How do you go about really demystifying statistics?

Tim: Well, that’s a very good question, how to demystify statistics. Sometimes it’s simply a case of finding out what the statistics are when previously there aren’t any, or the statistics aren’t being reported. So something I’m working on at the moment, for example, is the idea of food miles, how far our food travels. And it’s often that those who care about such things, the local food movement environmentalists tell us we should be very, very worried about how far food travels. So, of course, a simple question is, well first how far does food travel, and second is, well how much of a problem is that. So that’s one thing is to simply actually ask well, hang on, we have a story here, we have a concern but we don’t have any numbers, can we go and find the numbers? And sometimes it’s easy, sometimes it’s difficult, but sometimes the simple process of gathering the numbers and working out how big or small a problem is tells its own story.

At other times, we simply pick up the reports that the headlines have covered and we ask ourselves well what’s going on here. At the moment, we’re working on the report on hazardous drinking that made the headlines recently. Put the stockbroker belt as the place, the most hazardous drinking capital of Britain, and so what we do is a report for the listeners, and we tell them what the report really says, and usually what the report really says is something different. So there we revealed a secret some of the newspapers didn’t bother to tell you, and we’ve told the story behind the story, and I think that’s also engaging - people want to be involved.

At other times, we’re just trying to put a human face on a story. I mean this evening I’m going to interview a software engineer mathematician, basically a nerd, who’s made millions writing computer programmes to carry out foreign exchange trades. There’s not really statistics behind this but what we’re trying to do is ask how much money can your maths degree really make you in the City, and that’s a human interest story, but of course it’s about numbers as well, so sometimes that’s the way we do it.

But, you know, I’m new to this, this is the first time I’ve presented a radio series. I’ve worked before on television and in print, and what’s exciting to me is it’s obvious that the producers who work on the show are fantastically experienced and fantastically creative, so I’m looking forward to them teaching me new ways to bring numbers to life.

Do you believe that economics, numbers and statistics can solve every day problems?

one of the best ways to understand the world is through statistics

Tim: I think they can because ultimately what are statistics but an attempt to understand the world around us, and many of the decisions that we make, to make smart decisions, we need to understand the world. So a decision about where to send your child to school, or whether you should be buying or selling your house, is going to of course depend on your personal circumstances and your own judgment, but it might also depend on whether you can trust school league tables, or whether you can believe those indices that say that house prices are going up, or house prices are going down. So if you want to move beyond your own personal experience, if you want to move beyond anecdote and make sensible decisions, you have to understand the world. And one of the best ways to understand the world is through statistics. But statistics have their risks, they can mislead us, and hopefully More or Less listeners are going to be less misled than others.

I think the straightforward challenge for us on More or Less is to choose some stories that have everyday relevance. It’s not that this is a ‘how to do it’ programme, that every single story is going to make your life better. Some of them, I think, we report just because we feel that they’re of public interest or we feel they’re just fun. But I think we also need to pick stories, we need pick issues, where what we’re covering is actually going to tell you something that you can go out and use. I think that’s an aim for the programme and I’m fairly confident we’re going to be able to fulfil it.

Kevin, how does More or Less work for you from a teaching perspective?

Kevin: I think it works very well from a teaching perspective. There’s always a risk actually in thinking of this sort of thing from a teaching perspective. If people think that the point of the programme is to teach them stuff, they’d probably turn off in droves. It doesn’t actually come across like that, and I think that’s a major strength, because there are other programmes that have looked at this sort of thing that have come across as really rather too teaching - More or Less just doesn’t, and I think that’s a great strength of it.

I mean what we try to say in the written teaching material that we produced at the OU, is to try to tell interesting stories about things that are relevant to people’s lives and involve numbers and statistics, economic concepts and so on like that. It’s sometimes quite difficult for us to do that in a print medium because the production times are quite long. You know, what people are interested in this month, by the time the course material’s appeared in their post box, they’re probably not terribly interested in it any more.

More or Less is produced on a current affairs schedule, therefore it can pick up things when they’re in the news, and it can build on those, and in doing that, it can persuade people that the approaches of economics, the idea of using statistical data where statistical data is available, going and looking for it where they aren’t available and so on, can apply to a great wide range of things, and that sort of rubs off in persuading some people to say, well there’s a lot of technical stuff here that I’d actually like to know more about. So it complements things we do in teaching in a lot of different ways.

However, I have to say that’s not the main purpose of the programme, and it’s not the main purpose of the OU’s involvement with the programme, that’s concerned much more with just showing people how this is interesting and relevant to them, showing anybody who happens to tune into Radio 4, and that can be hundreds of thousands or millions of people.

If people want to get to grips with the numbers that hit the headlines, what are the key concepts we should get our heads around?

if you haven’t got the context of the number then you don’t really have anything

Tim: Well, context is the most important thing. If you haven’t got the context of the number, both in terms of its scale, its importance, but also in terms of the political context, who’s telling you this number and why are they telling it to you, if you don’t have that, then I think you don’t really have anything. But I think the second very important concept, when it comes to statistics, is that of causation. So here’s a straightforward example. Taller children have greater reading skills. So does that mean we should put children in a mangle and start stretching them, or put them on vitamin supplements? Well, no, taller children are older, and older children have better reading skills. So the fact that taller children have better reading skills hasn’t really told us anything.

Now, that doesn’t mean that you can’t come up with a meaningful statistical analysis of the relationship between a child’s height and a child’s reading skills. So you can compare tall ten year olds and short ten year olds, tall six year olds and short six year olds, so you can do good work. But if you’ve just heard a statistic reported on the news without that sort of concept being mentioned, then often what you’ve got is a spurious correlation, and the difference between a spurious correlation and causation is a very, very important one. And I just wish the media paid a bit more attention to it because often, as a listener, you’re hearing, you don’t know whether you’re hearing a report of work that has been done properly to take account of causation or work that hasn’t been.

So those are things. Simply the size, the scales, the number, is it a big number, very important. The political context, who’s telling me this, why are they telling me it, that’s very important. And then, are we hearing about correlation here, which is not that interesting, or are we hearing about causation, which is always very interesting but much harder to prove.

What do you want like listeners to take away from More or Less?

when somebody tells you a statistic, ask yourself, 'Why are they telling me that?'

Kevin: I think the thing I’d really like you to take away, is when somebody tells you a statistic, ask yourself, 'Why are they telling me that?' That’s it.

Tim: Well, the message would be that if you listened to More or Less, at the end of the half hour, you will not only have had some fun but you will also have learned a couple of things that you didn’t otherwise know about the news that week. So I’d like to think of it as a programme which is not only enjoyable but is actually going to give you the inside story on what the media are not really reporting and should be.

This interview was originally published in October 2007

  < previous   Page 3 of 3

Content last updated: 22/10/2007