The curse of policy-based evidence

Journalists and others must question the science used to justify policy decisions Copyright: Flickr/JeremyRyan

Send to a friend

The details you provide on this page will not be used to send unsolicited email, and will not be sold to a 3rd party. See privacy policy.

Journalists and scientists must guard against policymakers selecting science to legitimise pre-chosen and politically-motivated policies.

In the final session of last month’s World Conference of Science Journalists, Britain’s chief scientific advisor, Sir John Beddington, made a revealing confession.

He told how, when interviewed for his post, he had been questioned about what he would do if a prominent science journalist asked him to comment on a government decision that ran counter to his advice. His response was that he would avoid the question and thus, by implication, any embarrassment his answer might create.

Beddington’s reaction emphasises the political pressures facing even supposedly-neutral science advisors. A previous occupant of his post, for example, has admitted being pressed to withdraw an unguarded public comment that cuts in public funding for agricultural research were partly responsible for the spread of disease among UK cattle in the early 1990s.

But Beddington’s admission also highlights a growing concern among scientists and others (including experienced civil servants) that praise-worthy efforts to promote evidence-based policy are sometimes undermined by politicians seeking ‘policy-based evidence’ — research that can be used to justify politically-motivated action.

Bad science brings bad choices

The value of evidence-based policy is clear. Robust scientific findings, such as those linking human activity to climate change, or how animals are kept to disease outbreaks, can point convincingly to the need for immediate action.

Conversely, bad science used to make social choices inevitably brings bad decisions. The eugenics movements of the 1930s demonstrated that conclusively. So too have more recent claims by some scientists that HIV does not cause AIDS — an idea embraced for example by the former president of South Africa.

One reason for improving science communication skills within the developing world is to help people who provide decision-makers with scientific information to ensure that information is robust (for example, that it has been properly peer reviewed). This is important whether the information is direct scientific advice, or comes indirectly through the media.

But science journalists and other communicators must also be alert to the reverse situation, where politicians selectively refer to, and may even sponsor, research designed to endorse pre-determined action.

The UK Parliament explicitly raised this issue in a 2006 report, warning that ministers "should certainly not seek selectively to pick pieces of evidence which support an already agreed policy, or even commission research in order to produce a justification for policy: so-called ‘policy-based evidence making’" (see Scientific advice, risk and evidence based policy making[1.5MB]).

More scepticism please

Journalists have long guarded against the problems that can arise when financial interests are involved. Think, for example, of the automatic scepticism meeting industry-funded research that makes safety claims for controversial products, whether cigarettes or baby food.

We must apply the same scepticism to politicians who may be using science primarily to legitimise their actions — particularly when the quoted research has been paid for, either directly or indirectly, by a government whose actions are under fire.

And we should also be sceptical when a politician wheels out a friendly scientist — a fellow university student perhaps — to support action in a field outside the scientist’s professional knowledge or expertise. Respect for the authority embedded in the title "professor" should not mean automatic deference, nor does it excuse journalists and others from questioning whether the scientist speaks with authority based on relevant experience.

Sadly, deference often seems stronger in many developing countries than in the developed world, where the need to scrutinise scientists’ expertise is gaining acknowledgement.

Joint responsibilities

Scientists have a responsibility to help ensure that only sound science is used to inform policy-making. One of the roles of scientific academies, for example, should be to monitor government statements and complain, both privately and publicly, when these make inappropriate references to scientific research.

Journalists share this responsibility, but they need scientists help. Journalists can find separating appropriate from inappropriate science a challenge, given their frequent lack of technical knowledge and unfamiliarity with scientists’ procedures for making such judgements.

But what journalists can do is bring cases to public attention so democratic procedures can take corrective action.

All science journalists should be on watch against the danger of policy-based evidence. Indeed, all professional science communicators should be aware of the danger, and stand ready to expose it.

David Dickson
Director, SciDev.Net

See Letter to the editor.