Monday, 1 October 2012

Statistics... not as simple as they look


Here's an interesting fact: statistics are complex things.

Sure, they always look like they're making clear points. Inflation is up by this much; unemployment is up by this much; this politician is this far 'ahead' of this other one. But it's not so simple, as I've argued in print many, many times. You have to be able to interpret the data, to explore the context, to understand the history.

So when The Daily Telegraph reported the other day that there were 'only 100 mature cod left in the North Sea', it was a misleading picture. It misled your present correspondent, though not quite as badly as many in the mainstream press. What they really meant to say was 'one estimate of 13-year-old cod in the North Sea says that there might only be 100 left of that age'. Except, of course, that this isn't a very good measure of what it means to be an 'adult' fish, and there are many millions of four-year-old and above cod left in that part of the world. So this blog is sorry if it gave a false impression. We should have stuck to our previous, more optimistic and published conclusions about recent legislation - which does at least offer some sort of way out via reform of the Common Fisheries Policy and the establishment of Marine Conservation Zones. At least we're admitting it here. Many outlets don't.

All the same, Fishing Ministers everywhere have been quick to pounce on the snafu to calm public nervousness. Don't worry, they imply; there's years of fish left. We don't need to shrink the fishing fleet by much. You can keep on munching on your fish and chips. Well, it's not true. Measuring fish by the millions doesn't account for the effect that fishing them out as quickly as they mature. Numbers are still far below replacement levels, and we still need to fish less, and more responsibly. Some economists even think that a temporary freeze on all fishing would be the best way to go - and prove more profitable in the end, as stocks are replenished. So the message is still: don't eat lots of white chunky fish without the dark blue Marine Stewardship Council label on it. Otherwise, you're helping to destroy the world.

In a way, this brings us back to one of the most interesting controversies raging over the present US Presidential election. That is: are the polls 'skewed'? Some Republicans think they are 'over-sampling' Democratic-leaning sympathisers with President Barack Obama. Pollsters answer that this is nonsense: by and large, they just report the party balance of the sample they get. If Democrats become more enthusiastic (and the evidence is that they have since their party's convention in North Carolina), then their presence among respondents becomes more noticeable. Who's right? Well, given how accurate and non-partisan polls have been since the 1970s, the pollsters are probably right. But you have to delve into the past, and into the statistics, to find that out. You have to know the history.

So, yes, there aren't only '100 adult cod left in the North Sea'. But there sure as hell aren't enough, either.

A big thank you to readers: I just wanted to say that last month was again Public Policy and the Past's best month ever, with traffic up yet again by ten per cent on the month before. So to the growing band of loyal surfers: thanks, let others know about PPP, and here's to many more readers!

No comments:

Post a Comment