Latest Polls and Latest Odds

[harry94]

What sort of things happened in the final few weeks of the Quebec referendum?

Will all those things happen in Scotland?

Will any that are replicated have the same affect?

That's basically why I don't think it's that straight forward.

This referendum is likely to have a high turnout and with so many people who don't engage with politics voting it's pretty hard to judge what could sway them either way. I wouldn't be surprised to see both campaigns giving out free muffins or something to people who don't care much and getting them in the booths.

[H_B]

Is that a universal trend across all comparable referendums?

This was certainly suggested during a radio debate last weekend about this. Polling experts declare a "conservative trend" which suggests there is a move away from the riskier element on polling day.

I'm not sure what examples they were citing in support of this claim. Quebec is one obvious example.

[Mr Bairn]

This was certainly suggested during a radio debate last weekend about this. Polling experts declare a "conservative trend" which suggests there is a move away from the riskier element on polling day.

I'm not sure what examples they were citing in support of this claim. Quebec is one obvious example.

I suppose it's the same trend that tends to see incumbent governments improve their polling figures over the last 6 weeks or so in most election campaigns. In 2010 the Conservatives were expected to win a majority a couple of months before polling day, in 2011 the SNP were behind with a couple of months to go and ended up winning a majority, in the AV referendum it was neck-and-neck a couple of months before polling day but in the end the no side achieved a 2:1 ratio, and in the 2007 election the SNP, who won by a single seat, were much further ahead a few weeks before the poll.

[ayrmad]

There is absolutely no chance IMO that only 29% of our voting population plans to vote Yes.

No chance whatsoever. It's far higher than that. I simply don't accept that particular poll is correct.

ETA: Oh FFS they are still including Don't Knows. Grrrrr. OK in that case it's predicting around 42% Yes and 58% No. That is one of the lower estimates for Yes but I suppose it's in a reasonable ballpark.

~ Double the Don't Knows of the majority of polls as usual.

[ayrmad]

YES touching 2/1 with some bookies, NO touching 4/11 with some, the bookies must be doing their own polls

[Mr Bairn]

There is absolutely no chance IMO that only 29% of our voting population plans to vote Yes.

No chance whatsoever. It's far higher than that. I simply don't accept that particular poll is correct.

ETA: Oh FFS they are still including Don't Knows. Grrrrr. OK in that case it's predicting around 42% Yes and 58% No. That is one of the lower estimates for Yes but I suppose it's in a reasonable ballpark.

What's your problem with DKs being included? Seriously?

Although I'm not sure of the methodology used that's creating these higher DK numbers. It could of course just be the sample they use has a far higher proportion of DKs than the other pollsters, however the methodology used by polling companies is designed to minimise the potential for large sampling biases to occur(ie If you took a completely different random sample, the results should, in theory, fall within 3% of the control group)

I personally suspect there's more to it than that. My guess is that this particular pollster automatically includes anyone that answers "DK" into the DKs column, whereas some of the others probably investigate it further, asking if they're leaning yes, leaning no, or genuinely haven't decided one way or the other, with only the latter being shown as their top line "Don't know" figures.

The only reason for excluding the don't knows is because it's the easiest way to measure the pollsters against one another. Truth is we probably won't know who is right until a few minutes after 10pm on polling day when the exit polling figures are released.

Edited April 16, 2014 by Mr Bairn

[H_B]

No it isn't. An interesting article somewhere listed Quebec as an outlier. Normally Yes has to come from substantially behind.

You'll have examples proving this of course?

Or is this claim to be filed alongside such gems as :-

"Most Scottish graduates move to England"?

[H_B]

As I have shown elsewhere, of 34 referendums in democracies since the 1980s for which the numbers are available, opinion shifted towards the status quo in the closing stages of the campaign in 23. Some shifts were enormous: three referendums have seen support for reform fall by 40 percentage points during the campaign; another seven have seen it fall by at least 20 percentage points. By contrast, support for change rose in only eleven cases, and most of these rises were tiny; only one exceeded 10 percentage points.

The reason for this pattern is that voters are cautious. If they are unconvinced of the need for change, they tend to stick with what they know. Thus, the “don’t knows” in early polls tend to split towards the status quo as polling day approaches. Sometimes, Yes support falls directly too: the idea of change often sounds appealing on first blush, but as voters consider it more the doubts build and many voters switch.

http://www.psa.ac.uk/insight-plus/blog/scotland%E2%80%99s-referendum-debate-what-really-going

Edited April 16, 2014 by H_B

[Mr Bairn]

What was the one about 4 parties being able to form the Scottish government? I'm detecting a new bullshitter in the clown collective

[vikingTON]

The reason for this pattern is that voters are cautious. If they are unconvinced of the need for change, they tend to stick with what they know. Thus, the “don’t knows” in early polls tend to split towards the status quo as polling day approaches. Sometimes, Yes support falls directly too: the idea of change often sounds appealing on first blush, but as voters consider it more the doubts build and many voters switch.

Except for, err, the 11 cases in which 'voters' presumably weren't 'cautious': a full third of your sample. So hardly the most convincing pseudo-psychological analysis of voting patterns then.

[H_B]

What was the one about 4 parties being able to form the Scottish government? I'm detecting a new bullshitter in the clown collective

1) The claim was that there are 4 parties who can "genuinely" be the major party in a Scottish election. Unlike in the UK, where there are, emm, only 2. Unfortunately oaksoft has yet to name these 4 parties. We can but wonder who they are.

2) Added to the claim that the majority of Scottish graduates move to England to work.

"There is a lot of truth in that.The majority of graduates in Scotland go to London or somewhere else in England."

There was also the claim that 70% of Scots wanted this referendum. Another claim with zero evidence.

A quite staggering collection of stupidity.

[Mr Bairn]

It's crazy that this guy isn't even in the top 5 thickest clown collective posters.

I genuinely fear for him in more intellectually advanced company. He must stick out like a sore thumb.

[BlackIsleBud]

I'm a C U N T

FTFY

[BlackIsleBud]

I'm a C U N T too. Just like my bestest fwiend HB.

[Confidemus]

I genuinely fear for him in more intellectually advanced company. He must stick out like a sore thumb.

Is the retail store where you work a hotbed of intellectual thinking then?

[Mr Bairn]

I'm a student at a Russell group university. Intellectual debate tends to dominate any night that I'm sober.

[Confidemus]

I'm a student at a Russell group university. Cry wanking tends to dominate any night that I'm sober.

FTFY

[Mr Bairn]

^^^ Says the poster who lied about the situation of many sick and disabled people to score points for the yes campaign

[Mr Bairn]

What's wrong with security guards?

And low sixties. Nearly 3/4 of the way to a 2:1

[Confidemus]

^^^ Says the poster who lied about the situation of many sick and disabled people to score points for the yes campaign

You really are the most swivel-eyed of loons.