But numbers are not automatically informative. The volume of figures an organisation puts out is not a guarantee of accountability. Openness is only the first step. The figures have to be made to work for their living. Interpretation and comparison are needed to make the stats sing. The World Bank displays its achievement in the year to July 2011:

119,639 Bank documents released to the public.
25 million page views on the Open Data website.
4.5 million page views on the Access to Information external website.
1.02 million visits to the Documents and Reports website.
795,000 documents downloaded.

Are those numbers impressive? What do they tell us? We need some context. Of what total of documents do those released to the public relate: is the proportion large or small? So 795,000 documents were downloaded. But what proportion of visits to the website produced a download? Were those who downloaded a small proportion of total visitors – in other words did the World Bank like other such organisations have a much smaller concentrated audience than the total number of hits on its site might suggest? Anyway, how do these figures stack up when we look at the International Monetary Fund, say – in other words, benchmarking is needed. Brute figures can be dumb.

When global population figures are announced it is usually with anxiety about ‘population explosions’ and concerns that global resources cannot sustain continued rates of growth. You might be forgiven for momentarily thinking that maybe we -  the living  - outnumber the dead.  More probably your instinct would be that taken in their totality, the dead must outnumber the living.  But by how much? and how you would work that figure out?

We won’t make you wait for the answer……the ratio of dead to living is 15:1…..there are 15 people who have once lived  for everyone of us alive on earth today. It is estimated that 107 bn people have ever lived. BBC Magazine’s article ‘Do the dead outnumber the living?‘ gives more insight into the PRB’s approach.

Certainly the balance of opinion has not seemed to be in his favour. Lots of concessions, though.  There has been the occasional concession e.g. that he had made qualifying look easy. Good. These remarks were soon countered by reminders that he was was one of two England managers who oversaw the national team concede 4 goals in a World Cup finals game. Not so good.

However….some references were made to his goal average per game being a healthy 2-1 and, moreover, that he had the best wins-loss ratio of any England manager post-World War II.  (A quick look at the OPTA table below verified that his wins-loss ratio was indeed better even than Sir Alf Ramsey’s).

This all goes to show that whilst measuring performance the statistical way is important, if the stats are telling us one thing, we will be slow to listen if our intuition and senses are taking us in another direction. However, I think it is safe to assume that Fabio Capello’s successor is paying attention to the statistics. Statistically-speaking, Fabio Capello  has outperformed his immediate predecessors and whoever takes on the mantel next has a lot to live up to. 

Carmine Gallo, an American ‘presentation coach’ whose books include an analysis of the Steve Jobs style, offers some useful tips on how to get numbers over in a piece in Forbes.  Use your imagination with stats and you’ll make sales.

For example, instead of talking about 3 per cent market share, a speaker could get the relevant proportion of his audience to stand up – giving an immediate and convincing picture of the numerical relationship.

‘Statistics often don’t mean much if left on their own. In fact, the bigger the number, the more important it is to put into context. Have fun with it. Brainstorm ways to add context around the statistic and bring it to life.’

He talks about a Mexican fruit brand, which was saying that in the previous five years it had planed 30m trees. Make it live for your audience is the advice. That is s the equivalent of 60 Central Parks in New York, which makes for an arresting image.

In an upbeat resume of the figures, the BFI announced the gross value of box office takings was up together with the number of cinema admissions. Defining a ‘British film’ isn’t easy, when production finance comes from various sources, but basing the definition on where the film was principally shot, the BFI put Harry Potter and the Deathly Hallows Part 2 at the top of its list in terms of box office gross for 2011, followed by The King’s Speech and The Inbetweeners Movie, all proudly described as UK movies.

More space, more speed; more speed, more deaths? You can build an intuitive picture, but it isn’t necessarily so.

You can’t – says Michael Blastland in his BBC News Magazine blog - make an open and shut case in favour of congestion as a safety device, but the comparison is suggestive. (Congestion charge payers in the capital are getting a sort of insurance policy?)

But note, these figures are for serious, not all accidents. They don’t make urban traffic any friendlier. It’s not immediately obvious, especially if you are a cyclist in the capital, that heavy goods vehicles are anything but a threat

As Michael notes, maybe the most dangerous roads have good accident records and appear safe because cyclists and pedestrians avoid them. In other words, awareness of the figures – statistical consciousness – also matters, but not necessarily in any straightforward way.

