massively prior to the Edinburgh agreement defining the question. Link to comment Share on other sites More sharing options...
H_B Posted July 23, 2015 Share Posted July 23, 2015 Polling in 2012 .. With DK excluded (as was of course the case in your final figure of 44.7%) Jan 2012 - 44% Feb 2012 - 47% June 2012 - 37% July 2012 - 44% Aug 2012 - 44 % Oct 2012 - 34 % Oct 2012 - 35 % Oct 2012 - 45% Link to comment Share on other sites More sharing options...
H_B Posted July 23, 2015 Share Posted July 23, 2015 I absolutely did not handpick a poll, I actually merely pointed to several different polls from different pollsters that showed yes below 30 points, including SSAS, Ipsos and Yougov. Define starting position, polling methodologies from 2011 to 2012 varied massively prior to the Edinburgh agreement defining the question. Below 30 does not equal 20-25 now does it? Produce the starting point for each BPC poster please. We'll ignore for the moment the elephant in the room which is you mangling together polls with don't knows and without. Link to comment Share on other sites More sharing options...
H_B Posted July 23, 2015 Share Posted July 23, 2015 You hate admitting you are wrong. I get that. But you got a bit carried away here.. That's all. Yes absolutely did not begin life at 20-25 %. Why don't you just come clean and admit that was rubbish? Link to comment Share on other sites More sharing options...
renton Posted July 23, 2015 Share Posted July 23, 2015 Below 30 does not equal 20-25 now does it? Produce the starting point for each BPC poster please. We'll ignore for the moment the elephant in the room which is you mangling together polls with don't knows and without. Read my original point again, I said it started in the range 25-30%. The 20-25 is your own fiction. Link to comment Share on other sites More sharing options...
renton Posted July 23, 2015 Share Posted July 23, 2015 You hate admitting you are wrong. I get that. But you got a bit carried away here.. That's all. Yes absolutely did not begin life at 20-25 %. Why don't you just come clean and admit that was rubbish? As I said above the 20-25 is your number, not mine. XBL may have mentioned 23% I absolutely did not. Link to comment Share on other sites More sharing options...
H_B Posted July 23, 2015 Share Posted July 23, 2015 Read my original point again, I said it started in the range 25-30%. The 20-25 is your own fiction. Apologies. You are correct. You stated 25-30 . That is still nonsense however. Please provide the starting points for each pollster you are using as evidence For example IPSOS MORI has Yes at 39% in.. January 2012! Link to comment Share on other sites More sharing options...
renton Posted July 23, 2015 Share Posted July 23, 2015 Apologies. You are correct. You stated 25-30 . That is still nonsense however. Please provide the starting points for each pollster you are using as evidence For example IPSOS MORI has Yes at 39% in.. January 2012! Asking what question? First quesyion in sequence? With or without dks? How does that fit into the general trend? Link to comment Share on other sites More sharing options...
H_B Posted July 23, 2015 Share Posted July 23, 2015 Asking what question? First quesyion in sequence? With or without dks? How does that fit into the general trend? Given you are comparing the rise with the final figure which of course means no DKs the earlier figures also have to exclude DKS. Though the Jan 2012 IPSOS figure of 39% I gave you actually doesn't and its still 39%. Give me both. Including and excluding DK. Oh and the date of the poll. Link to comment Share on other sites More sharing options...
H_B Posted July 23, 2015 Share Posted July 23, 2015 If you wish to use only those polls with the exact wording of the final question , given that's what you are comparing the results to, that will be fine. Which will exclude all those before Nov 2012 I think it was Link to comment Share on other sites More sharing options...
H_B Posted July 23, 2015 Share Posted July 23, 2015 If you want to give me the earlier 'Should Scotland be.' results like the Jan 2012 IPSOS one that's OK too Link to comment Share on other sites More sharing options...
xbl. Posted July 23, 2015 Share Posted July 23, 2015 You know, I always wondered if there was a textual demonstration of the word "mewling". Link to comment Share on other sites More sharing options...
renton Posted July 23, 2015 Share Posted July 23, 2015 Given you are comparing the rise with the final figure which of course means no DKs the earlier figures also have to exclude DKS. Though the Jan 2012 IPSOS figure of 39% I gave you actually doesn't and its still 39%. Give me both. Including and excluding DK. Oh and the date of the poll. No they don't - it's a particularly brutal methodology with the phone polls/face to face polls given the huge numbers of DKs in those types of polls. Edinburgh agreement wasn't until October 2012, so until then no one knew if it was a single or two questions, no one knew the wording and no one knew the likely franchise. Therefore any polling prior to then is subject to massive error of who and what was asked, in what sequence and what part of any larger poll it was. In October 2012, you have YouGov and Ipsos showing Yes between 29-31% and PB at 37% immediately post to the Edinburgh agreement. Given that this is the logical starting point for the referendum polling, nothing there directly contradicts anything I've said (given these are snap shots with +/- 3% error), and are in line with the graph shown above going back to January 2013. Link to comment Share on other sites More sharing options...
