coyle final paper by pVErGYG


									           Statistics, Development and Human Rights

              Session I-Pa 4c

How Does Statistical Information Aid the
 Development of Appropriate Policies in

               Andrew COYLE

                     Montreux, 4. – 8. 9. 2000
                               Statistics, Development and Human Rights

How Does Statistical Information Aid the Development of
Appropriate Policies in Prisons?
Andrew COYLE
International Centre for Prison Studies, King’s College, University of London
75-79 York Road
London SE1 7AW, United Kingdom
T. + 44 (0)20 7401 2559 F. + 44 (0)20 7401 2577


     How Does Statistical Information Aid the Development of Appropriate Policies in Prisons?

       Despite the increasing openness which characterises prison administrations in some
countries today, prisons remain essentially secret places. To a certain extent, the very nature of the
prison demands that this should be so. Yet it is important that those who are responsible for
prisons should remember that they carry out their work on behalf of society. At the same time,
members of the public have an obligation to make sure that they approve of what is done in their
name behind prison walls. The existence of good statistical information about prisoners is an
essential prerequisite to ensuring that our prisons are as just, decent, and humane as possible.
       This article begins by examining the crucial role that the availability of comprehensive
statistical information plays in prison management and in the strategic development of better prison
practices. It goes on to consider the manner in which external monitors employ statistics to improve
prisons. The article concludes with suggestions about areas in which current statistical practices
could be improved.


      Comment l’information statistique contribue-t-elle au développement de politiques
pénitentiaires appropriées ?

      En dépit de l’ouverture croissante qui caractérise de nos jours les administrations
pénitentiaires de certains pays, les prisons restent des lieux essentiellement secrets. Dans une
certaine mesure, cependant, la véritable nature de la prison le requiert. Il est néanmoins important
que les responsables des prisons gardent bien à l’esprit qu’ils exécutent leur travail au nom de la
société. Parallèlement, le public a l’obligation de s’assurer qu’il approuve ce qui est fait en son
nom derrière les murs des prisons. L’existence de données statistiques fiables sur les prisonniers
constitue un prérequis essentiel permettant de garantir que nos prisons sont les plus justes, les plus
décentes et les plus humaines possible.
      Cet article commence par examiner le rôle crucial de la disponibilité de données statistiques
complètes dans la gestion des prisons et dans le développement stratégique de meilleures pratiques
pénitentiaires. Il examine également la façon dont les contrôleurs externes exploitent les

                                         Montreux, 4. – 8. 9. 2000

                                Statistics, Development and Human Rights

statistiques en vue d’améliorer les prisons. L’article se termine sur des suggestions concernant les
domaines dans lesquels les pratiques statistiques actuelles pourraient être améliorées.

1. Introduction

   The challenge of treating prisoners in a manner which is decent, humane and just is one which
those of us who work in prisons struggle with every day of our professional lives. Throughout 25
years of public service as a prison governor in the United Kingdom I was continuously involved in
attempting to improve conditions in prisons in Scotland and England by managing them according
to national and international standards. As a result of that experience, I have come to realise that the
possession of good statistical information about prisoners is an essential prerequisite to running
prisons in a just, decent, and humane manner.
   Permit me to give you two examples from my own experience. (Coyle, 1994). In the late 1980’s
Peterhead Prison in Scotland held the prisoners who had been assessed as the most disruptive and
dangerous in the Scottish prison system. All of them had been involved in riots, in taking hostages
or in escape attempts from other prisons. When I went there as Director in 1988 I walked into a
world where there was continual violent confrontation between staff and prisoners. The normal
uniform for staff each day included a riot helmet, body armour, and perspex shields. Prisoners were
locked up in isolation for 23 hours each day; a number of them wore only blankets and had covered
themselves in their own excrement. When we set about the difficult task of changing this coercive
environment we concentrated on the key aspect of the human relationship between staff and
prisoners. But this was not in itself sufficient. We also had to collect and analyse all available
information. Who were these men who had been assessed as the most dangerous and disruptive in
the system? What were the facts on which those assessments had been based? How much of the
information collected about them was based on fact and how much on suspicion? What did we have
to do to make them less dangerous and disruptive in future?
   My second example comes from Brixton Prison in London, where I became Director in 1991.
Brixton was built in 1819 and is the oldest prison in London. When I took command of the prison I
found that one large wing held 300 mentally disturbed prisoners, almost all of whom were pre-trial
and had been sent to prison for psychiatric reports. The atmosphere in that wing was of a medieval
madhouse. All day and all night there was continual shouting, wailing and banging, and an all-
pervading stench of stale food, urine and excrement. The prison was being used quite
inappropriately for men who should have been in the care of the health services rather than the
prison service.
   As we set about making radical changes, we found once again that we had to collect and analyse
whole sets of information. We were helped in our efforts to change the system by two studies which
had just been published in England. The first had found that 37% of convicted men and 56% of
convicted women in prison in England and Wales suffered from some form of psychiatric disorder.
(Gunn et al, 1991). A parallel study of remand prisoners had found that “remands in custody of the
mentally disordered were not carried out because of the nature of their offences, but because of their
need for social and psychiatric help”. (Dell et al, 1991). This study also discovered that a person
who was remanded in prison for psychiatric assessment was likely to spend longer in prison
awaiting trial than ordinary remand prisoners. These and other pieces of evidence were extremely
important for our discussions with other criminal justice and health agencies about why so many
mentally disordered people were locked up in prison.
   Many prison administrators in Europe and elsewhere in the world are faced with these or similar
challenges. The purpose of this paper is to examine how people who spend their working lives
dealing with prisoners every day use statistical information to introduce humanity into the prison

