The reading of opinion polls

The press just love opinion polls. And so, at the moment, do Better Together who are claiming that the gap between Yes and No is growing in No’s favour. It might well be and I am no expert in polls so I couldn’t possibly say one way or another. But, because I am not an expert I go and read the views of two who are.

The first on my list is pro-Independence James Kelly in his blog Scot Goes Pop (where he recently ran a fund raising effort to keep his blog going, hoping to raise in £2500 in 11 days, and received that amount, and more besides, inside of 24 hours) and the other is by the University of Strathclyde Professor John Curtice of What Scotland Thinks. Prof Curtice is perhaps Scotland’s leading psephologist (I must admit to having to look up that word) and has been a regular feature on BBC and STV for at least 20 years (if my memory serves me right) and is renowned for being impartial. I don’t know if he is in favour of Yes or No but I know his analysis of the results reflect very little bias that anyone has accused him of. I tend to visit James’ site because I find his style of writing more agreeable and entertaining, not because he delivers more favourable results for the Yes side (the two differ on methodology of analysis, but that is way over my head).

I am sure you have all heard the line ‘statistics, damned lies and statistics’ and are probably wary of polling reports in any case but I suspect very few reading this blog will have much idea of how polls work. I certainly only have the vaguest idea and I am aware that I appear to know more than many.

There is major significance in what questions are asked in which order and the precise wording can have a crucial effect on the results. The differences between conducting polls via telephone, email or face-to-face all impact the results. To be in any shape or form accurate, the polls have to be representative of the voting population and, as recent polls have typically had around a 1000 responses, considerable work has to be done to ensure that those polled represent a cross section of society on several layers; gender, age, socio-economic status, ethnicity, political persuasion (after all, there is no point asking a 1000 SNP members which way the intend voting and then claim ‘Scotland says…’) and all manner of things depending on what you are polling about. Then, once you ask the questions you have to apply all sorts of statistical number crunching to produce a meaningful result. Then, once all of that is done, some newspaper or some political organisation grossly over-simplifies the results and can claim just about anything the feel like from the debris. And it is even worse when one poll is compared against another if the two are conducted using completely different methodology (which has happened to produce BT’s results).

This is where James Kelly and Prof Curtice come in; they examine the methodology of the polls themselves as well as the results they produce. So please, next time you see the results of a poll printed in a newspaper or claimed by any political organisation, visit the websites above and read a bit more about what the polls actually say.

[A taste of how Better Together’s claims above may not be quite what they claim can be found here: and here]


About Hugh Wallace

Soldier, sailor, policeman, engineer, scientist, democrat, socialist, environmentalist, advocate of Scottish Independence
This entry was posted in Uncategorized. Bookmark the permalink.

One Response to The reading of opinion polls

  1. Pingback: Are Scots racist? | Are We Really Better Together?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s