Waving the white flag ?

Has Whitehall finally admitted defeat ?

Something with the long-winded title of “Consultation on improvements to Ministry of Justice statistics” was published in Feb 2011.

It is an acceptance, albeit in a roundabout way, of the criticisms  I and others have levelled at the way the Home Office has hitherto compiled criminal statistics – especially for rape.

Conviction rates for rape have always been devised in a way that would be unacceptable for any other crime. This anomaly is thankfully, at last, acknowledged and is to be killed off.

The publication written by Jil Matheson, the National Statistician also acknowledges on page 22, that false allegations of rape are a fraught subject area in need of greater examination and she recommends:

  • “ . . . .   that the Ministry of Justice commissions and publishes an independent research report to study the frequency of false allegations of rape compared with other offences and the nature of such allegations.”

Just how independent and how unfettered that groundbreaking  ‘research’ will prove to be is open to question. The Ministry of Justice had no information on false allegations in 2010 other than those I provided for the Nov 2010 review. No mention of its existence appears in this Feb 2011 paper.

The report comes in several smaller summaries and this one http://www.justice.gov.uk/downloads/consultations/moj-stats_consultation-a.pdf is a good place to start (the full list can be found at: http://www.justice.gov.uk/consultations/565.htm ).

The second really interesting part begins at page 26, Para 2.3 “End-to-end measurement and downgrading.” Here the thinking of the Min of Justice is revealed and their proposed amendments laid out.

The following, although abridged, is taken directly from the “Consultation on improvements to Ministry of Justice statistics” (pages 26 – 28):

  • Conviction  rates

The Stern Review recommended that the Min of Justice should work with the National Statistician to explore fully the issue of conviction rates in rape cases. As rape cases cannot be considered in isolation we have considered this issue alongside the review of Barriers to Trust in Crime Statistics.

During the Min of Justice’s investigations they have found at least four different definitions of a rape conviction rate being used which are as follows:

  •  A.  Convictions for rape and recorded crime

One method, used commonly in discussions of the Criminal Justice System response to rape, shows the number of people convicted of rape as a proportion of all rapes recorded. This is often incorrectly referred to as a conviction rate.

Thus while useful, this method is misleading in terms of presenting evidence on convictions for rape.

The Min of Justice will continue to make available the necessary data to allow users to construct this measure if they wish, but advises thatresults are referred to as a conviction to crime ratio.

  • B.  Convictions for rape and people prosecuted

Official convictions for all offence types produced by the Min of Justice are, strictly speaking, also not rates but ratios.

They are ratios of convictions for a principal offence (e.g. rape) over one year to prosecutions for that same offence. This is because offenders convicted in a reporting year are not always the same people who were prosecuted in that year as investigations and trials span more than one reporting year.

Common terminology refers to the ratio as a rate but the Min of Justice’s has from today begun referring to these correctly as conviction ratios.

[ NB statistical rates are often found expressed as per 1,000 persons or per 100,000 of either offences or persons. This allows for direct and accurate compositions to other countries with very different population sizes – RW ]

  • C.  Convictions for any offence and people prosecuted

An alternative method to the standard used by the Min of Justice presents the number of defendants convicted of any offence as a proportion of those prosecuted for rape.

The Stern Review published in 2010 drew attention to this more inclusive method of presenting conviction data which is important because it reflects instances of downgrading. For example, if someone is prosecuted for rape but convicted of a lesser offence, this conviction would be included.

  • D.  Convictions for rape and charges heard by jury

Finally, Cheryl Thomas’s paper (2010) presented a different method of calculating conviction rates [1] (see also Appendix A).

This considers the number of convictions against charges of rape heard by a jury following a ‘not guilty’ plea, rather than the number of people prosecuted.

In considering charges and not people prosecuted, this measure differs from the Min of Justice method, as illustrated by the example below:

  • Ten people are prosecuted for a total of 30 charges of rape. Juries pass a guilty verdict against 18 of these charges.
  • This translates into a 60% jury conviction rate based on charges heard but, if only four of the ten people prosecuted were found guilty of rape, Min of Justice statistics, which are based on people prosecuted, would report a conviction rate of 40%.

