A few thoughts on referendum opinion polling

In recent years I have tried to avoid blogging on scottish politics for reasons too complicated to go into. But after a week of reading silly tweets, press releases, blog posts, and articles I have been stirred from my silence. Here are, in no particular order, some thoughts on referendum polling.

1. The credibility of opinion pollsters for their general business is dependent on having a pretty good level of accuracy in relation to polling for high profile things, like big public votes. The pollsters therefore, in good faith, do their best to get the most accurate reflection of opinion. For either side in the referendum to claim X is in the pocket of the other lot is a bit cheap. Sometimes you don’t get the result you would like to hear because you are not hearing what the public likes.

2. There may be things that are difficult to take into account in the opinion polls for the referendum – particularly people who may vote who are not usually voting (because they’re not usually registered) – but the companies are trying to do their best.

3. if you are trying to derive any grand declaration of voting intent for an area from a regional subsample of around 120 voters you are an idiot. Those regional polls that have been worth looking at are those where a regional paper or TV station with regional coverage has had a proper job done by a pollster in the area. Last week yes and no folk issued press releases claiming [Region x] for [yes/no] based on samples of 100 or so folk (which was not mentioned). The yes ones emerged after the Survation poll at the start of the week, the no ones after the poll at the end of the week. Read the regional subsamples for each and marvel at the fluctuation of support in some areas. Then compare with genuine regional polling with proper samples that has taken place. If you are a political type issuing a press release like that think about your credibility before publishing.

4. If you are issuing a press release based on a self selective poll on a newspaper or magazine website you are an idiot. Votes on newspaper websites can be hijacked. This will not come as news to fans of television shows or pop groups or personalities. Look at the old best TV show of all time polls, look at what won, and wonder why work by Kneale, Potter, Bleasdale, Rosenthal, Davies &c didn’t appear at the top of the web based votes. I really enjoy doctor Who. I have a DVD and audio and book collection that testifies to it. Is it a better show than I, Claudius? aye, right.

5. If a web based poll purports to be from a site in a particular sector make sure before issuing the press release proclaiming great news for you that those who can vote in the poll actually come from that sector. If you issue a press release proclaiming sector X is for [yes/no] and it can be shown that a legion of one direction fans scrawling through web links can click an answer and comment then your credibility is hit.

6. A vote through social media is a waste of time. Either hijacking takes place or if the link to the poll isn’t shared by people on other social media with an exhortation to vote for this then when you hold up a mirror to a sample you have some control of selecting don’t be staggered if it reflects what you think. Don’t issue press releases or public declarations that your self selecting poll is clearly more accurate than one by opinion pollsters or people will question your motivation or sanity. I have over the years asked various polls on social media often about the pronunciation of scone. This is as informative as a poll on social media gets. A percentage is in favour of scone, another percentage in favour of scone. Beyond that I wouldn’t draw any conclusions.

7. Not every poll can be good news for the side you support whether the vote for the side you support goes up or goes down. Some aren’t good news for your side. Admit it. Not every news story or event is good news for the side you support. The other side gets good news too. Hell, both sides have some quite impressive people on view articulating their positions. That’s why some people are finding reaching a decision difficult. If you want to be taken seriously maybe being a bit less Pollyanna-like will help that.

8. A general thought to end with: people on both sides are taking their position for the best of motivations. They are taking their position because they think it’s for the best for the majority of folk. You might disagree with them, but they’re sincere. Maybe if everyone assumed that those on each side was equally well motivated it might make things a bit easier afterwards.

Advertisements

About loveandgarbage

I watch the telly and read when not doing law stuff and plugging my decade and a half old unwatched Edinburgh fringe show.
This entry was posted in Uncategorized. Bookmark the permalink.

One Response to A few thoughts on referendum opinion polling

  1. Pingback: Alex Salmond remains trapped in a currency quagmire with no way out in sight » Spectator Blogs

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s