Sir

Your News Feature “It's a scoop!” (Nature 426, 222–223; 2003) thoughtfully explores the dangers of excessive competitiveness among researchers racing to be the first to publish high-profile scientific results. Although this can lead to secretiveness or sloppy work, a much worse possibility is that it is encouraging scientific fraud or other forms of misconduct.

For example, I asked a colleague who was familiar with the details of a recent case of scientific fraud why — given the risks — he thought the perpetrators had done it. He drolly observed that the real question is not why a few scientists commit fraud, but why more don't do it. He went on to say that since the maximum penalty for getting caught (dismissal) was no worse than the routine penalty for not producing enough high-profile papers (no job), most junior scientists, at least, have nothing to lose by committing fraud.

I suspect that, although blatant cases of fraud in high-profile journals are probably rare, more subtle forms may be quite common: skipping a control or two, claiming that the same result was obtained in a repeat experiment that was not made, or failing to mention that some essential cross-checks did not give the required result.

It is almost impossible to determine how often this kind of thing is happening, but that is not a cause for complacency. The notion that fraud will always be detected when the results are shown to be false assumes that the fraudulent results will always stand out sufficiently to attract scrutiny. Less blatant cases may pass unnoticed, or the researcher trying to push his or her work just a little might actually have guessed right. Furthermore, how many researchers these days have time to repeat whole chunks of previously published work, or go back over older literature and critically examine every result for possible signs of fraud, let alone be willing to do anything about such suspicions?

What can be done? It probably requires changes in the way science is funded, and especially how junior scientists are employed. Some sort of international inspection body might help, similar to that used to find and punish cheats in the sporting world. And as individuals, we can all help to develop a scientific culture that discourages fraud or related misconduct as much as possible, while finding ways to increase the rewards for honest but effective competitors.