Archive for the ‘Statistics’ Category

‘90% of all data was created in the last 2 years’ – WTF??

July 9, 2013
Big data

Oooh big data!!

It’s been a while since I last posted (I have a good excuse, I was organising my wedding and then off on honeymoon!). But something’s been bugging me – a strange claim I’ve seen in a few places. Legend has it that 90% of all data was created in the last two years (ta @james_randerson for the link).

I find it difficult to get my head round this. I think the statement is supposed to convey the humungous amount of new data that is being created every day – from genetic sequencing to personal data recorded by smart phones. But it seems to be one of those statements where the story has been simplified so much as to be almost completely meaningless – what kind of ‘data’ are we even talking about? (more…)

The risk of accuracy

February 6, 2013
Percentage symbol

Watch those percentages… (image by NavBack)

I love reading about risk. As someone who finds maths a bit of a challenge, it always becomes more real to me when the numbers relate to the real world. Using numbers to explain the risk of something happening seems like an excellent way to harness the power of maths – numbers are objective, concrete, and not vague.

But as always in the messy real world, it’s not quite that simple. That’s why I was really inspired by this blog post from the Risk Science Center at the University of Michigan, explaining how extra levels of ‘accuracy’ – more numbers – can actually make risks harder to understand. (more…)

I am not a number – I’m a statistic

May 29, 2012
Number nine

Today’s post is brought to you by the number 9 (it’s my favourite)

It’s very annoying being treated impersonally, as ‘call number 18 in a queue’, or as a poorly defined Amazon customer (no I don’t want more baby toys, I only bought one for my friend). But we all exist as impersonal datapoints on many, many spreadsheets all over the place.

I’ve ranted with my friends about the strange decisions taken by insurance companies: why is insuring my car against fire and theft MORE expensive than comprehensive cover?? How is Catford classified as a flood plain?

But gradually I achieved (for me) a zen-like level of calm – I remembered that really I’m just a statistic. In this particular case, a woman in her twenties with a very old Ford Fiesta. And my friend is simply a Catford resident whose neighbours have made several claims for flooding (probably burst pipes, rather than Biblical inundations of south London). We’re all statistics. (more…)

NHS productivity – room for statistical improvement

February 13, 2012
Placard reading 'what do we want? Evidence-based change. When do we want it? After peer review'

Evidence-based placard

The words ‘statistics’ and ‘measures’ often strike boredom into the hearts of even the nerdiest. Especially when we’re talking about ‘measuring impact’ or any similar corporate jargon phrase.

But this week’s news about the uncertainty behind reports of ‘falling NHS productivity’ just shows how crucial and controversial these measures can be. You might think that before you make a huge change to the health system, you would want to know how it’s performing. The Lancet has shone a spotlight on the fact that it’s very, very difficult to measure this, and it’s not actually clear whether NHS productivity is going up or down.

This isn’t a trivial thing. Measures might be a bit soul-destroying at times, but they are literally the raw material of science, so ‘evidence-based’ policymakers should always be looking very hard at the measures they’re using. (more…)

Nuclear hysteria

March 27, 2011
Radiation dose chart by xkcd

Radiation dose chart by xkcd

Did you know that taking a transatlantic flight exposes you to far more radiation than spending a day near the Fukushima plant would? No, neither did I. Just one fact that doesn’t seem to have come out of the media coverage of the problems at the plant following the earthquake and tsunami in Japan.

Luckily the web comic xkcd has produced a brilliant diagram illustrating why we shouldn’t get too hysterical about this. Clearly there are huge problems for people in Japan, but so far their exposure to radiation has been quite low and is unlikely to cause any illness.

As one article on the BBC website pointed out, many perfectly healthy people in America pay hundreds of dollars for a full-body CT scan as a check up. These scans can expose you to a radiation dose equivalent to being 1.5 miles from the centre of the explosion at Hiroshima. And in case that sounds a bit hysterical, the article goes on to say that “because more than 70 million CT scans are carried out each year, the US National Cancer Institute has estimated that 29,000 Americans will get cancer as a result of the CT scans they received in 2007 alone.” (more…)

Census excitement

March 9, 2011
Tick box

I'm exercising my right to tick a box!

I just got home to find my 2011 census waiting for me on the doormat – with its attractive purple colour scheme and the impressive words ‘your response is required by law’ filling me with a sense of purpose.

I’m not joking – I am genuinely excited by filling out the census, although I’m not sure ‘nerd’ counts as an ethnic group. The census is a tool for a load of nerds at the Office of National Statistics to gather a massive glut of data, which they can gorge on for years to come. (more…)