Street lights vs crime: redux

UPDATE 04/04/12: Lights off, but no more crime in Hertfordshire, says Police Authority report

Following on from my last blog post (“Do brighter street lights make you safer from crime?“), a Guardian “Comment is Free” editorial was published on 26 December 2011 under the title “In praise of leaving the lights on“. This article came after the widely reported words of Stella Creasy MP who called for a halt to street-light switch-offs until studies had been carried out into the effect on crime of reduced light at night.

In my previous post I cited several US studies that have shown that switching street lights off at night does not result in an increase in crime, and indeed in many cases brighter street lights resulted in an increase in crime.

I have subsequently read several British studies that also support this view, such as the 1991 Home Office Crime Prevention Unit Papers No. 28 and 29, entitled respectively “The Influence of Street Lighting on Crime and the Fear of Crime” (pdf) and “The Effect of Street Lighting on Crime and Fear: A Review” (pdf). These reports state that

no evidence could be found to support the hypothesis that improved street lighting reduces reported crime (Paper No. 28)

and

improvements to street lighting can help to reduce the public’s fear of crime, but that they make less of a difference to the prevailing level of crime than many people would expect (Paper No. 29)

However the Guardian editorial cites

a 2002 study by the Home Office [Home Office Research Study 251, HORS251] [which] found that “improved street lighting led to… an overall reduction in recorded crime of 20%” (pdf).

This seems to contradict the earlier Home Office studies. Might things have changed in the intervening 11 years? It appears not. The author of the 1991 Paper No. 29, Dr Malcolm Ramsay, a Senior Research Officer in the Crime Prevention Unit of the Home Office, wrote a 2004 paper in the British Journal of Criminology criticising the methods used in the HORS251 study.

The arguments in Dr Ramsay’s research paper are predominantly of a statistical nature. The introduction states that

the [HORS251] review at first sight appears to be an appropriate statistical synthesis of all studies on street lighting and crime across the world. However on close examination, the statistical claims and methods are unfounded.

According to Dr Ramsay, not only does the 2002 HORS251 report

use methods that ignore the large variation (known as “overdispersion”) in the data and implicitly assume that crimes are independent events, which is implausible in the extreme

but it also

is not comparing like with like, for the individual studies, in general. This is because brighter street lighting is applied to more crime ridden areas and the comparison areas are less crime-ridden and this will lead to an effect known as regression to the mean.

Further problems are apparent with HORS251. For example small studies are excluded for no good reason. Dr Ramsay’s paper concludes:

Crime reduction is frequently presented as a potent argument for increased lighting – here it has been shown that there is no scientific basis for this claim.

I for one welcome any new data on this issue, but in the light of the above criticisms of HORS251 it is unfortunate that the Guardian used it in a comment piece in praise of leaving the lights on.

Thanks (again) to @darkskyscott of the International Dark-sky Association for links to these reports, and to the Campaign for Dark Skies for the information in their “Lighting and Crime” section.