In Lake Wobegon, the fictional town invented by the US humorist Garrison Keillor, “all the women are strong, all the men are good-looking, and all the children are above average”. In keeping with Keillor’s gentle dig at the inflations of self-bias, if Lake Wobegon had research laboratories you can be sure that all the experiments would work, all the results would be significant and all the scientists would work safely.

This week, Nature reports the initial analysis of results from the first international survey of scientists’ attitudes and behaviour towards lab safety, conducted by the University of California, Los Angeles, together with Nature Publishing Group (see page 9). The analysis hints at a Lake Wobegon bias in perceptions about safety: one-third of scientists say that safety is more important to them than it is to their colleagues, with only 2% voting the other way. Although most respondents say that their labs are safe places to work, they simultaneously report behaviour, such as frequent lone working, that seems to belie that confidence.

The survey was done to improve understanding of lab safety culture. Health-and-safety officers have long complained of a lack of international data. It would be premature to draw immediate conclusions from the quantitative results — for example, almost half the respondents reported being injured in the lab — because few other comparable data have been collected. But the results do caution against complacency.

So, as you return to your laboratories in the New Year, look around the benches, observe your own working practices and those of your colleagues, and evaluate your relationships with supervisors and safety officers. Not everyone can be above average — but awareness of how perception clashes with reality can help lift standards for all.