Policy-based evidence making

Policy-based evidence making is a term which has been coined to describe the way Government reviews in particular are constructed so as to give the outcome on which those commissioning the review have already settled. Department for Transport Research Report 30 is an example of this: by selecting a review team and terms of reference which exclude all views not known to be in sympathy, an "independent" review can be said to support the pre-defined policy in full.

Other examples of policy-based evidence making include the BMA's internal report used to justify a reversal of its stance on helmet compulsion in 2006. This report was written by a BMA staffer from material clearly hand-picked by the helmet lobby (most notably BeHIT). This is evident from the fact that several claims were included which are (a) provably false and (b) only known to have been stated by BeHIT. An example is the figure of 50 child cycling head injury fatalities per year, an "estimate based on under-reporting". The level of under-reporting of child deaths in the UK is as close to zero as makes no odds, of course.

The term is also used by some to describe what they see as the heterogeneity of reports by "the establishment". In some cases, of course, the establishment will all agree for the simple reason that they are right; the more vocal and isolated the minority, the more likely this is to be represented as a conspiracy (see Wikipedia and Wesselygate - the CFS activists attack the mainstream research on CFS as policy-based evidence making).

As ever, the use of the term is not indicative that either side is right. Unless it's me castigating the Government for yet another report designed to prove that what cyclists really need is to ride on the pavement wearing a plastic hat, of course, in which case naturally I am right and they are reverse-engineering their evidence from the conclusions. So there.

Moral: Think for yourself. And don't forget the bit about thinking, either.