                                          Montreux, 4. – 8. 9. 2000

                                Statistics, Development and Human Rights

    The article begins by examining the crucial role that the availability of comprehensive statistical
information plays in prison management and in the strategic development of better prison practices.
It goes on to consider the manner in which external monitors employ statistics to improve prisons.
The article concludes with suggestions about areas in which current statistical practices could be

2. Statistical Information and the Day to Day Management of Prisons

   What kind of information do prison administrators need for the proper daily management of
prisons? As always in this field, it is helpful to begin this discussion with reference to the
international human rights instruments that govern record keeping in prisons.
   Both Rule 7 of the United Nations Minimum Standards for the Treatment of Prisoners and Rule
8 of the European Prison Rules require that in every place where persons are imprisoned, a
registration book must contain information concerning each prisoner’s identity, the reasons for
commitment to prison and the dates of admission and release. These instruments also require prison
authorities to record details of each prisoner’s criminal offence as well as his or her physical and
mental capacities.
   These requirements represent the bare minimum of information necessary to manage a prison.
The management of a humane prison requires much more. In order to develop and maintain good
prison practices, prison administrators must have complete and accurate information about all
aspects of prison life including the race and gender of prisoners in their care, all health information
and details about daily regimes.

      Register of Prisoners

   Compiling an accurate register of prisoners is an essential first step to maintaining security and
control in the prison. Without this information, prison administrators could not perform their most
important duties: keeping untried prisoners separate from convicted prisoners, keeping young
prisoners separated from adults, keeping women separate from men and keeping petty offenders
separated from prisoners who have committed serious offences.
   Keeping accurate records of receptions into prisons can also help prison managers to plan for the
future. Statistics from past years or months can assist in predicting the number of prisoners likely
to be admitted at certain points. With this knowledge, reception staff can then make sure that
essential items such as food, clothing and bedding are on hand.
   The first piece of information needed for every new prisoner is the legal warrant for his or her
detention. This warrant should state the date of first detention and also the date on which the
warrant expires. The prison director has no right to hold any prisoner for whom there is no warrant
executed by the proper legal authorities. Information about the length of time that individual
prisoners will be in custody will also be useful to the authorities in planning regimes which are
appropriate for the needs of these prisoners. Information about the date of release means that the
authorities can help prisoners to prepare for this in a way which will facilitate their return to civil
   I remember having a discussion a number of years ago with a senior prison official in a
developing country about what was needed to drive forward the process of prison reform in that
country. When asked what his first priority would be, the official replied that it would be to have a
computer in every prison linked to headquarters. This seemed rather a luxury in a system which had
difficulty in providing every prisoner with the basic necessities of life. The official had a logical
answer. It was that many of the problems of the prison system in that country were caused by high
levels of overcrowding. The majority of prisoners were awaiting trial. Because of poor record
keeping many of them missed their due date to appear in court and an equally large number were

