Open question to Alun Evans and Les Reid at The Coventry Telegraph. (2 Viewers)

[SkyBlueCharlie]

On 16th October you published the following article http://www.coventrytelegraph.net/news/news-opinion/could-ricoh-arena-sold-coventry-6191587 that was accompanied by a Yes/No poll headed: Should the council sell the Ricoh Arena to Sisu so Sky Blues can play in Coventry?

When do you expect to report and comment upon the results of this poll?

When you do so, will you also report upon the total number of votes cast?

 

---

[asb]

The answers that will be provided by the poll will be meaningless. It will, however, not stop both sides crowing about some minor victory by spinning the result, even if Mr Reid and the CET take the results no further and make no reference to them.

The whole methodology of the poll, and the inherent flaws of the delivery, leaves the results being as scientific as flipping a coin.

The poll used is termed SLOP, that is a Self-Selected Listener Opinion Poll. It is a favourite of the media and they like to suggest it offers valid results, but sadly it does not.

Some basic flaws are:-

Possible to vote more than once. I tested this by voting once for Yes and once for No. I do expect some on here to suggest that it is only the No voters that have voted more than once, and others to suggest that it is the Yes voters. I can only state that both options are open to multiple spam voting.

The only people that are going to vote are those that feel strongly about the issue. Strength of feeling does not equal an impregnable position of truth, nor does it offer a balanced view.

The question is so narrow that a vote for either side could be based on numerous conditions. Voting for an idea based on an assumption is flawed unless the position you are voting for is clearly defined.

I really have had enough of such poorly executed polls being used as irrefutable evidence of strength of support for an idea.

The results of the poll are invalid and can never show support for either side.

*I feel the need to add that this is NOT an attack on the journalistic standards of the CET or those of Mr Reid.

 

[SkyBlueCharlie]

asb,

Thank you for such a detailed and illuminating response. I must admit that I was aware of the limitations of this type of poll hence my, to a degree, mischievous, question. Where I disagree with you is that is that I think this does reflect poorly on the CT. As a responsible organisation they should be aware of the limitations you outlined so why use this type of poll?

The only answer that I can think is that they were looking for a specific response that they could use to justify the article that it accompanied and would be depending on general ignorance concerning the validity of the results.

It would be interesting to hear a response from the CT if only in the interests of the transparency they regularly demand from others.

 

[ohitsaidwalker king power]

The answers that will be provided by the poll will be meaningless. It will, however, not stop both sides crowing about some minor victory by spinning the result, even if Mr Reid and the CET take the results no further and make no reference to them.

The whole methodology of the poll, and the inherent flaws of the delivery, leaves the results being as scientific as flipping a coin.

The poll used is termed SLOP, that is a Self-Selected Listener Opinion Poll. It is a favourite of the media and they like to suggest it offers valid results, but sadly it does not.

Some basic flaws are:-

Possible to vote more than once. I tested this by voting once for Yes and once for No. I do expect some on here to suggest that it is only the No voters that have voted more than once, and others to suggest that it is the Yes voters. I can only state that both options are open to multiple spam voting.

The only people that are going to vote are those that feel strongly about the issue. Strength of feeling does not equal an impregnable position of truth, nor does it offer a balanced view.

The question is so narrow that a vote for either side could be based on numerous conditions. Voting for an idea based on an assumption is flawed unless the position you are voting for is clearly defined.

I really have had enough of such poorly executed polls being used as irrefutable evidence of strength of support for an idea.

The results of the poll are invalid and can never show support for either side.

*I feel the need to add that this is NOT an attack on the journalistic standards of the CET or those of Mr Reid.

asb,

Thank you for such a detailed and illuminating response. I must admit that I was aware of the limitations of this type of poll hence my, to a degree, mischievous, question. Where I disagree with you is that is that I think this does reflect poorly on the CT. As a responsible organisation they should be aware of the limitations you outlined so why use this type of poll?

The only answer that I can think is that they were looking for a specific response that they could use to justify the article that it accompanied and would be depending on general ignorance concerning the validity of the results.

It would be interesting to hear a response from the CT if only in the interests of the transparency they regularly demand from others.

