Brexit Poll 1/2/16

[Mario149]

turbobloke said:

If 'what a group of people think' was relevant, then it would be relevant.

I'll remember that the next time I have need of legal advice. I'll engage a firm, listen to what the consensus of the lawyers I have hired say, then ignore them and do my own thing.

[Mario149]

don4l said:

Could you explain why you think that the opinions of a statistically insignificant group of economists are of any relevance?

History would suggest that they are nearly always wrong.

Did you see Eagle on the telly last night.

I think that we can agree that she did a fantastic job.

Because their economic predictions align with every major economic authority and hold more water than an a statistically irrelevant group consisting of BoJo, Nige and Mike who are telling us it will all be fine.

They're not nearly always wrong, otherwise they wouldn't have jobs and people would train a monkey to push a key to make the decision and save a shed load of money.

With you on Eagle though, saw her last night, absolutely dire.

[Mario149]

turbobloke said:

Mario149 said:

Murph7355 said:

Absolute nonsense.... With hindsight (I'll come back to that) we categorically know it was wrong to state anything definitive from the polls as the polls ended up not reflecting what actually happened in the GE.

And yet, despite being catastrophically wrong, they were in fact only 5% or so out.

Which completely altered the outcome. Outcomes matter.

Look at it that way and your perspective improves

That may be relevant if it was an even-ish split, except if you remove 5% from the stats we're looking at, you still have ~70%+ predicting the same outcome.

[Mario149]

CaptainSlow said:

Mario149 said:

So if it's that bad, *show* me the bias and how all the Leave economists were left out! Show me how Leavers did not respond because of X,Y and Z. I'm happy to concede the poll is unlikely to be bang on, but just choosing to stick your head in the sand and say it's not relevant is naive at best. Especially when combined with the fact that no significant economic authority agrees with you. Given Leave is apparently going to be economic nirvana after a few years, you would think that even if the poll was giving dodgy results, we'd see a little more support for that cause.

I don't need to show you the bias, you need to read up a bit more about sample testing. That's the point. This incredibly low response rate would lead anyone conducting a serious sample test to reject the results. The results may be right, the results may be wrong...nobody knows. It's not naive, it's the way any testing works and to use results from a 17% response rate would lose anyone their credibility.

A 17% response rate is faaaaaaar more than you get for many other polls. By your logic, we should be ignoring all of them and they're all a waste of time.

CaptainSlow said:

Coupled with the way the Remain camp word the results "90% of economists say...." when it should be "90% of economists who stated an opinion said..." is misleading. A number of companies have used similar wording in similar situations and have been pulled up by Advertising Standards.

How about 100% of economic authority groups of note who have been sampled agree that Brexit would be damaging to the UK economy.

[Mario149]

don4l said:

Mario149 said:

don4l said:

Mario149 said:

CaptainSlow said:

It is absolutely a flaw in the results of this poll. As response rate of 17% would fail any statistical testing, this is fact, so indeed the poll isn't right or wrong...it's just not worth considering.

Oh come off it, you're starting to sound silly now. 17% / 630+ responses is perfectly good enough to get a good idea of what a group of people think. We're not talking about asking 5 people out of a group of 25. The GE polling is always done on a sample of about 1,000 or so people (or a few hundredths of 1% of the population), is far more complex to process and even with its errors in 2015 the polls were only about 5% out.

No, it is you who is looking silly. If you had actually read the report you would have noticed the following:-

IpsosMORI said:

Technical note

639 respondents completed the online survey between 19th to 27th May 2016. The survey was undertaken online with invitations sent out to non-student members of the Royal Economic Society (RES) and the Society of Business Economists (SBE). Reported figures are based on unweighted data and so should only be taken as representative of those who responded. Overall 3,818 invitations were sent out, with a response rate of approximately 17%. For questions where respondents were asked to provide reasons for their answers, respondents were presented with a pre-defined prompt list.

They are quite clear. The data are unweighted and therefore cannot be extrapolated.

I did read it. So as I said a couple of posts up, if the unweighted data is so bad and significantly excludes those in favor of Brexit, show me. And if the poll is so useless and not representative, why are IposMORI even publishing it? No wait, I know, they're a lefty organisation right?

It would appear that 90% of people on this thread know that you are completely ignorant about statistics.

IpsosMori published the data because they were paid to do it. When they published it they said that the response rate was too low to draw any conclusions. They specifically warned that the data was unweighted, and therefore we should not try to extrapolate the data.

The vast majority of the invited participants were members of the Royal Society of Economists. 60% of their members are not even UK based. There does not appear to be any requirement to be an economist to join.

As Peter north pointed out, the remain campaign is an exercise in ignorance. Your continued insistence that 88% of economists want to remain just proves that he is absolutely correct.

Fine, you think the poll was flawed and useless. So be it. Now show me a better source of information that has a better sample rate, is weighted etc, whatever you want that you think makes it more reliable, that contradicts it. I am genuinely curious as this is one thing I have not seen from the Brexit side.

[Mario149]

don4l said:

Mario149 said:

Fine, you think the poll was flawed and useless. So be it. Now show me a better source of information that has a better sample rate, is weighted etc, whatever you want that you think makes it more reliable, that contradicts it. I am genuinely curious as this is one thing I have not seen from the Brexit side.

And you are unlikely to see it any time soon.

Economists rarely agree about anything, and when they do, they are invariably wrong.

Lol, that is just hilarious. All the major economic bodies are saying Brexit is a bad idea financially, but it's okay because they're all wrong and BoJo, Nige and Mike definitely know better.

Like I replied to turbobloke the other day, it's like me going to a firm for legal advice, getting told by all the partners to do the same thing for my own good, then doing the exact opposite because they've all lost a case at some point or another so they must all be wrong on this and my mate down the pub thinks he knows better too.