                                          Montreux, 4. – 8. 9. 2000

                                Statistics, Development and Human Rights

held beyond their due date of release. The official argued that a proper computer system would
allow his officials to make sure that everyone arrived in court on the proper day and that everyone
was released at the proper time. His case was that, in addition to meeting the demands of justice,
this would result in a significant reduction in the pressure on the prison system and an improvement
in conditions for prisoners.
   The failure to keep a complete register of prisoners can have dire consequences, as is illustrated
by an incident that ocurred last year in West Bengal, India. Prison officials there discovered a man
who spent 37 years as an inmate at the state Presidency Jail without ever coming to trial. (Press
Trust of India Limited, 7/12/99). The prisoner had been accused of murder in 1962, but his trial
was postponed after he was found mentally unsound. Afterwards, prison officials simply forgot
about him. It was only when the National Human Rights Commission brought his case to the
attention of local authorities in December 1999 that he was transferred from the jail to a psychiatric
hospital. He had been imprisoned since January 28, 1962.

      Health Care in Prisons

    At the beginning of this article I made specific reference to my own experience with mentally
disordered prisoners. Prison directors the world over could tell of similar experiences.
Consideration of health matters in prison goes far beyond mental health. The international human
rights instruments make it clear that prisoners are entitled to medical treatment of the same quality
as that received by the rest of the community. For example, Principle 9 of the UN Basic Principles
for the Treatment of Prisoners provides that “Prisoners shall have access to the health services
available in the country without discrimination on the grounds of their legal situation”. Without
complete information about the health of prisoners, prison staff cannot implement appropriate
policies for treating illness or controlling communicable disease.
    In many countries prisons have levels of overcrowding which are simply unbelievable, with
prisoners spending long periods locked in poorly ventilated cells with hardly any natural light and
little access to sanitation. In such environments it is no surprise that disease and ill-health flourish.
    The efforts of statisticians in collecting information about health in prisons have resulted in great
publicity about the spread of diseases like tuberculosis, hepatitis and AIDS. In Russia 96,000
prisoners, almost 10% of the total, are estimated to have active TB. (GUIN Report, 1999). In
Puerto Rico, where there are eleven and a half thousand prisoners, 94% have Hepatitis C. This
figure sounds unbelievable but it comes from a study by the Centers for Disease Control in Atlanta
published in 1999. In Brazil in some prisons one in five of the prisoners has HIV infection. They
do not have enough doctors so prisoners are given the job of caring for those with AIDS. (Amnesty
International, 1999). In the United States more than one quarter of a million of their nearly 2
million prisoners have a mental illness, according to a report released by the Department of Justice.
(U.S. Department of Justice, Bureau of Justice Statistics, 1999).
    Publication of these facts about public health epidemics in prisons enables prison administrators
to appeal to government ministers for more resources.


   In almost all countries ethnic minority groups are over–represented in prisons. In some countries
it will be black people, in others Roma or gypsies, in others aboriginals or other indigenous groups.
Prison managers have to be constantly aware of the number of minorities in prison and to be
vigilant about the potential for disparities in treatment. England and Wales provides a good case
study. The number of men and women from ethnic minority groups in English prisons is out of all
proportion to their numbers in the general population. In 1998, black people made up 11.8 per cent

                                          Montreux, 4. – 8. 9. 2000

                                Statistics, Development and Human Rights

of the prison population compared with under 2 per cent of the general population. (Research
Development and Statistics Directorate, Home Office, 1999). In other words, for every 100,000
persons in the general population there were 185 white persons in prison, 160 people of Asian
descent and 1,245 black people. (Prison Statistics, England and Wales, 1998). These figures need to
be pointed out to police, prosecutors and judges who are responsible for putting people through a
criminal justice process which leads to imprisonment.
   But the problem of race does not end at the prison gate. What goes on inside prisons must also be
scrutinised. In the Prison Service for England and Wales, this important task falls to the Race
Relations Committee which exists in every prison. This is a cross-professional group whose
purpose is to address racially motivated incidents and ensure that all prisoners have equal
opportunities to use facilities and services. It is the task of the Race Relations Officer at each prison
to collect statistics and to report back to the Committee.
   There are several sets of data which help this committee to find out whether there is racial
disparity in the treatment of prisoners. One is the racial break down of the number of prisoners who
are subject to disciplinary sanctions. In England and Wales not only are minorities over-represented
in the general population; they are also over-represented in punishment and segregation units. The
United Kingdom is not alone in this respect. In the United States, nearly 15% of black federal
prisoners are held in maximum security segregation, as compared to 7.9% of white federal
prisoners. (U.S. Department of Justice, 1998).
   Another way of assessing racial equality or disparity is to examine how the most coveted work
assignments are made. Working in the kitchen, for example, is a popular assignment in many
prisons. Being a wing orderly is another popular placement. In contrast to proportions in the
punishment unit, it may be that majority ethnic groupings are over-represented in these areas. As
with many other aspects of prison life, it is often difficult to be precise about issues such as racial
discrimination. Measurements such as these can go a long way towards providing indicators which
can be objectively verifiable.
   The importance of the work that statisticians do is not only in collecting such information but
also in analysing it. Prison managers need statisticians to tell them, for example, why it is that
prisoners from ethnic minorities are more likely to be housed in punishment cells than those from
majority groupings. Is it because they commit more disciplinary offences; is it because staff are
more likely to identify them when they do; or is it because they are more likely to receive harsh
punishment? These are just some of the issues in which the work of statisticians can be invaluable
to good prison management.

