DEVELOPMENT PROPOSAL
It’s interesting that we are discussing the reliability of polling organisations, especially those with close ties to political parties and media organisations. From the earliest days of this Coffee House Wall I have been talking about beginning a project that would poll a large representative sample of the electorate, asking questions that are not normally considered by the polls we see in the papers, and which would try to be rigorously honest about the results and which would be entirely independent.
The intent would be to take into account the usual criteria of age, social class, location and previous voting record, but this project would also consider ethnic background, identified religion and immigration status. It seems necessary to distinguish between the opinions of those recently migrating to the UK, or being members of immigrant families, and those who are members of the indigenous population. Perhaps they are not different, perhaps they are. We don’t get to see at present. It would also be intended that the site provided for a constant live record of registered members voting intentions, and that the registered base would be large enough for this result to be statistically valid.
This would run as a seperate sister website with links between the two of course but would not be adding complexity to this site. I won’t mention the name I have in mind as someone will steal it.
I am mentioning this here, not to lean on those who have already been supportive, but because there are thousands of people who read this site and may consider it an interesting idea that is worth supporting. To begin with a detailed specification is required. Some aspects of which will be presented here for open discussion.
Funding is needed to develop this project. Reasonably significant funding. If you think this is an idea worth supporting and which could provide a corrective to some of the other polls routinely produced as evidence of the electorates views, then do please make a donation now using the various Paypal links on the site. If there is sufficient support then it will at least allow a specification for the site and project to be produced.
I wish you well with this project – it is undoubtedly needed.
However, I fear you will be prevented in your aims. The beliefs and intentions of too many of those that cannot tick British Caucasian (were such a tick box ever to be permitted!) are greatly at odds with the traditional beliefs of the nation and its residents of long standing. The establishment will not look favorably upon any attempt to expose these aims and beliefs to a greater scrutiny.
Just look how long it has taken to permit any discussion of immigration without outraged howls of racism across the media. We still cannot openly and freely discuss the steady Islamisation of the western world without fear of physical violence, let alone verbal attack.
Thanks for the encouragement EC. It is appreciated.
But I can’t just do nothing. I am not yet 50 (June 4th!). I’m willing to take the risk while I can. To be honest I think that allowing the voices of different communities to be heard, even if what is said is unpalatable, cannot be considered racist or xenophobic. It is the silencing of all of these voices in one headline statement by the usual pollsters which is racist. It denies all of us, in our different backgrounds, a voice.
Peter
one of the failings of the current polsters is failiing to track different attitudes in different areas. The rise of UKIP has been patch across England, and non existant in Scotland and Wales. But we see the headline figures which fail to take account of these geographical differencies.
And their sample size is too small, just 10-15 from each county.
(Even a county will have differencies. Warwickshire has the old mining areas round Nuneaton and the rural areas round Stratford. Northamptonshire has huge rural areas, and Corby in the middle.)
But unless these differencies are recorded, then polls will become increasingly inaccurate.
Just my 2p worth (Should be 2d but hey ho)
Hi Alexsandr, that is why I want to start this project. These differences should be recorded, and would be recorded by the system I hope to create, and woul dbe reported.
The population of England and Wales is 56,000,000. The adult population is about 44,000,000.
To be 99% confident of getting polling results that are +/- 2% accurate I would need a sample size of about 4000 people.
For a whole county like Warwickshire, adult population of 420,000, I would need a sample of about the same amount of people, because the population is smaller, so a larger sample is needed to be sure of accuracy. Bringing it down to 95% confident of results that are +/- 3% needs 1000 people.
The latest poll at YouGov for the Sun had 1914 participants. And in Scotland only 147. So the maths suggests that there is a 95% chance that the Scotland results are representative with a +/- 8% accuracy.
The only poll that I am familiar with is YouGov, I have been completing their surveys for years and I am convinced that their surveys of voting intent or political opinion are corrupt.
My reason for thinking this is that when I first started I indicated how I was of ‘the right’ since then all of the surveys that I am asked to complete are with regard to consumerism never politics.
I think that they deliberately weed out rightwing opinions.
Yes, I also think there is a problem with YouGov. Which is why the system I am hoping to be able to develop would not weed out people based on their politics etc.
Peter, you sound eager and young and idealistic. To get the thing up and running would cost more than you can get from any donations by those visiting your site. What you may like to do is conduct abit of research on the scarcity, poor quality, reliability of existing polling, suggest ways of getting it done better then approach in person a rich man or a woman, try convince such person to fund it.
You may also like to find someone who’s been involved in some of the private polling done by the major parties to get an idea of the cost, and what’s involved. You may find just setting up and maintaining the sample will need more than your idea of segmentation by age, voting intentions and stuff. The whole field has moved on, is quite sophisticated today.
Hi Baron, Thanks for the comments. I am not young or idealistic. I am old and want to get something done before it is too late. I am pretty aware of polling and statistics. I’d not say that most political polling is very sophisticated. It is used by our politicians and seems designed to produce a given result.
I agree about it needing more than donations. I do need to find sources of funding. But I look at a lot of polls and I think that it is not beyond anyone who has some statistics background.
Peter,
I think that Baron is correct about the startup costs involved.
Have you heard about the recent advent of “Crowd funding” ?
There will be literally 1000’s of Stats Grads on the job market in a couple of weeks time.
Got a good name for the site, though. How about “Honest Polling”?
I am not sure the start up costs are so high. They are higher than I can bear myself. But web development is what I do for a living when I am not trying to get more visitors here. I shall programme the entire project myself. If you look at the poll results of YouGov it is not more complex than the Stats A Level I did years ago. It is the truth which matters.