Evidence and objectivity

Even the best evidence is subject to interpretation bias, says The Scientific Alliance.

Scientists may lay claim to objectivity, but we should always be cautious in accepting the conclusions of others. Take, for example, a news item from this morning: Pesticide ban to save bees ‘based on flawed research’. The argument made by Norman Carreck, who works at the University of Sussex on bee research, is that other scientists had used unrealistically high levels of neonicotinoid insecticides in experiments with bees, so wrongly concluding that these chemicals are a likely cause of a decline in numbers.

To add further interest to this intellectual spat, one of the scientists accused of drawing misleading conclusions is a colleague of Mr Carreck’s at the University of Sussex, Dave Goulson. The importance of this is that Prof Goulson’s work was one of a number of studies which were influential in the introduction of a temporary ban on the use of neonics in the EU. This took effect last year and has been blamed for extensive losses of oilseed rape through attack by flea beetle.

Mr Carreck’s criticism is that his colleague’s study actually looked at a worst-case scenario rather than being typical of the exposure bees would actually receive in the field. Prof Goulson, not surprisingly, defends his results, claiming that even higher levels of pesticide had been found in pollen than his team had fed to bees. Mr Carreck, however, also criticises other studies on the basis that bees were fed a single dose of pesticide rather than receiving it over the extended period of time representative of foraging. This, he said, is comparable “to the different effects on a human of drinking a bottle of whisky over one hour, 24 hours or longer”.

The rights and wrongs of this affair are not clear-cut without going into the detail of all the papers, but the story does sound alarm bells when we consider a report from December: Scientists accused of plotting to get pesticides banned (previously commented on in Misuse of science). It would be easy to conclude that there was an inherent bias in some of the studies, with researchers setting out to gather evidence to support a belief or expectation rather than being scrupulously objective. This sort of behaviour is reinforced by the peer-review process, which is supposed to guarantee that only good-quality research gets published but in reality tends to support received wisdom rather than challenge it.

It is understandable that people interpret and use data to support their case; it happens all the time. It is perhaps more surprising that the public in general haven’t cottoned on to the fact that scientists are as prone to this as anyone else, and still trust scientists more than most other groups. The phrase ‘blinding people with science’ is not too far from the truth: as soon as something is quantified, it begins to acquire an aura of truth.

Economists are prone to similar biases, arguably more so, since the dismal science is actually more of a black art than a science. Admittedly, economists do quantitative analyses which are often quite rigorous, but the outcome is highly dependent on the assumptions made. It is interesting, then, to see another headline; Wind power ‘adds resilience to UK energy market’. This refers to a report produced on behalf of RenewableUK by Cambridge Econometrics (The impact of wind energy on UK energy dependence and resilience). The headline figures are that wind power saved the UK £579 million in fossil fuel costs in 2013 and a generating mix with a high proportion of wind would result in a price rise of only 4% for electricity, rather than an 8% increase if end-of-life coal and nuclear plant was replaced by gas.

Looking at the study a bit more closely, the figure for 2013 savings in coal and gas costs takes no account of the cost of wind energy. It also takes a DECC figure of 28.4TWh of electricity generated by wind (7.9% of the total) and assumes that an equivalent proportion of gas and coal generation was avoided. So, since we know that coal and gas backup for intermittent wind generation has to be ramped up and down rather inefficiently and ‘spinning reserve’ kept as standby, the actual amount of fossil fuel would be somewhat less. By how much, we don’t know, but it is clear that there is less and less displacement of conventional generation as the proportion of renewables increases.

To summarise that, we can say that the figure of £579 million saved in fossil fuel costs is, on one hand, inaccurate because it does not account for the inefficiencies associated with backing up renewables and, on the other hand, misleading because the cost of wind-generated electricity has to be set off against it. The other fact is that wind energy relies on subsidies to compete with thermal generation, paid for by consumers. It goes against experience to believe that renewables have actually lowered prices, which is the implication of the report.

The second claim is that more wind energy would make the UK’s electricity supply more resilient, quoting a 4% relative drop in electricity prices if more wind was in the mix. However, this is based on DECC’s high gas price scenario, a 41% increase in price in real terms. Even for a point halfway between the high and central price scenarios, ‘high wind’ and ‘high gas’ give equal annualised costs. And this is before allowing for the additional backup and infrastructure costs for wind. Resilience is itself a misleading word in this context: wind makes the energy supply less rather than more secure, whatever the price may be.

He who pays the piper, calls the tune, and in this case the tune is very much a renewable one. A worst case scenario has been used to give an apparent modest advantage to wind energy, sufficient to generate positive headlines.

But, fortunately, there are people who are prepared to challenge misleading information. Take this recent case: Vicar wins ban on anti-fracking claim. In this case, a retired vicar used the Advertising Standards Authority to prevent a local activist group from issuing leaflets giving unsubstantiated and exaggerated claims about the supposed dangers of fracking. The damage had doubtless already been done, but small victories such as this should be celebrated. Long live independent thought.

 

Martin Livermore

The Scientific Alliance

St John’s Innovation Centre

Cowley Road

Cambridge CB4 0WS

_____________________________________[



Read more

Looking for something specific?