This forum is about wrong numbers in science, politics and the media. It respects good science and good English.
I thought number watchers might appreciate getting back to basics with an instance of a patently wrong number I spied being bandied about in the national press.
The BBC News website has recently published the following article on a draft Commons bill which is intended to give various branches of government new powers for electronic snooping.
Theresa May is promoting the idea that the bill is aimed at "saving lives". And then, the following sentence appears, unquoted, unsourced, unsubstantiated and not elaborated upon:
"Of the 30,000 estimated cases last year where the police made an urgent request for communications data, between 25% and 40% of them resulted in lives being saved."
I found this statement quite alarming. Presumably they can only be referring to the prevention of intentional deaths, and not accidents or deaths from natural causes. In the latest year for which statistics were available, 2009, there were 722 murders in the United Kingdom. There is a fluctuation, year to year, but they seem to be significantly down from a high of 1220 in 2002 (although whether this is due to there being generally less violence, or great strides forward in trauma medicine, is perhaps an open question). In the same year, there were 4245 suicides, for a total of under 5000 intentional deaths.
25-40 percent of 30,000 is 7500-12,000. 7500-12,000 incidents of at least one life being saved? If we assume that the figures for intentional deaths were similar in 2011 to what they were in 2009, then the suggestion is that the security services' and the police's requests for digital data snooping are alone responsible for reducing the rate of intentional deaths in Britain by a factor of three or more.
I do believe that violent criminality is down-played in this country, but maybe the government and the BBC would like to try and pull the other one? How could even a basically numerate journalist regurgitate such a statement unchallenged, and how could an editor allow it to be printed?
I think most people would interpret that sentence as meaning that several thousand lives are being saved every year by communication data requests made by the police. But really you would need to know the total number of these communication data requests that are being made on average as part of a specific police investigation.
The following document throws a bit of light on it:
The relevant bit looks like the following paragraphs:
"Home Office published statistics show that there are in excess of 4 million crimes reported annually and around 1.4 million of these will fall into the serious crime category. Criminals will often use many communication devices at any one time and will regularly change them. A significant murder or organised crime investigation can involve up to 500 communications data requests or more often mainly subscriber checks.
Communications data helps focus an investigation by identifying possible suspects, and is also critical in confirming alibis, ruling people out of further enquiries, and finding witnesses. Many tens of thousands of communications data requests are made every year in urgent threat to life situations: e.g. to find a vulnerable or missing person or in kidnap situations."
From the above it looks like there could be as many as 500 communication data requests per investigation by the police, so the 7500 to 12000 figure would actually correspond to a much smaller number of lives being saved.