Good survey that. The big questions for me though are, who are the 1% who still support the regime and how much destruction will have to be done before the penny finally drops?
As someone who deals with survey data every day I must say I am impressed with the thoroughness of this. When doing something like this it is very easy to think the data is telling you something thats its not and false conclusions are a common stumbling block. This seems well thought out with sensible caveats and no unreasonable assumptions. Well done to the trust and hats off to whoever carried this out, especially if you are a volunteer.
Impressive survey - well done. I would query the point about sales likely to be much above 20% as the multiple tickets purchases (I purchase 3) was not covered and will, I would expect, mitigate the online bias to some extent. This should be very worrying for the CEO.
As someone who deals with survey data every day I must say I am impressed with the thoroughness of this. When doing something like this it is very easy to think the data is telling you something thats its not and false conclusions are a common stumbling block. This seems well thought out with sensible caveats and no unreasonable assumptions. Well done to the trust and hats off to whoever carried this out, especially if you are a volunteer.
@Weegie Addick takes the lion's share of credit I think.
It's interesting but, objectively, you have to question the validity to some extent. If I'm reading correctly 1,519 S/T holders responded - that's only 15% of the (estimated) 10,000 S/T holders. Of those 82% say they won't renew. I think most statistician's would say that's too small a sample number to be able to accurately estimate that therefore 82% of all season ticket holders feel the same. I think it would take a bigger number of the 10,000 to have responded to have a higher degree of confidence. It's still impressive that so many feel that way - but you can't really assume that the other 8,481 S/T holders do. A better number counter than me - @cantersaddick ? might have a better feel for the accuracy.
Good survey that. The big questions for me though are, who are the 1% who still support the regime and how much destruction will have to be done before the penny finally drops?
Probably that fat pervert who works in the club shop.
As someone who deals with survey data every day I must say I am impressed with the thoroughness of this. When doing something like this it is very easy to think the data is telling you something thats its not and false conclusions are a common stumbling block. This seems well thought out with sensible caveats and no unreasonable assumptions. Well done to the trust and hats off to whoever carried this out, especially if you are a volunteer.
@Weegie Addick takes the lion's share of credit I think.
It's interesting but, objectively, you have to question the validity to some extent. If I'm reading correctly 1,519 S/T holders responded - that's only 15% of the (estimated) 10,000 S/T holders. Of those 82% say they won't renew. I think most statistician's would say that's too small a sample number to be able to accurately estimate that therefore 82% of all season ticket holders feel the same. I think it would take a bigger number of the 10,000 to have responded to have a higher degree of confidence. It's still impressive that so many feel that way - but you can't really assume that the other 8,481 S/T holders do. A better number counter than me - @cantersaddick ? might have a better feel for the accuracy.
Really? I think you'll find that most surveys have significantly smaller sample sizes. By my reckoning you'd want about 400 out of 10,000 to start taking the results seriously. Getting more than three times that means that there's very little margin for error here.
It's interesting but, objectively, you have to question the validity to some extent. If I'm reading correctly 1,519 S/T holders responded - that's only 15% of the (estimated) 10,000 S/T holders. Of those 82% say they won't renew. I think most statistician's would say that's too small a sample number to be able to accurately estimate that therefore 82% of all season ticket holders feel the same. I think it would take a bigger number of the 10,000 to have responded to have a higher degree of confidence. It's still impressive that so many feel that way - but you can't really assume that the other 8,481 S/T holders do. A better number counter than me - @cantersaddick ? might have a better feel for the accuracy.
Really? I think you'll find that most surveys have significantly smaller sample sizes. By my reckoning you'd want about 400 out of 10,000 to start taking the results seriously. Getting more than three times that means that there's very little margin for error here.
agree with Stig. Sometimes you see ad's on TV and in the small print along the bottom it will say '89 out of 120 women surveyed agreed'. Think there are a few more than 120 women in the UK.
It's interesting but, objectively, you have to question the validity to some extent. If I'm reading correctly 1,519 S/T holders responded - that's only 15% of the (estimated) 10,000 S/T holders. Of those 82% say they won't renew. I think most statistician's would say that's too small a sample number to be able to accurately estimate that therefore 82% of all season ticket holders feel the same. I think it would take a bigger number of the 10,000 to have responded to have a higher degree of confidence. It's still impressive that so many feel that way - but you can't really assume that the other 8,481 S/T holders do. A better number counter than me - @cantersaddick ? might have a better feel for the accuracy.
