Here’s a 5 minute film I did on Newsnight last week, about how politics needs better data. Specifically, it’s about how politicians misuse statistics, how we can stop them, and how we can generate better evidence on what works, and what fails.
If you’re interested in more on this topic, well… there are some good examples of dodgy government stats on this site, and in the chapter on Dodgy Government Statistics in my latest book. For more about randomised trials of government policy, here’s a Radio 4 documentary I did on them, here’s a Cabinet Office paper I co-authored on them, and they crop up in this that I did for the Department of Education. If that’s not enough then I’ll write some more.
I took great pride in the fact that my book Bad Science was first reviewed by Viz and the British Medical Journal. In that vein, here’s a 90 minute interview I did on stage with comedian Richard Herring in the Leicester Square Theatre. I have bad hair, we cover a lot of dorky material – epidemiology, medical politics, smears, govt statistics, evidence based policy – he keeps steering the conversation back to my penis. I think it’s an interesting format.
Here’s a talk I did last year that’s just popped up online. The Cochrane Collaboration is a fabulous organisation, producing gold standard “systematic reviews” summarising all the data that’s ever been collected on important questions in medicine. Cochrane have become great by inviting criticism: for example, they run the Silverman Prize, for the best essay or paper pointing out stuff that they’ve got wrong. At their 21st birthday conference, they asked me to do a talk on what they should do next. I said they should get better at talking to patients; better at talking to policy makers; and better at talking to machines. Here’s the talk: Read the rest of this entry »