I bet you both voted though....

 

[shmmeee]

The answers that will be provided by the poll will be meaningless. It will, however, not stop both sides crowing about some minor victory by spinning the result, even if Mr Reid and the CET take the results no further and make no reference to them.

The whole methodology of the poll, and the inherent flaws of the delivery, leaves the results being as scientific as flipping a coin.

The poll used is termed SLOP, that is a Self-Selected Listener Opinion Poll. It is a favourite of the media and they like to suggest it offers valid results, but sadly it does not.

Some basic flaws are:-

Possible to vote more than once. I tested this by voting once for Yes and once for No. I do expect some on here to suggest that it is only the No voters that have voted more than once, and others to suggest that it is the Yes voters. I can only state that both options are open to multiple spam voting.

The only people that are going to vote are those that feel strongly about the issue. Strength of feeling does not equal an impregnable position of truth, nor does it offer a balanced view.

The question is so narrow that a vote for either side could be based on numerous conditions. Voting for an idea based on an assumption is flawed unless the position you are voting for is clearly defined.

I really have had enough of such poorly executed polls being used as irrefutable evidence of strength of support for an idea.

The results of the poll are invalid and can never show support for either side.

*I feel the need to add that this is NOT an attack on the journalistic standards of the CET or those of Mr Reid.

I tried voting twice but there's no evidence it works. If you inspect the voting element it only reveals a pre-loaded number when you click, your vote isn't represented in that tally. I tried clearing cookies to no avail as well. Both of these are fairly standard security measures on any polling system. Admittedly IP changes would work, but that's a bit more hassle and harder for the non-tech savvy to do.

 

[asb]

Where I disagree with you is that is that I think this does reflect poorly on the CT. As a responsible organisation they should be aware of the limitations you outlined so why use this type of poll?

The only answer that I can think is that they were looking for a specific response that they could use to justify the article that it accompanied and would be depending on general ignorance concerning the validity of the results.

I can not even guess reasons for the poll, and any possible usage of the data gathered by the CET, past the basic and most likely, a point of interest to promote discussion, not something to be taken seriously.

I can only offer the view that such polls are common place within the media. The media are not generally academics, and are not held to the same peer reviewed validation. They see such polls as cheap and quick solutions, and have no requirement to ensure validity of the data. The news is simply an editorial view of the events they describe and is rarely seen as irrefutable proof.

I have seen these polls being used for nothing but entertainment, a talking point. I have also seen them being held up as some form of rigorous scientific study offering 100% truth, by academics as well as newspapers.

I can only at this stage comment on the methodology so I await any usage of the data before I can judge that.

It would be interesting to hear the answers you ask of the above

 

[shmmeee]

I can not even guess reasons for the poll, and any possible usage of the data gathered by the CET, past the basic and most likely, a point of interest to promote discussion, not something to be taken seriously.

I can only offer the view that such polls are common place within the media. The media are not generally academics, and are not held to the same peer reviewed validation. They see such polls as cheap and quick solutions, and have no requirement to ensure validity of the data. The news is simply an editorial view of the events they describe and is rarely seen as irrefutable proof.

I have seen these polls being used for nothing but entertainment, a talking point. I have also seen them being held up as some form of rigorous scientific study offering 100% truth, by academics as well as newspapers.

I can only at this stage comment on the methodology so I await any usage of the data before I can judge that.

It would be interesting to hear the answers you ask of the above

While this is nice and all, it's not really useful to claim "this isn't perfect, therefore it's not valid". As I said, there's no evidence you can multi-vote, the article was the most read on NewsNow and CET which suggests a fair number of those interested in CCFC read and (due to the low barrier of entry) probably voted, and let's be honest the question was fairly clear. Those who want the club back regardless of owners would vote yes and those who do not want Sisu to get the Ricoh on the cheap would vote no.

So while it's not perfect, it's certainly a valid data point for discussion.

 

[asb]

I tried voting twice but there's no evidence it works. If you inspect the voting element it only reveals a pre-loaded number when you click, your vote isn't represented in that tally. I tried clearing cookies to no avail as well. Both of these are fairly standard security measures on any polling system. Admittedly IP changes would work, but that's a bit more hassle and harder for the non-tech savvy to do.

