Martin Wainwright
Friday November 21, 2003
The Guardian
Over-anxious Britons are placing a blind, almost medieval, faith in brighter streetlamps and security lighting as crime deterrents, according to a statistical analysis which raises questions about Home Office research.
Government advice that surveys in Britain and the US show better lighting to have “no negative effects and demonstrable benefits for law-biding citizens” is flawed and unjustified, according to a study to be published in the British Journal of Criminology.
The paper follows arguments at the Home Office over a summary of research on street lighting and its effects on crime. Independent criticism of the summary’s use of statistics was initially expected to be incorporated as a “serious health warning” but appeared only as a brief addendum.
The scientist behind the criticisms, Paul Marchant, a statistician at Leeds Metropolitan University, said yesterday that the conclusions of the paper – The Effects of Improved Street Lighting on Crime; Home Office Research Study 251 – were unfounded and based on statistical mistakes.
“They have all the trappings of the right sort of official information, but they contain major errors. Not only does the paper fail to demonstrate that improved street lighting decreases crime, its figures could equally well show that it increases it.”
The row comes at a time of growing concern about light pollution, with a warning about its effects, particularly on astronomy, sounded last month in a report by the Commons select committee on science and technology.
Dr Marchant said that pressure for supposedly “crime-deterring” extra street lighting was worsening the problem, abetted by householders’ security lights whose effectiveness was also a matter of faith.
“It may not be too fanciful to make comparisons with our medieval ancestors’ measures against witchcraft – tokens hung over the door and so forth,” he said. “We know the ‘witches’ are out there, and we turn to deterrents which we think will scare them off. But we do not know. We do not have the evidence.
“Meanwhile light pollution has substantial environmental consequences. If any decision is taken to increase lighting, it needs to be taken on the best possible evidence.”
The Home Office said officials had been made aware of the issues and had asked the report’s authors to look at them in detail. A spokesman said: “The need for further analysis of statistical variance calculations was accepted and an addendum has been added to reflect this change. But this did not substantially affect the report’s conclusions.”