The Indy’s Rape Statistics: Wrong Again!

I’ve lost count of the number of times I’ve had pick up The Independent over basic statistical errors and today they’ve made another absolute howler, albeit that at least part of the blame appears to lie with Labour’s Shadow Attorney General, Emily Thornberry:

Emily Thornberry, the shadow Attorney General, whose office uncovered the figures in the House of Commons Library, said: “Everyone with an interest in the protection of women is encouraging rape victims to come forward and it seems that they are starting to. It is therefore profoundly disappointing that we have not seen a surge in the number of decisions to prosecute. In fact, the proportion of rapes being sent for charges, prosecuted and convicted have been shrinking every year under this Government.

Last month the CPS trumpeted the fact that police referrals to them for rape had increased by 8.3 per cent to 5,850 in 2013/14. But the annual report into violence against women and girls omitted to explain that since reports to the police had soared from 16,357 to 20,725 last year – the number of cases taken on to prosecution should have been significantly higher.

The small numerical increase in referrals does not keep up with a huge surge in reports, meaning victims are proportionately more likely to have their case dropped by the police than they have been since records began.

NO.

Based on the most recent audited statistics from the Crown Prosecution Service, the average lead time for rape cases from the point at which an alleged offence occurred to the point at which charges are preferred against a suspect is just over 430 days.

The reason that the small numerical increase in referrals in 2013/14 has not kept up with the large surge in reports in the same year is not because there has been a fall in the number of referrals but because the lead time in rape cases means that the vast majority of last year’s referrals relate to alleged offences that were reported to the police in 2012/13, and some cases even earlier than that.

Right now, we don’t know what effect the surge in reported cases in 2013/14 will have on the CPS figures for referrals because the vast majority of those cases will not begin to appear in the CPS data until this year (2014/15), figure that will not be published by the CPS until (from memory) around January/February 2016.

Add in the fact that, at this stage, we have no idea at all what proportion, if any, of that “surge” in reported cases stems from people coming forward to report historical offences in the wake of the publicity surrounding Jimmy Savile, Rolf Harris, etc – and for what should be obvious reasons (i.e. lack of physical evidence, etc) historical offences are likely to take longer to investigate and may be much less likely to lead to charges than contemporary offences, not least because in some cases the alleged perpetrator may already be deceased – and it should be clear that it may be 2-3 years before we actually get anything approaching an accurate assessment of the impact that last year’s rise in reported cases will have on the number of cases the CPS takes forward to prosecution let alone the number of actual convictions those cases generate.

Although the figures themselves are correct, the interpretation placed on them isn’t because both Thornberry and the Indy have failed to take into account the lead time between offence and charging decision.

Sorry.

 

  • mattwardman

    One number I would like to get my head around is how much of that increase relates to Yewtree and similar.

    Reports to the police up by 4000+ in 2013-14.
    YewTree identified some 589 allegations (Wikipedia) up to 2012-2013

    eg will the latest Cliff Richard allegation appear in these numbers? Did the historic Saville allegations appear?

    Equally I’ve been seeing some commentary about falling “conviction rates” (allegation to conviction, presumably ignoring the time lags you mention). How much of *that* fall relates to historic allegations and the difficulty of prosecuting them.

    That’s before we even get into working out how allegations from 20-40 years ago should impact on current resources/services/practices etc, never mind the issue of unreliable memory and coroboration requirements stripped out of the system etc.

    Phew.