      Prison Regimes

   There are many other issues affecting the way prisons are managed and their regimes developed
which need to be subjected to objective measurement. There are a number of key indicators as to
the health of a prison regime and the relationship between staff and prisoners.

   One of these is the use which is made of the formal disciplinary process. Why is it that in one
    prison the segregation unit is always full whereas in a neighbouring prison of similar size and
    with a similar catchment, very few prisoners are sent to segregation?
   Another important measure can be the number and type of grievances filed by prisoners. This is
    a good example of the care which is needed when analysing prison statistics. A large number of
    grievances from one prison may indicate that something is amiss. However, a small number of
    grievances will not necessarily indicate that all is well. Instead, it may be that complaints are
    suppressed or not allowed. In this context, as with disciplinary figures, it is also important to
    analyse the source of complaint. Are many of them being made by the same individuals or by

                                          Montreux, 4. – 8. 9. 2000

                                Statistics, Development and Human Rights

    prisoners from a particular wing or living unit? Are they against staff in general or are the
    complaints frequently against the same member of staff?

   Prison directors are likely to pay close attention to the number and seriousness of assaults by
    staff on prisoners, by prisoners on staff and by prisoners on prisoners. Another important
    indicator of the safety of the prison may be the frequency with which prisoners deliberately
    injure themselves.

   Directors will also wish to measure the amount of time prisoners spend out of cells each day.
    How much access do they have to fresh air and showers? How often can they make contact with
    their families and under what conditions? What about the observance of religion and access to
    their clergy?

   Simply measuring time out of the cell or the accommodation unit is not in itself enough. There
    should also be an objective formula for measuring what prisoners do when they are out of their
    cells. Do they simply stand around doing little or are they engaged in activities such as useful
    work, education or sport?


   Prison management, of course, is not only about prisoners; it is also about staff. In assessing how
well a prison is managed, the directors need to have a wide range of statistics concerned with
personnel. How many, in the first instance, in ratio to the prisoners? How many staff work directly
with prisoners and how many are in “backroom” jobs? What is the ratio of security staff to staff
working on regimes and programmes? How many staff come from ethnic minorities and how are
they distributed among the different grades?
   As with prisoners, there are a number of indicators about whether staff feel safe as they go about
their work and whether they have job satisfaction. One of the best of these is the level of sick
absence. Staff who are afraid or dispirited are more likely to have a series of minor illnesses and to
take regular days off work.

3. Using Statistical Information to Re-think Prison Policies

   The work of statisticians can also inform the wider debate about the use of imprisonment and can
facilitate international comparisons.

    Encouraging Public Accountability For What Goes on in Our Prisons

   Despite the increasing openness which characterises prison administrations in some countries
today, prisons remain essentially secret places. It can be argued that prisons are the last great
secretive institutions in a democratic society; their high walls and fences are built to keep the public
outside just as much as to keep prisoners inside. To a certain extent, the very nature of the prison
demands that this should be so. Yet that secrecy brings with it the need for public accountability
because prisons are part of the fabric of modern society every bit as much as other institutions such
as schools or hospitals. It is important that those who are responsible for prisons should remember
that they carry out their work on behalf of society. This means that members of the public have the
right to know how prisons are being managed. Indeed, it can be argued that they have an obligation
to make sure that they approve of what is done in their name behind prison walls. Statistical
information can help to make prison administrators publicly accountable for what they do.