You are correct. There is no evidence that such results are recorded each time you click. I base my opinion on the knowledge of the systems they are using, although not directly how they use those systems. Also on voting early in the cycle when votes changed the percentage on clicking. All of which is no proof of being able to vote more than once, but can make the whole process questionable when combined with the other issues.

I also noticed that the comment boards for the CET do allow and record, keep recorded, votes up and down, even with no IP changes.

So it is plausible to suggest that their mastery of the user identities on their site is not the best.

The fact that you are aware of the ability to change IP to vote more than once shows that there is a way even if they were monitoring IP addresses.

For the vote to be valid then they need to validate the identity of the each and every voter. That is more work than required for what they are most likely after.

As long as it is not taken seriously and the first person to quote the results as proof is firmly challenged then it's not really an issue.

 

[shmmeee]

You are correct. There is no evidence that such results are recorded each time you click. I base my opinion on the knowledge of the systems they are using, although not directly how they use those systems. Also on voting early in the cycle when votes changed the percentage on clicking. All of which is no proof of being able to vote more than once, but can make the whole process questionable when combined with the other issues.

I also noticed that the comment boards for the CET do allow and record, keep recorded, votes up and down, even with no IP changes.

So it is plausible to suggest that their mastery of the user identities on their site is not the best.

The fact that you are aware of the ability to change IP to vote more than once shows that there is a way even if they were monitoring IP addresses.

For the vote to be valid then they need to validate the identity of the each and every voter. That is more work than required for what they are most likely after.

As long as it is not taken seriously and the first person to quote the results as proof is firmly challenged then it's not really an issue.

Validity is not a binary state. Obviously it would be even more valid if everyone submitted to an ID check, but it doesn't make this methodology useless.

 

[asb]

and let's be honest the question was fairly clear. Those who want the club back regardless of owners would vote yes and those who do not want Sisu to get the Ricoh on the cheap would vote no.

So while it's not perfect, it's certainly a valid data point for discussion.

There is an assumption in that view.

You assume that those that voted No think that SISU will get it on the cheap.

You assume that those that voted Yes want the club back regardless of owners

I assume for a counter argument that the point that people add caveats to their votes means the question is unclear.

When you get people stating "Yes..but only with a fair price" then the original question has been altered to include an extra premise. Some that voted would have been swayed by their view on the most likely asking price. Some would think yes as long as a fair price is offered, some will vote no as a fair price will never be offered.

The lack of clarity of question forces people to choose an assumption which alters the possible selection of yes or no. It is possible to clear out the inherent flaw in that by removing the assumptions.

All because lots have voted does not make it a valid vote. All because the CET has published it does not make it irrefutable.

And for the record I have no idea what the current state of the voting is, so I could be challenging a piece that I could use to support my view on CCFC.

 

[shmmeee]

There is an assumption in that view.

You assume that those that voted No think that SISU will get it on the cheap.

You assume that those that voted Yes want the club back regardless of owners

I assume for a counter argument that the point that people add caveats to their votes means the question is unclear.

When you get people stating "Yes..but only with a fair price" then the original question has been altered to include an extra premise. Some that voted would have been swayed by their view on the most likely asking price. Some would think yes as long as a fair price is offered, some will vote no as a fair price will never be offered.

The lack of clarity of question forces people to choose an assumption which alters the possible selection of yes or no. It is possible to clear out the inherent flaw in that by removing the assumptions.

All because lots have voted does not make it a valid vote. All because the CET has published it does not make it irrefutable.

And for the record I have no idea what the current state of the voting is, so I could be challenging a piece that I could use to support my view on CCFC.

Oh I agree with your concerns, I was merely pointing out that your assumptions can be countered with equal and opposite assumptions. As I say it means you take it with a pinch of salt (like ALL scientific enquiry) but it doesn't mean you throw the baby out with the bathwater.

 

[asb]

Validity is not a binary state. Obviously it would be even more valid if everyone submitted to an ID check, but it doesn't make this methodology useless.

