South of the River – slow news and statistics


Compass-SouthComment logo 2“Nothing happened in South Leeds today”. That was almost an article on this blog on Tuesday. We managed to fill the void with a riveting report about an email account getting hacked.

Have you noticed how there is always just enough news to fill a newspaper or a news broadcast? The Six O’clock News is always 30 minutes long, they never cut it short because there isn’t much to talk about. The South Leeds Life version of this is the three stories that drop into your email in box every weekday morning between nine and ten (if you are a subscriber).

Slow news seems to be a regular phenomenon at the beginning of January. It’s as if everyone has a collective hangover and no one really wants to get started again as they slouch back to work like Shakespeare’s schoolboy “with his satchel … creeping like a snail”.

I first became aware of it in 2000 when I heard an academic I knew being interviewed on The World At One. He explained later that his rather dry report on northern council estates happened to be published on 2 January. Journalists had scoured it like hungry jackals and found a short paragraph that suggested Councils should, in a few extreme cases, consider demolishing unpopular and badly designed estates. As a result, it became the top story and the BBC called in ministers and opposition politicians to pontificate on whether all council estates in the north of England should be demolished.

The Yorkshire Evening Post published a classic slow news story last Monday: “Revealed: Leeds’s (sic) top 10 crime hotspots”.

There’s nothing wrong with a background piece, nothing wrong with digging in the statistics to find an interesting story. You can do some very good journalism on days when you’re not drowning in press releases, speeches and the effects of the weather. In my view the wrong question had been asked and we ended up with a bit of non-story.

The top 10 crime hotspots were the locations where most crimes were committed and recorded in the Police’s figures. They were the places where most people gather – supermarkets, a hospital and a rock festival. Now there is an underlying level of criminality in the population so the more people you have in one place the more crime you will have. This is why I say it is a non-story.

It’s almost very interesting, it certainly raised questions for me. What was missing was some context. How did the list compare with the year(s) before? Is supermarket crime increasing? Are people still shoplifting high value items to make money, or are they stealing bread because they’re desperate. Another way of looking at it would be to compare the numbers of people in each place so you could compare the rate of crime, not just the total number.

This is not the most important story of the year, but it points to a bigger problem and reminded me of an article I saw in the very serious science journal Nature back in November. The article was called “Policy: Twenty tips for interpreting scientific claims”. It was aimed at politicians, but mentioned the journalists that report on them as well.

The issue is that politicians have to make decisions about difficult or contentious issues. They are often presented with scientific evidence to inform them – for example climate change or badger culling. But do they understand how scientific information is gathered and the limitations of the information? Do they know which questions to ask of the information?

I’ve discovered in recent years that a lot of people don’t understand the scientific method. The last 200 years have seen massive progress for humankind fuelled by scientific discoveries, everything from the steam engine to Teflon on saucepans, never mind the internet. Many people have translated that into a view that science, that scientists know things for certain. Then when something goes wrong – the thalidomide scandal or Chernobyl – they flip and say all science is rubbish.

In fact the scientific method is a system for testing ideas and improving understanding but it rarely if ever reaches a point where something is absolutely certainly known and understood. You refine your theory by testing it and it’s the results you weren’t expecting that help you make real progress. So in a sense scientists are always proving themselves wrong.

Kepler measured the movements of the planets in our solar system and Newton came up with the theory of gravity to explain that. Einstein overhauled that theory with Relativity in the twentieth century, but the phenomena he described can only be seen in the behaviour of galaxies. This is on a much bigger scale and one that Kepler could not see or measure, because he didn’t have Jodrell Bank or the Hubble space telescope.

The important thing is that Einstein moved our knowledge on, but that didn’t make Newton wrong. Newton can explain our everyday experience and the car you drive and the house you live in are designed on the basis of Newtonian physics.

Re-reading the article in Nature I realise that it’s a bit heavy going, but it is worth persevering with. A lot of it is about statistics, because that is how results are measured and reported – think about testing new medicines. Politicians (and journalists) don’t need a degree in statistics, but they do need to know about sample size, bias, control groups and randomisation. One point I particularly liked is that scientists are human and sometimes under pressure to come up with the “right” result. It’s about applying the scientific method, testing what is presented to you, asking questions, being critical.

Jeremy Morton Aug13It’s too easy to cynically quote Churchill’s phrase “Lies, damn, lies and statistics”. Interpreted properly statistics can tell you some very interesting truths.

I’ll be back next week with more of my views from South of the River. If you’re on Twitter, you can follow me: @BeestonJeremy.