Learning to understand hidden data

I was watching some videos today and stumbled across a few by a guy named Ben Goldacre. Goldacre is a doctor, who has become an epidemiologist: which is the science of how we know in the real world if something is good for you or bad for you. That may sound fairly abstract, but it’s actually something we all experience many times a day.

Have you ever convinced yourself that having another block of dark chocolate is ok, because it’s good for you? Or perhaps another glass of red wine? Or taking fish oil? These are all acts that have probably stemmed from an epidemiologist at some point; who has completed a survey, crunched the data, and decided that the red wine you’re now about to drink is going to reduce your chances of getting cancer.

The contention of Goldacre’s talks are this; that the surveys and data that we hear repeated to us through various media channels are often flawed scientifically and that we should be aware of this. Further, we should keep in mind that most things we believe are merely points of view that we’ve held for one reason or another and should be remembered as so. A point of view is, after all, just a view from a point. (Thanks Col!) Further to this, things react differently when they’re being observed. But that’s a topic for another day.

But so what? What does this mean for us, really? What is the macro trend here?

We’re going to see more and more data pouring into our lives – even if we don’t know about it. Data sits below the consciousness of most citizens. Do you ever question whether the pill given to you by your doctor successfully passed it’s double blind placebo test as well as beating out the competition drugs that were already on the market? I certainly don’t.

And so, it’s important to realise that as things get more complex in terms of the data we become exposed to through our more quantified lives, things are largely still uncontrollable. I love how in the video Goldachre comes back to the traditional view; If you want to live longer then you should eat more fruit and veg, be fit, not drink or smoke and keep your weight down. But you might still die early.

It’s becoming more and more important to understand the differences between mysteries and puzzels.

4 thoughts on “Learning to understand hidden data”

  1. Hey Steve – meaning through data is definitely complex and getting more so as our methods for analysis get better. Already the media are happily ignoring the difference between a correlation and a causal relationship. You might find this article interesting :) http://www.wired.com/medtech/drugs/magazine/17-09/ff_placebo_effect?currentPage=all&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+C2C-InTheNews+%28Feed+-+Coast+to+Coast+-+In+the+News%29

  2. Hi Steve. Interesting topic. When you get around to that discussion for another day about how things react when being observed l direct you to “The Hawthorne Effect” which l think was one of the first reported incidences of this.  

Comments are closed.