Independence - how would you vote?

[SpungoMcGoo]

Their methodology for the Indy referendum is completely different from all of the other pollsters, and obviously flawed

Their findings are completely different from all the other pollsters.

Have you wondered why all of the other independent polling data shows a Grand Canyon of a gap between Yes and No, and Panelbase shows just a very wide river? Is it a Unionist conspiracy?

If you want to call 8/9% a "very wide river", good for you. But you seem to have STILL not explained what's so wrong about their methodology.

[H_B]

Don't be bloody ridiculous. They're polling public opinion, that's what they do.

If you poll public opinion at a Scottish Conservatives party conference, you are polling public opinion.

Would that, or would that not, be a reasonable way to determine Scottish opinion on the Independence referendum?

[H_B]

It did indeed turn out to be wrong,

A child could have told you that it was wrong. Getting it wrong wasn't what made a total c**t of you. Getting it wrong whilst being sneering and condescending about it is what made a total and utter c**t of you.

Amusing though.

[Davi3j]

Their methodology for the Indy referendum is completely different from all of the other pollsters, and obviously flawed

Their findings are completely different from all the other pollsters.

Have you wondered why all of the other independent polling data shows a Grand Canyon of a gap between Yes and No, and Panelbase shows just a very wide river? Is it a Unionist conspiracy?

Goalpost shifting.....I was talking about past results

[Ad Lib]

You didnt realy explain why it isnt true though, did you? ... not having a long track record isint the same as getting it wrong compared to the other polls, is it?

It's only been compared with one solitary past result. It is therefore not "no less reliable" than polling companies that have thoroughly developed a detailed portfolio of polling skills and methodology over (in some cases) more than a decade.

[SpungoMcGoo]

A child could have told you that it was wrong. Getting it wrong wasn't what made a total c**t of you. Getting it wrong whilst being sneering and condescending about it is what made a total and utter c**t of you.

Amusing though.

You might have had a case for gloating had you pointed to the correct figures in the first place. But trust me, I won't be losing any sleep over the shame.

Anyway, still waiting for the flaws in Panelbase methodology. They DON'T poll at the Scottish Conservatives conference, they have a balanced and weighted demographic the same as everyone else. Are you going to actually explain, or just keep repeating the empty smear and hope everyone gives up?

[SpungoMcGoo]

It's only been compared with one solitary past result. It is therefore not "no less reliable" than polling companies that have thoroughly developed a detailed portfolio of polling skills and methodology over (in some cases) more than a decade.

I'm bewildered as to why you're insisting that the absence of proof is contra-proof.

[Ad Lib]

Don't be bloody ridiculous. They're polling public opinion, that's what they do. Whether they're recording public opinion about a tin of beans or a change to the constitution is utterly and completely irrelevant. The game remains the same, it's the pitch that changes.

Jeezo, I'm getting tired of your "I know fucking everything, me" attitude.

"Polling public opinion" is, at best, in this analogy, "hitting a ball with an implement". We could be playing tennis, lacrosse, badminton, squash, cricket, hockey, polo or shinty for all we know.

I sure do hope someone says what it is soon.

Jesus Christ it's been spelt out about 40 times. Their figures are derived from those who are certain to vote at Holyrood elections. Even though Holyrood election turnout (50% in case you're wondering) has absolutely f**k all to do with the demographics of campaigning, issue or behaviour of a referendum on Scottish independence.

[Ad Lib]

I'm bewildered as to why you're insisting that the absence of proof is contra-proof.

I'm not. There are two distinct arguments being made.

1. Panelbase does not have a track record yet. Their polling methodology should therefore not be blindly assumed as appropriate or good practice as a default position. Its sole performance so far has been to be roughly in line with other pollsters on Holyrood elections, which given their methodology focuses on Holyrood certainty to vote is probably not surprising.

2. Panelbase's methodology for independence referendum polling is wildly different from the established pollsters in the field, and they are essentially the solitary outlier in having the contest averaging somewhere between an 8 and 9% gap (or a 4-5% swing). All the other major pollsters have a trend gap which is larger than that, and hardly any of them have as remarkably high a "don't knows" category as Panelbase, which may in part be down to the way Panelbase collects its data (online, where "don't know" will be a drop-down menu or whatever option already, rather than verbally prompted over the phone, which is the preferred data collection method of (most of) the other pollsters).