H_B Posted July 23, 2015 Share Posted July 23, 2015 In October 2012, you have YouGov and Ipsos showing Yes between 29-31% and PB at 37% immediately prior to the Edinburgh agreement. Given that this is the logical starting point for the referendum polling, 3. The logical point is it? Why didn't you pick Jan 2012? That was also before the Edinburgh Agreement and before the wording was agreed. Oh that's right.. Because you don't like the IPSOS figures from Jan 2012 which asked the exact same question they did in Oct 2012! Or indeed June 2012. Whose results you also don't like as they don't fit your narrative. So when you said Yes 'started' at 25-30% what you actually meant was that they started at this point at an arbitrary point of your choosing in 2012 , using a different question to that asked in the end and including DKs in the figures? Link to comment Share on other sites More sharing options...
renton Posted July 23, 2015 Share Posted July 23, 2015 The logical point is it? Why didn't you pick Jan 2012? That was also before the Edinburgh Agreement and before the wording was agreed. Oh that's right.. Because you don't like the IPSOS figures from Jan 2012 which asked the exact same question they did in Oct 2012! Or indeed June 2012. Whose results you also don't like as they don't fit your narrative. So when you said Yes 'started' at 25-30% what you actually meant was that they started at this point at an arbitrary point of your choosing in 2012 , using a different question to that asked in the end and including DKs in the figures? Read what I said (edited to add: actually don't - I said prior to the agreement when I should've said post), any polling prior to October 2012 is subject to all kinds of theoreticals, poor questions and questionable weightings. Only after then can any poll start to be taken seriously for the referendum to be fought, not some hypothetical situation. Even then pollsters chopped and changed throughout, which is why I try not to look at one pollster in isolation and why I don't like the average poll of polls - it creates a false average between disparate pollsters that can obscure any meaningful trends. There is no doubt to me that YG's Kellner correction had a role in that now famous 51/49 poll, where the other pollsters were showing a narrowing of No leads, they over shot because the Kellner correction made their methodology very sensitive to opinion changes in one sub sample. You also have different weightings - some using turnout filters, others not. The types of polling makes a big difference as well, excluding DKs in PB and YG doesn't make a huge difference as there never was that many to start with, whereas excluding them from TNS/Ipsos does, as their polls regularly returned something like 20% DKs. So when I say yes is on 25-30% at the start, I mean the start of the referendum proper, after October 2012 when pollsters began using some semblance of the actual question, although not even then the actual question (that practice didn't take hold until well into 2013). When the franchise was known and the pollsters started asking it on a regular enough basis to show clear trends. At this point, as shown by the graphs I've previously posted, you see an increase of 10%-30% dependent on which pollster you use, I've overstated nothing, and all you have is your own conviction - shorn of any evidence - that Yes were awful, beaten comprehensively and you were right all along as you always are, on every subject. That's the long and short of this. Link to comment Share on other sites More sharing options...
DeeTillEhDeh Posted July 23, 2015 Share Posted July 23, 2015 While all of this might seem interesting it's pretty irrelevant imho. Apart from the fact that most polling organisations have changed their methodologies over this time (like comparing apples and oranges) there's also the fact that the one hard poll (the referendum) had a margin larger than the polling organisations were predicting. I suspect that there may be more changes to methodologies in particular to deal with DKs or those reluctant to give an opinion. My own gut instinct is telling me (like it did before the referendum) that pollsters are underestimating the number of secret No voters. Link to comment Share on other sites More sharing options...
renton Posted July 23, 2015 Share Posted July 23, 2015 While all of this might seem interesting it's pretty irrelevant imho. Apart from the fact that most polling organisations have changed their methodologies over this time (like comparing apples and oranges) there's also the fact that the one hard poll (the referendum) had a margin larger than the polling organisations were predicting. I suspect that there may be more changes to methodologies in particular to deal with DKs or those reluctant to give an opinion. My own gut instinct is telling me (like it did before the referendum) that pollsters are underestimating the number of secret No voters. Recent polls use recalled referendum vote so force folk into giving that preference before weighting by it. Designed to shake out such voters. Prior to that change but post referendum the same pollsters were showing 50-52 in favour of yes, after that (imo sensible) change they are sitting at 47-48 yes, a modest increase since the referendum. Which given the margin by which most pollsters over shot the actual result is consistent with the offset introduced by that methodology change suggests these secret No voters are worth about 3 points (ish) Link to comment Share on other sites More sharing options...
Mr Bairn Posted July 23, 2015 Share Posted July 23, 2015 A bit like why he claims to be an SNP voter. Which fools absolutely nobody. Did he even claim to have voted SNP this time? Link to comment Share on other sites More sharing options...
Fide Posted July 23, 2015 Author Share Posted July 23, 2015 If anything has been demonstrated on this thread, it's that H_B sure knows how to party. Link to comment Share on other sites More sharing options...
Gordon EF Posted July 23, 2015 Share Posted July 23, 2015 While all of this might seem interesting it's pretty irrelevant imho. Apart from the fact that most polling organisations have changed their methodologies over this time (like comparing apples and oranges) there's also the fact that the one hard poll (the referendum) had a margin larger than the polling organisations were predicting. I suspect that there may be more changes to methodologies in particular to deal with DKs or those reluctant to give an opinion. My own gut instinct is telling me (like it did before the referendum) that pollsters are underestimating the number of secret No voters. I reckon the No demographic probably didn't get picked up as much by pollsters as easily as the Yes demographic. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.