Home » Journalism » Lies, Damned Lies and Damned Liars: The Truth About Statistics

Lies, Damned Lies and Damned Liars: The Truth About Statistics

Statistics

To quote the immortal genius that is Homer Simpson “you can prove anything with facts!”

He’s right, you can.

As long as you don’t know what you’re doing.

Today I read a free newspaper on the train. In it, there was a small article of 50-100 words about a study from New Zealand. Within there was one quote statistic. Out of the 160-odd patients in the sample it was found that stroke victims were 2.3 times more likely to have cannabis in their system.

I do not debate the validity of this statement, or do I have any reason to suspect the paper is inherently flawed. What I do debate is the interpretation of this statistic which was stated in the very first line of the article.

“CANNABIS USERS TWICE AS LIKELY TO HAVE A STROKE.”

Now, anyone with a rudimentary grasp of statistics or numbers can tell you that that figure does not produce that conclusion. Not. Even. Close.

What it does suggest is that people who have a stroke are 2.3 times more likely to lead a lifestyle in which they smoke cannabis. It does not mean what the headline says. Cannabis will not make you have a stroke.

Not only does it break the fundamental rule of misrepresentation, it also breaks the cardinal sin:

CORRELATION DOES NOT EQUAL CAUSATION.

Just because two numbers seem to rise and fall together does not mean they connected in any way whatsoever.

Now, I am not suggesting the person who wrote the article was lying, though that is a possibility. I am suggesting the most likely reason for this is that no-one has a clue how to handle statistics.

For all the qualifications you need to become a journalist, an ability to understand statistics isn’t one of them. Yet look in a paper! In almost every article you will find one statistic or another. Often out of context, often completely misrepresented by the headline.

To make matters worse, the average person on the street doesn’t know how to use them either. It leads to this horrendous state of affairs where people can use pretty much any numbers they want to say anything they want because people don’t know enough to either use or interpret statistics properly.

So the next time you read a statistic in a newspaper, or see one on TV, or in an advert just ignore it. 94.3% who do are smart for it. And that’s a fact!

Advertisements

4 thoughts on “Lies, Damned Lies and Damned Liars: The Truth About Statistics

  1. I took a statistics class in my first semester of Uni – it was perhaps one of the best decisions I made for electives. I am now of the belief that anywhere a mean is reported, a standard deviation – at least – should be reported with it.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s