                                          Montreux, 4. – 8. 9. 2000

                                Statistics, Development and Human Rights

      Encouraging Rational Debate About the Use of Prison

   In most countries nowadays there is great public concern about law and order and the need for
protection against criminals. The public’s concern is often fuelled by certain elements of the media
which paint a picture of a society in which law and order has broken down and the criminal justice
agencies are failing to protect the public. There is a belief that, if only more people went to prison
and if only prison regimes were harsher, society would be a safer place. This belief is reflected in
the unprecedented increase in prison populations in a number of European countries.
   Objective data can be used to introduce sanity and rationality into the debate about what our
response to crime should be. There is little, if any, solidly based research evidence of a direct
relation between higher rates of imprisonment and reduced crime levels. Of course, it is possible if a
high enough proportion of all people, particularly young men, is locked up that there will be a
reduction in crime. But that begs all sorts of questions about the value systems of a society which
chooses to go down this path. Those of us who work in the criminal justice system have an
obligation to explain the reality to the public and to the media.
   The research that criminal justice statisticians produce can enable prison administrators to gain
support for policies that will encourage a more balanced use of imprisonment. One can point to two
types of programmes in this field that depend for their existence on the collection and publication of
statistical evaluation. They are alternative sentencing programmes and conditional release

   Alternatives to prison are based on the idea of social reintegration for offenders who do not pose
    an immediate threat to the safety of civil society. They can include community service,
    suspended sentences, fines, curfew, compensation to victims or a combination of these
    sentences. Research has shown that if alternative sentencing programmes are to be successful,
    the public must be well informed about them and support them. (Stern, 1999).

   Conditional release programmes enable prisoners in many countries to leave prison for short
    periods of time to arrange housing and seek employment in preparation for release. Early
    release schemes sometimes attract a great deal of criticism from opposition politicians. States
    that use conditional release have to recognise the public’s concern that some prisoners released
    early may abuse their privileges and use early release as an opportunity to commit further
    crimes. This public concern is perfectly understandable and has to be answered. Studies from
    Canada and from France have shown the high success rates of these programmes and have been
    instrumental in re-assuring the public. (Larocque, 1998; Burricand and Hatal, 1997).

   More research evaluating the effectiveness of these and other programmes continues to be

      Making International Comparisons

   Statistics can permit a state’s prison service to compare itself with other countries. This is a
particularly useful exercise when we compare countries with similar economic trends and crime
trends, but very different criminal justice policies. For example, the Canadian population growth
trends and crime trends are remarkably similar to those in the United States. But the increase in the
prison population in the United States, up almost 30% in the last four years, has been nearly triple
the 10% growth experienced in Canada. (Correctional Service of Canada, 1998).

                                          Montreux, 4. – 8. 9. 2000

                                Statistics, Development and Human Rights

   Within Europe there are interesting variations. For example, there are more prisoners serving life
sentences in England and Wales than in the rest of Western European countries combined. (Council
of Europe Annual Penal Statistics, 1998; Home Office, Prison Statistics, England and Wales, 1998).
Of the 29 member states of the Council of Europe for which figures are available, only Romania,
Estonia and Lithuania have higher rates of imprisonment than England and Wales for people under
the age of 21. (Social Exclusion Unit, Policy Action Team 12, 2000). It would be useful to have an
analysis of why the number of persons entering prison in France has remained stable in recent years
and now shows some signs of reduction in contrast to the trend in a number of other countries.
(Burricand and Hatal, 1997).
   This might be the point at which to make a plea that statistics on prison subjects should be
collated in a uniform manner. At present, differences in the way in which figures are gathered often
make international comparisons difficult or impossible. Each year the Council of Europe reports on
prison populations, sentencing, and crime trends in member states and publishes its findings.
(Council of Europe Annual Penal Statistics, SPACE, 1996-1998). Unfortunately countries use
different methods to record their prison populations. For example, some countries do not include
juveniles or remand prisoners in their prison population statistics. At the International Centre for
Prison Studies, we recently did some work on comparative lengths of sentence in different
countries. Our task was made much more difficult by the different recording mechanisms used. In
England and Wales we count the number of prisoners serving sentences of more than four years,
whereas many other European countries count the number of prisoners serving sentences of more
than three years. In similar vein, the Council of Europe figures on conditional release in member
states do not specify whether they refer to temporary or permanent conditional release.

