Although The Knot Garden strives to be self-reliant, it acknowledges its dependency on nature. In particular, it appreciates the important role played in plant pollination by the many species of bees that visit in the Spring and Summer. The bees are the life force of The Knot Garden. Without them, it would not exist.
Having learned from Ben Goldacre’s book just how much bad science there is in the world, I am keen, in the spirit of balancing things up a little, to find some examples of good science. I recently discovered a candidate project in the form of a study published last year by the Centre for Ecology and Hydrology.
The purpose of the study was to determine the extent to which the treatment of oilseed rape with neonicotinoids has impacted bee populations in the UK. Evidence collected during the study suggests a clear correlation between neonicotinoid use and declines in bee colonies, although there is a caveat that the outcome should be seen in the context of other environmental factors such as habitat loss, climate change and mite infestations.
On the one hand, we need to be wary of taking at face value the sort of headline statistics favoured by the press in their relentless pursuit of sensationalism. On the other, we should resist the temptation to simply reject all scientific studies on the grounds that simple truths cannot be proven using statistics. In the absence of a register of research studies operated along the lines suggested by Ben Goldacre in his book, we need to have a way of critically assessing these papers for ourselves.
Given the limited time I have for evaluating scientific studies, I decided to adopt a high level framework that would help me to quickly determine a study’s value. Here are the criteria that I selected, together with my take on how the neonicotinoids study fares.
First we have impartiality. Any study sponsored by an organisation that has a commercial interest in the outcome should be ruled out immediately. The Centre for Ecology and Hydrology was set up specifically to act as an independent research agency and has adopted impartiality as a core guiding principle. And the project was subject to the scrutiny of an independent scientific advisory committee.
Second comes credentials. Are the individuals well qualified to perform the research? The scientists who took part in the project have impressive CVs. Those for five of the seven authors can be read on the website of the Centre for Ecology and Hydrology.
Third , we need to examine the scope of the research. Were the scale, duration and depth of the project adequate to allow reliable conclusions to be drawn? The neonicotinoids study was conducted over a period of eighteen years, covered over 4,000 square kilometres of farmland and examined populations of bees of 62 different species. Detailed information about the study can be found online.
As a fourth element, we should consider how well the study was designed and how effectively its conditions were controlled and its measurements taken. This is the hard part and I have not had the time to perform more than a cursory evaluation. However, one of the authors helpfully gives a detailed analysis of the statistical methods used in one of his blog posts.
Naturally, The Knot Garden is very keen to understand whether neonicotinoids harm bees. But, beyond that, it advocates good science. I believe that this important study, with its implications for the future of bees, qualifies as such.