3. Having regard to 1 and 2, and insofar as we should be skeptical of pollsters anyway, we should be particularly suspicious of anything that Panelbase turns up on referendum polling.

[H_B]

If you want to call 8/9% a "very wide river", good for you. But you seem to have STILL not explained what's so wrong about their methodology.

It has been explained time and time again on this thread.

By excluding from the survey those who have no interest in voting in the Scottish Parliament election, you are excluding Scots who think that the Scottish Parliament is a complete and utter waste of time and is a glorified town council, for example.

There are plenty of people who don't bother their arse to vote in the Holyrood election. They may or may not vote in the referendum.

[Davi3j]

This isn't true though. Panelbase don't have a long track-record in UK or Scottish polling, and they've only been vindicated as being within the margin of error of the other pollsters on one occasion in one kind of poll, namely the Holyrood elections. There is good reason to believe that their metrics, whilst about par for the course for Holyrood, are not adequate for an independence referendum. Chief amongst them being their methodology and weighting criteria, which has been criticised by people both sides of the divide (me and renton on the Yes side and H_B in the no camp).

It's only been compared with one solitary past result. It is therefore not "no less reliable" than polling companies that have thoroughly developed a detailed portfolio of polling skills and methodology over (in some cases) more than a decade.

Thanks for admitting your earlier statment was untrue, nice of you to clear that up

[H_B]

You might have had a case for gloating had you pointed to the correct figures in the first place.

I did.

I have quoted the Ashcroft figures several times on this thread already. I knew what they were. Hence when I said your claim was "utter lies" it was because it was utter lies.

[Baxter Parp]

If you poll public opinion at a Scottish Conservatives party conference, you are polling public opinion.

Would that, or would that not, be a reasonable way to determine Scottish opinion on the Independence referendum?

Sweet Jesus Christ, what an incredibly stupid thing to say. That would be a good way to determine Scottish Conservative opinion, not the general publics. Which is why polling companies wouldn't do such a thing. What idiot point do you think you're making?

[H_B]

Sweet Jesus Christ, what an incredibly stupid thing to say. That would be a good way to determine Scottish Conservative opinion, not the general publics.

So Scottish conservatives aren't part of the general public?

Your point is that Panelbase poll public opinion. Either it matters how they choose who they poll or it does not.

Hint: It does.

[Baxter Parp]

"Polling public opinion" is, at best, in this analogy, "hitting a ball with an implement". We could be playing tennis, lacrosse, badminton, squash, cricket, hockey, polo or shinty for all we know.

I did this shit at college. Polling companies adapt their game depending on what they're working on, just as players adapt to different pitches. You're wrong, just accept it and move on.

[Ad Lib]

Thanks for admitting your earlier statment was untrue, nice of you to clear that up

Negatives galore. Please read as "as reliable" for "no less reliable". Panelbase are less reliable than other pollsters because they don't have a track record. They predicted one election with fairly uncontroversial weighting criteria for that election, with results within the broad range of general established polling.

One swallow does not a summer make.

[Baxter Parp]

So Scottish conservatives aren't part of the general public?

Your point is that Panelbase poll public opinion. Either it matters how they choose who they poll or it does not.

Hint: It does.

Of course it does, I never said it didn't. They're a professional polling company, you're a complete tool. They should know what they're doing, you clearly don't.

[H_B]

I did this shit at college.

Your HNC isn't going to impress anyone here I'm afraid,

[SpungoMcGoo]

By excluding from the survey those who have no interest in voting in the Scottish Parliament election

I'm confused. 16% of respondents in their latest poll didn't vote in the 2011 election. That doesn't sound much like "excluding Scots who think that the Scottish Parliament is a complete and utter waste of time and is a glorified town council".

[Ad Lib]

Of course it does, I never said it didn't. They're a professional polling company, you're a complete tool. They should know what they're doing, you clearly don't.

That's just the point though. A spectrum of established pollsters, both those affiliated to and those not affiliated to, Panelbase's commercial competitors, have raised their eyebrows at their decision to base 2014 polling on those certain to vote at Holyrood elections. What they're saying, politely, is that Panelbase "doesn't know what it's doing".