4. Measuring the Performance of Prisons Through Independent Inspection

  Returning to the international instruments for the management of prisons, Rule 55 of the United
Nations Standard Minimum Rules requires that:

       There shall be a regular inspection of penal instruments and services by qualified and
       experienced inspectors appointed by a competent authority. Their task shall be in particular
       to ensure that these institutions are administered in accordance with existing laws and
       regulations and with a view to bringing about the objectives of penal correctional services.

    In some of the countries, the inspection is the responsibility of specially appointed judges, whose
task is to come into prison on a regular basis to make sure that all rules are being observed and to
listen to prisoners’ complaints. In other countries it is the task of the local procurator. The
experience in most countries which have such a system is that this inspection is at best irregular and
superficial. The best systems of inspection are those that allow inspectors to gather information,
including statistical data, about each prison and to publish it.
    A good example of what can be achieved by an independent inspectorate is the work done by
Her Majesty’s Chief Inspector of Prisons for England and Wales. The Chief Inspector in England
has become a person of considerable influence in relation to the supervision of prisons. This is due
in large measure to the fact that the report which he makes to the Home Secretary following each
prison inspection is published along with the Home Secretary’s response to his findings. In
addition to prison inspection reports, the Chief Inspector also publishes periodic thematic reviews
on important issues such as suicide in prisons, life sentence prisoners, and women in prison.
    Another very important element of the public scrutiny of prisons in England and Wales is
exercised by members of the public who make up the Board of Visitors for each prison. Members
of the Board of Visitors for a prison represent the local community in scrutinising what goes on in
the prison. They are unpaid and do all their work in a voluntary capacity. Members have the right

                                          Montreux, 4. – 8. 9. 2000

                                Statistics, Development and Human Rights

to visit any part of the prison without exception at any time and to interview prisoners or other
members of staff in confidence. In order to carry out their work effectively, they need to have
access to proper statistics.
   Countries which are members of the Council of Europe have additional forms of accountability.
Those countries are subject to investigation by the European Committee for the Prevention of
Torture (CPT). Since 1990, the CPT has conducted over 70 inspections of places of detention in
countries across the European continent. Members of the CPT are appointed by state members and
come from a variety of backgrounds, including law, medicine, prison affairs and politics. In the
course of their visits they are assisted by independent experts. During its visits to each country, the
CPT has unrestricted access to all places of detention and complete freedom of movement within
them. The reports which are made as a result of such investigations are submitted in confidence to
the government concerned. The tradition has grown up over the years that countries should allow
these reports to be published alongside their own response. A number of these reports have been
highly critical.
   All of these inspectors rely heavily on statistical reporting to evaluate the prisons they visit.
Independent investigators will typically ask to review the register of prisoners. They will review
record-keeping of incidents, riots, and suicides. They will make sure prison authorities keep
records of minority groups in the prison. Particularly telling are records of the number and content
of complaints.
   These are matters which are also of interest to other independent experts who have the authority
to visit prisons. They include the UN Special Rapporteur on Torture and the Special Rapporteur on
Prisons in Africa.

5. Future Challenges for Statisticians: How to Measure Humanity in the Prison Setting?

   Prisons are primarily about people, the prisoners who are held in them and the staff who look
after the prisoners. It is, therefore, no surprise that some of the most important indicators about
whether a prison is humane and its regime rehabilitative are difficult to quantify statistically. There
is an understandable temptation for prison administrators and government ministers to focus on
indicators which can be easily measured rather than on those which go to the heart of the
institution’s operations.

      What is to be Measured?

   One of the difficulties for the prison service is that it is at the end of the criminal justice
spectrum. In most cases the authorities are likely to have tried every other solution with repeat
offenders and only when they have failed is the person sent to prison. At that point politicians and
public commentators begin to expect quick solutions. They expect the prison to succeed where all
other agencies have failed. The fact that not all prisoners go on to be honest and law-abiding
members of society after release is often interpreted as reflecting a failure of the prison system.
Why, so the argument goes, can the prison system not deflect all those in its care from a future life
of crime? Surely it should be possible to measure the success of the prison through simple
indicators such as re-conviction rates?
   In many countries re-conviction rates for released prisoners are disappointingly high. Of all
prisoners discharged from prison in England and Wales in 1995, 58% were re-convicted for an
indictable offence within two years of their discharge. The figures are worse for 14-16 year olds
who come out of detention in England and Wales. Seventy-seven per cent of young male prisoners
under 21 discharged from prison in 1995 were re-convicted within two years. (Home Office, Prison
Statistics, England and Wales, 1998).

