The experiment is one of the best forms of narrative we have, and one of the most underused. You state your aim, you say how you are going to test it, and then you test it. Results, conclusion and potential closure. As a process it is massively underused in the very area where it seems to fit best, documentary (science documentary in particular). The reason Supersize Me is so compelling is that at its heart it is an experiment.

Aim: To see how healthy McDonalds foods are.

Apparatus: A personable ginger bearded chap Morgan Spurlock and lots and lots of McDonald Outlets.

Method: He eats all his meals, three times a day, in a McDonalds for thirty days. He must supersize his order if asked. He will eat everything on the menu during this period. Before starting he will have a full medical, and will be checked by three specialist doctors are intervals during his experiment.

Result: Poor health.

Conclusion: Yep, McDonalds ain’t all that healthy for you.

The beauty of course is both Spurlock?s force of genial personality when doing it, and the nice way he allows his story to lead him into other stories (there are entire films in the marketing of McDonald?s, in school dinners, in the rise of obesity). But it reminded me that the experiment, as a way of testing a hypothesis, is almost never used in television documentary anymore. Hopefully Supersize Me will rectify that.