For the purpose it is currently being used it fits,as a talking piece. If that purpose changes to become a suggestion of a demographic vote then the methodology is flawed.

Shall we hold elections by similar methods?

Such methods are not generally used in serious research because of the validity of the data.

The reason they are used in the media is because they are quick, easy, and they produce a result. That is all you can state about them.

 

[shmmeee]

For the purpose it is currently being used it fits,as a talking piece. If that purpose changes to become a suggestion of a demographic vote then the methodology is flawed.

Shall we hold elections by similar methods?

Such methods are not generally used in serious research because of the validity of the data.

The reason they are used in the media is because they are quick, easy, and they produce a result. That is all you can state about them.

To be fair there's all kinds of problems with the electoral system, I wouldn't hold that up as a bastion of good practice.

Like I said, you can level the argument at all kinds of research, but it doesn't make it invalid, it just means that you should be aware of it's limitations.

 

[asb]

Like I said, you can level the argument at all kinds of research, but it doesn't make it invalid, it just means that you should be aware of it's limitations.

In academia there can be seen the view that "only my research is valid, and those of the people that support my view". I and others can be guilty of such generalisations

The original response to the post was highlighting some of the limitations of the methodology used. There can be a confusion on such limitations and what that means for the weight of evidence.

In simple terms a Mori Poll should not be placed alongside of a website poll as equal weighting. They both produce data but not in the same way, with the same internal validity.

What I see happening (and this is way beyond the issue with SISU, CET, ACL, etc) is people grabbing the data as a sledge hammer of truth, when in reality it's not even a toffee hammer of opinion. Where there needs strong caution of conclusion, there can be weak expression of limitations of result instead.

This all leads to more arguments and more division, and no one can be certain that what they are holding is the truth.

The other issue of CET reporting the results, I don't see the point in a major "breaking" story based on the quality of the poll. I do hope it is not sold as such, as it just adds poor weapons to either side, depending on who wins the vote. As an additional sentence in any of the numerous CET stories it might find a place. It just doesn't deserve its own story.

 

[SkyBlueScottie]

Mischievous question..... Ooh you're a card.....

 

[torchomatic]

I voted YES and NO at home and at work. Plus I did it again on my phone. There's no evidence it doesn't.

I tried voting twice but there's no evidence it works. If you inspect the voting element it only reveals a pre-loaded number when you click, your vote isn't represented in that tally. I tried clearing cookies to no avail as well. Both of these are fairly standard security measures on any polling system. Admittedly IP changes would work, but that's a bit more hassle and harder for the non-tech savvy to do.

 

[Nick]

To be fair you could spam it quite easily if you could be bothered. Would seem like a hassle to fix a nothing poll?

 

[Godiva]

To be fair you could spam it quite easily if you could be bothered. Would seem like a hassle to fix a nothing poll?

The poll itself is rubbish, but the result is bound to be in the paper. It will be quoted over and over. Used in numerous debates and discussions.
Think about it ... the effort is well worth it.

Remember the latest rubbish poll that got quoted twice by Les Reid?

 

[SkyBlueCharlie]

The poll itself is rubbish, but the result is bound to be in the paper. It will be quoted over and over. Used in numerous debates and discussions.
Think about it ... the effort is well worth it.

Remember the latest rubbish poll that got quoted twice by Les Reid?

It's been fairly well detailed by both asb and shmmeee why this type of post is flawed so the question is why use it? The answer is, as you say Godiva, for it to be used in the CT at sometime. My opinion was that the question was slanted in such away to provide the support for the latest petition raised to put pressure on CCC. The problem is that the results do not do that and as such, in my opinion, the results will be quietly shelved, never to see the light of day (or at least print anyway).

The real issue I have is that the CT, which prides itself on the quality and neutrality of its reporting resorts to using such a flawed tool to potentially distort what is a very valid question.

 

[dongonzalos]

I would say people are giving it a bit too much thought.

Is it really surprising that a lot of people are very concerned about trusting SISU with the Ricoh. Also that a lot of people are not keen to be broken by blackmail?