                                          Montreux, 4. – 8. 9. 2000

                                Statistics, Development and Human Rights

   In response to these high reconviction figures, countries like England and Wales and New
Zealand are studying ways to lower reconviction rates. In the last three years alone, the Home
Office has published six Statistical Bulletins and two Research Studies on the subject of re-
conviction rates. (Research, Development and Statistics Directorate, Home Office Statistical
Bulletins, Nos. 1/97, 2/97, 5/97, 6/97, 5/99, 19/99; Home Office Research Studies Nos. 187 and
192). The theory is that re-conviction rates are the best measure of the successes and failures of a
state’s penal policy. In fact reconviction rates are a simplistic and arbitrary measure of prison
regimes. Re-conviction rates measure only crimes which have resulted in a conviction. They do
not, for example, take into account the relative seriousness of the subsequent offence. Nor do they
take any account of other factors which may have led an individual to commit further offences.

       Measuring Processes or Outcomes ?

   Too much dependence on quantitative measurements can be counter-productive. It is not
sufficient to emphasise the need to meet key performance standards and at the same time to ignore
the influence of the prison environment as a whole. A good example of this issue is the problem of
suicide in English prisons. In May 1999, in response to an increasing number of suicides in prisons
around the country, the Chief Inspector of Prisons published a comprehensive thematic review
discussing every aspect of the problem of suicides in prisons. (HM Chief Inspector of Prisons for
England and Wales, 1999). The Prison Service has since reviewed its policy for minimising the risk
of suicide and self-harm. This policy is now very highly thought of. Prison staff are well-informed
about suicide risk factors, methods that prisoners use to take their lives and proper procedures to be
followed in the event of a suicide in prison. But still, prisoners keep hanging themselves. There
were a record 91 suicides in prisons in England and Wales in 1999. Why is this so? It is because
prison policies do not operate in a vacuum. They must be understood in the context of a prison
culture that has gone wrong and that no longer meets the needs of many vulnerable prisoners.
   Another illustration of this principle comes from a visit earlier this year by the Chief Inspector of
Prisons to the segregation units at a large London prison. In his subsequent report the Chief
Inspector wrote that he had never seen a punishment unit where the relationship between prisoners
and staff was as troubled as it was at this prison. (HM Chief Inspector of Prisons for England and
Wales, 1999). The Chief Inspector concluded that the prison management focused too much
attention on Key Performance Indicators, Performance Standards, and bugetary considerations, and
not enough attention on the actual correct and humane treatment of prisoners. Such is the difficulty
of evaluating regimes by a purely quantitative method.

      Measuring Quality

    One of the greatest challenges for statisticians working in the field of criminal justice is to
develop a set of tools which can take account of the human dynamics of a prison. These will allow
the collection and analysis of information about some of the less quantifiable aspects of prison life.
It is these qualitative aspects of daily life which really make the difference between a prison which
is decent, humane and just and one which is intolerable.
    The need to develop mechanisms which are capable of measuring quality of performance as well
as quantity is now well recognised in many sectors of public service. The National Agency of
Corrections in the Netherlands, for example, is attempting to improve the quality of its
administration by using the methods of the Institute for Dutch Quality throughout its prisons.
(Reference: personal communication with the author). The extensive system of surveys of both
staff and prisoners which have been developed in the Scottish Prison Service over the last ten years
is a model which other prison services would do well to study. (Wozniak, McAllister, 1992;

                                          Montreux, 4. – 8. 9. 2000

                                 Statistics, Development and Human Rights

Wozniak, Gemmell, Machin, 1994; Wozniak, Dyson, Carnie, 1998). In this context let me also say
that at the International Centre for Prison Studies we have developed what we have called a
Humanity Audit. The Humanity Audit will try to assess the human dimension of prison life and the
way in which the human beings in prison, whatever the colour of their uniform, treat each other.

6. Conclusion

    It is a cliché to say that in many prison systems success is usually measured by absence of
failure. Generally speaking, prison systems are not comfortable in planning ahead. Their real skill
lies in solving problems once they arise, rather than in preventing the problems from arising in the
first place. This leads prison personnel to say that a good day in prison is a day where nothing
    This attitude will not be sufficient in future. Instead, there will have to be a culture in which staff
are capable of planning ahead in order to deliver more positive results. This planning will be based
on good information about what has happened in the past, how it has been dealt with and what the
results, successful and otherwise, have been. In order to achieve these ambitions prison
administrators are likely to have an ever-increasing need for good data and for statistics which will
be of use to them in fulfilling their difficult public service.


