Many are won over by Freakonomics' charms, including Huw Richards, who meets the duo behind the phenomenon.

One measure of the success of Freakonomics , an original take on economics by journalist Stephen Dubner and University of Chicago professor Steven Levitt, is that a study guide is available less than a year after the book was first published.

Students can download it from the book's website. Teachers must go to the publishers and obtain a code before downloading the companion volume, which reveals the answers - a sensible precaution against students getting hold of answers to questions they may be set in tests. But Dubner admits: "We kind of hope some people do cheat, because a lot of what we do is about cheating."

That comment epitomises their irreverent approach and their fascination with human behaviour. They examine potentially contentious social phenomena - for example, their conclusion that the decline in crime in the US since the early 1990s is an unintended consequence of the legalisation of abortion is about as provocative as you can get. They subject hypotheses to rigorous regression analysis designed to sort the mythical from the significant and causation from mere correlation, but without taking political positions - or, as Dubner puts it, "We don't have a horse in any race."

³ÉÈËVRÊÓƵ

They argue that, far from being a "dismal science", economics can both entertain and, by returning to its roots as a science of human behaviour, illuminate issues far beyond its usual remit. In a characteristic inversion of stereotypes, the academic Levitt is happy if readers find the book entertaining, while the journalist Dubner hopes it will change how they think.

Levitt points out that, unlike political science, economics is defined by methodology rather than subject matter. That methodology, Dubner argues, "is a toolkit you can apply to thinking about almost anything".

³ÉÈËVRÊÓƵ

Many rookie economists find the subject as confusing as a foreign language. Levitt, coming upon it as a student, spoke it fluently from day one: "I was just good at it. I'd always thought the way an economist does." The subject matter came less easily. "I wasn't terribly interested in markets and interest rates. What had always interested me was ideas and finding the answers to questions. And I like simple ideas and questions, whereas there was a tendency to think that the more complex the argument, the better the work."

He denies that this makes him a pioneer: "I'd like to claim that I'm a trendsetter, but I'm not. I'm far more an indication of trends, with a whole generation of economists extending the range of the subject. This is partly because a lot of traditional problems are very well understood."

What sets him apart is being very good at what he does - and his working partnership with Dubner. Levitt was hardly toiling in obscurity when Dubner interviewed him in 2003. He was deemed worthy of the time, space and resources The New Yorker devotes to a profile. He had also won the biannual John Bates Clark medal for the best American economist under 40 - previous winners include Gary Becker, Milton Friedman, Joseph Stiglitz and Paul Krugman.

But he had no plans to write a book - there was no need given his productivity in research papers, the hard currency of academic economics - let alone a book that currently outsells Dan Brown's The Da Vinci Code on Amazon.

If Levitt supplies the ideas, the exposition and intriguing chapter titles - such as "What do schoolteachers and sumo wrestlers have in common?", "Why do drug dealers still live with their Moms?" and "How is the Ku Klux Klan like a group of real-estate agents?" - have the ring of gifted journalism.

³ÉÈËVRÊÓƵ

The collaboration rests on mutual appreciation. Dubner talks of the pleasure of "climbing into the mind of the warped data detective", while Levitt says: "When (Dubner) came to see me he'd read everything I'd ever written. He thinks like an economist, he's respected by economists and has an extraordinary talent for communication."

The underlying assumptions are those of the working economist: that incentives are the strongest influence on human behaviour, that marginal analysis is more relevant than looking at averages, and that datasets are more useful than anecdote or surveys. The aims are still those of an academic paper: to take a phenomenon and use the economics toolkit to explain it. But the analysis is buttressed by memorable anecdotes rather than equations - for example, the idiosyncrasies of name-giving. Every child christened Yale in California in the 1990s was white; the Harvards and Princetons were exclusively black.

As well as topping bestseller charts, they can point to "three or four e-mails a week from students thanking us for stopping them hating economics" and a growing list of universities - including Georgetown, where the book is used to teach public policy, and Berkeley - that are using the book as a set text.

³ÉÈËVRÊÓƵ

They have begun work on a second book, but its completion may be four or five years away because of the time serious data analysis can take: "For every idea that comes to something, there are nine or ten that do not,"says Levitt. He has, for instance, never been able to come up with a satisfactory analysis for traffic flows. Subject matter likely to be covered next time includes sport (a shared enthusiasm), health and prostitution.

There will, undoubtedly, be imitators - and Levitt and Dubner accept that their success owes something to the public taste for statistically based argument, demonstrated in particular by journalist Malcolm Gladwell's The Tipping Point: How Little Things Can Make a Big Difference , published in 2000. And this may not be the only way in which they are emulated.

Acknowledging the benefits of his partnership with Dubner, Levitt says: "I'd expect to see more academics doing this in the future."

Freakonomics by S. D. Levitt and S. J. Dubner is published by Penguin, ?8.99.

³ÉÈËVRÊÓƵ