Personally don't think that's too bad a sample at all. In my line of work (can't say what) we use a survey of 22000 households and gross that up to the entire UK population. Official statistics and policy decisions are based on this. Where the sample gets sketchy is when we are looking at a particular group of people all in a certain set of circumstances. We then try and inform decisions/ produce analysis on a handful of data points.
From what I can see here (the final product- if given raw data and time I could delve deeper although I am sure there is no need) and within the caveats given I see no reason to question the conclusions.
I have previously offered to sense check/ be a second pair of eyes on the production of these stats. It seems they have this in good hands.
This sort of thing to me perfectly illustrates the strengths and purpose of the Trust, and it's position in the spectrum of protest activities.
I've been critical in the past but this survey alone justifies the membership fee. Excellent work to all (especially @Weegie Addick by the sound of it!)
Absolutely agree with comments on sample size being more than sufficient. It's relatively straightforward to calculate but based on the calculator on Survey Monkey a total population of 10,000 with a 99% confidence level and a 5% margin for error would need 625 responses.
Size of the sample is absolutely robust - the potential for exaggeration will lie in the enthusiasm of responders as opposed to non-responders, since CAST can't easily adjust for that. But it's valid in itself and much better than no survey evidence.
Absolutely agree with comments on sample size being more than sufficient. It's relatively straightforward to calculate but based on the calculator on Survey Monkey a total population of 10,000 with a 99% confidence level and a 5% margin for error would need 625 responses.
Exactly. And a 1,900 sample from a 10,000 population gives 95% confidence in a 2% error margin.
Absolutely agree with comments on sample size being more than sufficient. It's relatively straightforward to calculate but based on the calculator on Survey Monkey a total population of 10,000 with a 99% confidence level and a 5% margin for error would need 625 responses.
Exactly. And a 1,900 sample from a 10,000 population gives 95% confidence in a 2% error margin.
Agree. Did we actually reach 10000 St holders? There was a lot of build up to it with a "big player pledge planned" but then nothing happened. I'm pretty sure if we reached the milestone theclub would have had a bit of a fanfare. But there was none. I guess just another thing we couldn't quite get over the line.
No I think his mind set is stuff them I will carry on, there is just no sense in any of his actions. Has a great club on its knees, a long battle ahead one we must win.
It's interesting but, objectively, you have to question the validity to some extent. If I'm reading correctly 1,519 S/T holders responded - that's only 15% of the (estimated) 10,000 S/T holders. Of those 82% say they won't renew. I think most statistician's would say that's too small a sample number to be able to accurately estimate that therefore 82% of all season ticket holders feel the same. I think it would take a bigger number of the 10,000 to have responded to have a higher degree of confidence. It's still impressive that so many feel that way - but you can't really assume that the other 8,481 S/T holders do. A better number counter than me - @cantersaddick ? might have a better feel for the accuracy.
The size of the survey is way more than required for it to be statistically robust. The potential for error, as stated in the report, creeps in due to potential bias in the make-up of the respondents and whether or not it is a representative set. It is a problem with a passive as opposed to targeted survey set, putting it there for people to complete leads to the potential that those completing it are of a similar mindset which introduces a bias into the results.
I can say this with confidence having spent a good deal of time working with a stats professor from Cambridge University structuring music usage surveys and statistical analogies previously and we played around with sample sizes a lot and found that if it was structured correctly then you could reduce the survey size to fractions of a percentage of the total population without any material change in result.
As an example, the BARB reporting panel is made up of just over 5k homes, this represents 26m homes in the UK and £billions in ad revenue are spent using this data each year.
Comments
From what I can see here (the final product- if given raw data and time I could delve deeper although I am sure there is no need) and within the caveats given I see no reason to question the conclusions.
I have previously offered to sense check/ be a second pair of eyes on the production of these stats. It seems they have this in good hands.
I've been critical in the past but this survey alone justifies the membership fee. Excellent work to all (especially @Weegie Addick by the sound of it!)
in any of his actions.
Has a great club on its knees, a long battle ahead one we must win.
I can say this with confidence having spent a good deal of time working with a stats professor from Cambridge University structuring music usage surveys and statistical analogies previously and we played around with sample sizes a lot and found that if it was structured correctly then you could reduce the survey size to fractions of a percentage of the total population without any material change in result.
As an example, the BARB reporting panel is made up of just over 5k homes, this represents 26m homes in the UK and £billions in ad revenue are spent using this data each year.