Amnesty International (1999). ‘No One Sleeps Here Safely:’ Human Rights Violations Against
Detainees in Brazil.

Burricand, C. and Hatal, C. (1997). Ten Years of Probationary Sentences. Infostat Justice No. 49.
French Ministry of Justice, Paris.

Council of Europe. (1996-1998). SPACE: Council of Europe Annual Penal Statistics. (ed. Tournier,
P.) Strasbourg.

Boe, R. and Muirhead, M. (1998). Have Falling Crime Rates and Increased Use of Probation
Reduced Incarceration? Forum on Corrections Research, 10. Correctional Service of Canada,

Coyle, A. (1994). The Prisons We Deserve. Harper Collins, London.

Dell S et al. (1991). Mentally Disordered Remand Prisoners. Home Office, London.

GUIN. (1999). GUIN Report: Monthly Prison Statistics, Moscow.

Gunn J et al. (1991). Mentally Disordered Prisoners. Home Office, London.

HM Chief Inspector of Prisons for England and Wales. (1999). Thematic Review: Suicide Is
Everyone’s Concern.

HM Chief Inspector of Prisons for England and Wales. (1999). HMP Wandsworth: Report of an
Unannounced Short Inspection.

Home Office. (1998). Prison Statistics: England and Wales.

                                           Montreux, 4. – 8. 9. 2000

                              Statistics, Development and Human Rights

Larocque, B. (1998). Federal Trends and Outcomes in Conditional Release. Forum on Correctional
Research, 10. Correctional Service of Canada, Ontario.

Press Trust of India, Ltd. (December 7, 1999). Delay in Dealing With Undertrials Not to Be

Research Development and Statistics Directorate. (1999). Statistics on Race and the Criminal
Justice System: A Home Office Publication Under Section 95 of the Criminal Justice Act 1991.

Research Development and Statistics Directorate. (1999). Explaining Reconviction Following a
Community Sentence. (Ed. May, C.) Home Office Research Study 192.

Research Development and Statistics Directorate. (1999). Reconvictions of Offenders Sentenced or
Discharged from Prison in 1995. Home Office Statistical Bulletin No. 19/99.

Research Development and Statistics Directorate. (1999). Reconvictions of Offenders Sentenced or
Discharged from Prison in 1994. Home Office Statistical Bulletin No. 5/99.

Research Development and Statistics Directorate. (1998). Reducing Reoffending. (Ed. Goldblatt, P.
and Lewis, C.) Home Office Research Study 187.

Research Development and Statistics Directorate. (1997). Restricted Patients : Reconvictions and
Recalls by the End of 1995. Home Office Statistical Bulletin No. 1/97.

Research Development and Statistics Directorate. (1997). Life Licencees : Reconvictions and
Recalls by the End of 1995. Home Office Statistical Bulletin No. 2/97.

Research Development and Statistics Directorate. (1997). Reconvictions of Prisoners Discharged in
1993. Home Office Statistical Bulletin No. 5/97.

Research Development and Statistics Directorate. (1997). Reconviction Rates of Those
Commencing Community Penalties in 1993. Home Office Statistical Bulletin No. 6/97.

Social Exclusion Unit, Policy Action Team 12. (2000). National Strategy for Neighbourhood
Renewal, Report of Policy Action Team 12: Young People, 24.

Stern, V. Alternatives to Prison in Developing Countries, 28-52 and 73-85. International Centre for
Prison Studies and Penal Reform International, London.

U.S. Department of Justice, Bureau of Justice Statistics. (1998). Sourcebook of Criminal Justice
Statistics, Table 6.51.

Wozniak, E. and McAllister, D. (1992). The Prison Survey: Scottish Prison Service Occasional
Paper No. 1.

Wozniak, E., Gemmell, M. and Machin, D. (1994). The Second Prison Survey: Scottish Prison
Service Occasional Paper No. 10.

                                        Montreux, 4. – 8. 9. 2000

                              Statistics, Development and Human Rights

Wozniak, E. and Dyson, G. and Carnie, J. (1998). The Third Prison Survey. Scottish Prison Service
Occasional Paper No. 3.

                                        Montreux, 4. – 8. 9. 2000


To top