Despite the impressive-sounding technical detail – “women have an average parking coefficient of 13.4, while the average male score was just 12.3” – this survey is really just a bit of fun. It does, however, raise an interesting statistical question. What do we mean when we say one group is better than another?

Let us think about something a little less controversial – and easier to measure – than parking skills. Something like heights. We can say, without offending anybody, that men are taller than women – or, more precisely, that men are taller than women on average. But that tells only part of the story. It is obviously not the case that all men are taller than all women. Less obviously, there will in fact only be a handful of men who are taller than all women. (A quick Google search puts the world’s tallest woman at 7’9”.)

So how should we express this difference in height?

A simple approach would be to say that the difference in average heights for men and women is about 5” (roughly correct in the UK). But the range of men’s heights either side of the mean is much greater than 5”, as is the range of women’s heights, so tall women may have much the same heights as short men. Without knowing more precisely how spread out heights are, the difference in means of 5” tells us very little. The statistical tool we need to measure the spread of a distribution is the standard deviation.

The standard deviation for adult heights is roughly 3”, and these heights  follow a Normal curve, so we can do some calculations. We can calculate that about two-thirds of women will have heights within 3” of the mean, and that about 90% of women will have heights within 5” of the mean. And it follows that about 5% of women will have heights above the mean height for men.

So a useful summary might be this: men are on average taller than women, with 95% of women being shorter than the average man. Or, if you prefer: though men are on average taller than women, 5% of women are taller than the average man.

The NCP survey doesn’t seem to give standard deviations, so let us illustrate the point with a guesstimated figure of 2.5. Let us assume a Normal shape too, and then we can do some calculations. A standard deviation of 2.5 would mean that 95% of women’s parking scores were in the range 8.4 to 18.4, with 95% of men’s scores in the range 7.3 to 17.3. Clearly that would mean a great deal of overlap. And a further calculation would give us this: women are better parkers than men on average, with 67% of women out-performing the average man. Or: compared with the average man, two-thirds of women are better at parking – and one-third are worse.

That is a better indication of what we mean by ‘better’.

Neil Sheldon has taught at The Manchester Grammar School for 40 years. He is a Chartered Statistician and Fellow of the Royal Statistical Society. He has been an RSS Guy Lecturer since 2007.  He is also course leader for the Certificate in Teaching Statistics offered by the RSS Centre for Statistical Education

So what would the basics look like? Here’s a Minimum Statement. Minimum is the word. No one expects journalists to be senior wranglers. Their skill set has to include imagination, presentation …words. But too often journalism educators have left numbers out. And the result, some would say, has been impoverished reporting of social police, crime, science, climate change …fields where you need to be able to assess risks, calculate proportions and (most important) spot other people’s errors and exaggerations.

Recent figures show applications by students living in England to study at university from this autumn are down, and there’s interesting comparative data from Wales, Scotland and Northern Ireland. But one year’s figures could be a blip. A mite of caution is a basic requirement – the alternative is breathless enthusiasm for any given set of figures, with the risk of misleading readers, viewers and listeners.

Take a look at our dozen or so cautionary points and let us know if they sound like the basis for a journalistic career in the 21st century.

It looks as if a fair numbers of candidates (the Police Foundation is regularly updating runners and riders) will have a background in policing and criminal justice. Some – Alun Michael has thrown his hat in the ring in South Wales – may even have been government ministers with policing responsibilities. Others –Keith Hellawell the former drugs ‘tsar’ has been mentioned — served as chief constables.

But an expert background may not immunize candidates against stretching the connection between the police and crime. The relationship between police numbers and activity and crime levels is far from simple. Most police work has nothing to do with crime, but transport and traffic, social order and as a first port of call for members of the public with nowhere else to turn.

We hope, as the campaigns hot up, to help public, press and PCC candidates take a cool look at the data. That’s not just crime stats, but all material relevant to assessing the work police do, much of which isn’t about crime. Our hope is for debate to be conducted on the basis of evidence and attested knowledge rather than speculation or assertion.

We’d welcome your thoughts about what such a statement should contain. Put them down here or contact d.walker@rss.org.uk