Statistics. What are they good for?

In July 2015 on July 28, 2015 at 7:50 pm


Sarah: Statistics. What are they good for? While the answer is not ‘absolutely nothing‘, it’s kinda hard to nail down nonetheless.

Yes, it’s about numbers. It involves maths. It boils down to complex analyses. I’m afraid you’re gonna need a spread sheet.

But before you run away in a screaming fit of ‘whoah, that’s not for me!’, give me a wee chance to explain why statistics — also known as ‘stats’ — really matter.

Like last week’s post, it boils down to being human.

Over millions of years of evolution, us humans have developed some opaque but useful ways of making sense of our world. We look for patterns. We look for evidence that supports what we already believe. We aim for the path of least resistance.

All this makes sense when you’re living in a forest or a dessert and you’re hunting food every day and you’re keeping your eyes peeled for wild animals and you’re trying to stop your children eating poisonous mushrooms and you don’t understand why a visiting stranger with a snotty nose who sneezes all over your family can lead to numerous deaths in your community.

But if you live in developed world where you have the luxury of not being in survival mode 100% of the time, you can use science to make sense of your world.

Statistics is the tool we need to take the bias that comes with being human out of a scientific analysis.

It’s hard to explain, but maybe pulling in an expert will help. Last week I had the chance to chat with a man who knows a lot about statistics: Professor Corey Bradshaw.

I asked him quite simply:

“Statistics. What does it even mean?”

Here’s what he said:

“As humans, we tend to taint what we see with what we want to believe.

And what we want to believe is influenced by experience, by belief systems, by biased sampling — that means selective recall of information — in our own brains.

We tend to ignore evidence if it doesn’t fit.

Statistics is a tool that allows us to separate a perceived pattern from what’s really going on. It allows us to remove the bias. It allows us to make an objective conclusion about what the real relationship is. You can’t do that by eye.”

The proper application of statistics allowed Professor Bradshaw to extract a meaningful and unbiased conclusion from an enormous body of data describing extinction of large mammals, and climate records from the world’s northern continents around 10-50 000 years ago.

The conclusion was that small warm periods in the climate were directly linked with animal extinctions. It’s a conclusion that the human mind alone could never have reached amongst all the noise and distraction of thousands of data points.

You can read more about how climate change caused extinction of mammoths, bison and hairy rhinos in my article for The Lead South Australia here.

[image thanks to Adrian Sampson on flickr]


Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s

%d bloggers like this: