Forums Forums National News UK wants to stay in the EU.

At least according to the Independent's latest poll, and by a considerable margin of 10%

http://www.independent.co.uk/news/u...emain-ten-points-leave-bmg-a8114406.html

Whilst polls at not that accurate, they are not likely to be out by that much, and it is not the only poll to have shown this trend recently although it has shown the biggest lead for remaining.

I think it pretty clear that the British people have spoken, and said "We've changed our minds. We now want to stay in!" Now the implications of leaving are beginning to sink in, and the slow-motion car-crash that is the government's hapless 'negotiations' haven't helped.

I wouldn't oppose another referendum, but I don't see the point. The poll results are clear enough and parliament seems to be realising that it doesn't HAVE to go along with brexit if the terms are not to its liking. It is unlikely that they WILL be any good, and they will decide that brexit IS the worst possible bad deal and stop it.

I'd be quite happy if the honourable members did the job of kicking it into touch without us having to lift a finger.

Everyone can clutch at straw polls.
From the researchers own website



http://www.bmgresearch.co.uk/indepe...t-height-of-brexit-negotiation-tensions/

Well, this seems to be a master class at straw clutching.

First, this isn't just one poll. Although it shows the biggest majority for remaining, there have been previous pools showing smaller majorities for staying in.

Second. if there is a movement towards remaining. the most recent post WOULD be the biggest, wouldn't it. That movement is clearly taking place, like it or not, and it appears to be accelerating.

Third, what the poll shows is that committed brexiters and remainers have not changed their views. What has happened is that the don't knows or don't cares during the referendum has had time to reflect and have realised what a crock brexit is and want no part of it.

Finally, if you want another referendum, that's fine by me. But perhaps you would now prefer to ignore the voice of the people because you are afraid the result will not be to your liking?

Polls are a joke these days, ask Cameron and May!

Leave or Remain single choice

Leave the EU (74%, 20 Votes)

Remain in the EU (26%, 7 Votes)

Total Votes: 27 Voting on this poll ends: 20th Dec 2017 10:38am

I don't think they are any more of a joke these days than they have been in the past. Certainly far less of a joke than the Referendum campaign was. But if you think that, is then lets have another referendum. Expensive and time consuming and - as the last one showed - involving lies and propaganda, but if that's what you want...

In fact the polls were pretty accurate in May's case. They clearly reflected the hemorrhaging of support from her as her appallingly bad campaign proceeded and towards the end clearly indicated a hung parliament.

Granny, you are not competent to run a poll. I will ignore it as it is unlikely to be representative.

You do realise the poll comes from the forum software and not Granny?

Of course I do. But that doesn't make it any more competently run. You need to weight the responses according to location, age, employment status etc. etc. to get an accurate result. No way will Granny or the forum software do any of that.

Granny was the one who posted this 'poll' though, not the forum software.

Exactly the sort of factors used to fiddle figures to get the "required" outcome.

Could you explain how those factors could be used to get a more "accurate" result? If any adjustments are used, how do you obtain accurate adjustments?

Well, I would have thought you would be able to work work that out for yourself. Its pretty obvious. Here's an example.

Suppose you have 5,000 people doing a poll and 1,000 of them happen to be retired. You know from other data that only 5% of the population are retired, so using the raw data will give a result heavily skewed to the views of the retired. so you discard the data from a randomly chosen number of these retirees to bring their percentage down from 20% TO 5%.

You can call this fiddling, but it is routinely done, and - not too surprisingly it usually yields a result which matches the actual outcome. Not always, of course. Views can change very rapidly and if the poll is done two days before the election it will not be able to take into account any outcome affecting events which occur in those two days.

I really cannot think why you imagine companies like Opinium and Yougov - whose reputation as honest pollsters is what they are actually selling - would fiddle anything. What do you think they are likely to gain from it?

For those interested. This site gives a free sample size calculator. Thus if you want to be 99% sure a result is accurate to within 3% and you have a population size of 65,000,000 then you'll need to poll 1,849 people, a tiny percentage.

However, if your population size is 1,000, then you'll need to poll 649 people. Well over half. Magic? No simple mathematics.

In my experience people that resort to insults either aren't sure or know they might be wrong.

How about just spreading your knowledge without the insults, more people might believe you!

You know from your above two posts where the fiddles lie, you can uses a much smaller population to "correct" the results of a bigger population. It is also very subjective as to what "corrections" are carried out and what is ignored. Under your regime you would have to do correlation tests on everything to make sure what is independent and what isn't otherwise some "corrections" would be getting counted more than once - or would they, there have been many discussions about that.

Then you have decide what confidence interval you should use, another subjective and/or manipulative choice. In my time and fields the UK preferred 97.5%, the Americans preferred 95% but 90% was used as well, who was right and who was wrong? Of course they are just indications, just as a small poll of 20 people on WikiWirral would be.

So rather than woffle on about your latest reasons for a poll on here not being a valid , in an effort to cover your indiscriminate original reasons i.e. the above in red, you might be interested to know that there are many on Wiki Wirral who are highly qualified, have their own businesses, a wide age range, and people from all walks of life. The truth is every category you listed is filled by Wikiwirral members. Maybe in your little world of arrogance , and 'no one is better than me' attitude , you can rest assured that you are greatly misjudging.

I cannot for the life of me see any motive for pollsters to fiddle their polls. It would quickly destroy their businesses. Nor can I see how it could be done without it showing up. The polling companies publish the raw data as well as the result of these weightings and they can be checked to see that they comply with industry standards. Excluding the odd typo, they invariably do. Go look at pollingreport.co.uk for instance. They usually provide links to all the data. If you want to, you can examine the corrections and see exactly what has been done with the raw data. I have yet to see any evidence at all of 'fiddling'.

DD talks about 'adjusting the population size' but a far as I can see this is more nonsense The population size is fixed by the British People - at around 65,000,000 and doesn't even appear in the calculations and adjustments done by the polling companies except to give an indication of how many people they have to interview in order to get the accuracy they require.

My point is that a poll is a lot more complicated than you or granny seem to realise. It isn't just a matter of counting the pros, counting the cons, and publishing the result. (Which is what the 'poll' here does) Therefore, granny's poll is likely to be pretty well worthless. It is nothing whatever to do with how clever or gifted the people here are. The problem is that they (including me) are a group who like discussing things online which introduces a bias. I suspect we are mostly of retirement age. which introduces more, and if we were a particularly well qualified and clever group as Granny claims, then that in itself would introduce further bias. Also, we all live on the Wirral -
a further source of bias. Certainly, the probability of the people who respond to granny's 'poll' being a representative cross-section of the UK electorate approaches zero.

The main source of bias lies in the questions themselves. It can be tricky to ask a question which doesn't include bias. "Would you prefer to be in the EU rather than out of it?" may well a return very different result from " Would you prefer to be out of the EU or in it?" for instance. If you look at the published information it will invariably show the actual question asked, allowing the reader to decide for himself if there is unconscious bias there.

Finally, it's worth remembering that attempting to trash a properly conducted poll says more about your attitude to the result than I think you would want to reveal. But if yo want to do it, then you will have to go into a lot more detail than vague unsubstantiated accusations. The devil, as always, is in the detail.