Cynics might say that in looking at 70,000  trials of offences committed by black and minority ethnic men but tried by “all white” juries an in-built bias was being sought perhaps for political considerations or to undermine the process for future changes.  Prof Cheryl Thomas’s survey included 4,000 jury rape trials. The reader has to be cautioned against the media and academia who these days are casual about using the word ‘rape’ which can mean a variety of offences from rape to less serious sexual offences.

[ Observation – Neither the Home Office in the past nor – it would appear from the above – the Min of Justice of today, have got their heads around and addressed the issue of “serial” offenders and multiple offenders (as opposed to the re-offending criminal or career criminal). On page 26 can be found references to those convictions involving multiple defendants and multiple offences but not a singular defendant facing multiple and identical charges.]

  • Moving forward in presenting conviction data  

It is the Min of Justice‘s view that both the conviction to crime ratio and also the conviction rate used by Cheryl Thomas have added to the knowledge and understanding of the particular circumstances of rape. However, they have not been replicated across all crime types and do not lend themselves easily to this wider approach.

The Min of Justice statisticians believe it is essential that all measures we publish should be published consistently for all offences. It is possible to publish the other two conviction rates for all offences:

  • The number of people prosecuted for Offence A that are convicted of Offence A and what percentage of all prosecutions for Offence A this comprises: the ‘within offence’ conviction rate;
  • The number of people prosecuted for Offence A that are convicted of any criminal offence and what percentage of all prosecutions for Offence A, this comprises: the ‘all offence’ conviction rate.

Neither measure on its own provides a complete picture of conviction rates as the first measure can under-estimate conviction rates while the second can overstate them.

We also find in these pages the dilemma faced by the CPS and police, i.e. rather than supporting a case for prosecution against the original charge, the evidence better warrants prosecution for a less serious offence, e.g. a reported crime of rape may receive a charge of serious sexual assault as the evidence was insufficient to prosecute for rape. This process is known as ‘downgrading’.

The final part of this section asks:

  • Q4. Do you agree that conviction rates should be measured using two measures: a ‘within offence’ conviction rate which shows the proportion of people prosecuted for an offence who are found guilty of that offence; and an ‘all offence’ conviction rate which shows the proportion of people prosecuted for an offence and found guilty of any offence ?
  • Q5. Do you agree that until the data linking necessary for full conviction rates to be published is completed, that we continue to publish the conviction ratio ?

Appendix A



Professor of Judicial Studies
Director, UCL Jury Project
Co-Director, UCL Judicial Institute

The recently published follow-up study, Are Juries Fair? (2010), tackles sensitive and controversial issues about the fairness of jury decision-making for the first time in this country. It examines whether all-White juries discriminate against Black and minority ethnic defendants, whether juries rarely convict on certain offences or at certain courts, whether jurors understand legal directions, are aware of media coverage of their cases or look for information on the internet about their cases during trial. The empirical study involved over 1000 serving jurors in three areas of the country and over 68,000 jury verdicts across all Crown Courts in England and Wales {“As well as an academic, she is also a documentary filmmaker and has produced programmes for the BBC, Channel 4, ITV, Discovery and PBS” –  and worked for French government – how do you get that well-connected ?}.

This extract may help put her work into context:-

All-white juries do not discriminate

By Paul Sullivan, from insidetime issue March 2010


Juries overall appear ‘efficient and effective’ and convict on almost two thirds of all charges presented to them, according to a ground-breaking study by the Ministry of Justice into the secrets of the jury room. Paul Sullivan reports The study, in which 70,000 trials were analysed, conducted by Professor Cheryl Thomas of University College London, found that contrary to popular belief and previous government reports, juries actually convict more often than they acquit in rape cases. A previous Home Office study claiming that jury acquittals were more common than convictions was based on only 181 verdicts in a few courts. The current study was based on more than 4,000 jury rape verdicts between 2006 and 2008.

– – snip – –

[1]  Cheryl Thomas, “Are Juries Fair ?” London,  Ministry of Justice (2010).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: