Examination of Witnesses (Questions 100
TUESDAY 17 JUNE 2008
Q100 Dr Whitehead: A question for
Professor McGuire. You have written very extensively on the value
of evidence-based practice in interventions for offenders. How
do you think the Ministry of Justice is doing as far as progressing
implementation of its effective practice and how do you think
it stands alongside what you have written on the matter?
Professor McGuire: I have a sense
that it has lost some momentum, that the policy has been superseded
by concern about investment in imprisonment, and so on, and perhaps
an over-emphasis on managerial structures at the expense of the
face-to-face content of work that people working directly with
offenders have to do and perhaps not enough emphasis on the skills
development, the education, the dissemination of knowledge concerning
what will be more likely to be effective. I think it is rather
unfortunate that that momentum, which I think in fact was reasonable,
in fact probably accelerated at too speedy a pace at one stage
and perhaps some of the findings that were disappointing that
came from it were a function of that, although I cannot prove
thatthat is a hypothesisbut perhaps that caused
some loss of enthusiasm for it in senior decision-makers. At the
moment I feel as if there is a bit of a vacuum there perhaps,
or that it has lost its strength.
Q101 Dr Whitehead: So you might say
the evidence of evidence-based policy proved to be uncomfortable.
Professor McGuire: It is very
difficult to say if it was uncomfortable. I think some parts of
it are probably uncomfortable. If you look at a large report written
a few years ago for the Solicitor General, Canada, about the relationship
between lengths of prison sentences and the likelihood of recidivism,
taking other factors out of the equation such as risk levels and
so on, you probably would not be building more prisons. That would
be uncomfortable for many people, I think.
Q102 Dr Whitehead: I was sort of
coming to that in terms of thinking about the current evidence-base
specifically for reducing reoffending. Do you think there are
particular weaknesses in that evidence base and, bearing in mind
the apparent lack of information and evidence as far as NOMS is
concerned on, for example, the cost of community orders and, indeed,
even the completion of sentences, do you think that perhaps skews
the evidence-base towards certain outcomes?
Professor McGuire: I think it
probably does. I think there is a very large evidence-base. Unfortunately,
the bulk of it is North American, and the transfer from science
to practice is difficult enough without there also being an international
movement of information to take place. There are many weaknesses
in that evidence, many gaps in it, the quality of some of the
research is not good, but perhaps the biggest weakness is the
one that was touched on when Dr Chitty was giving evidence about
transfer of information to practice, and there are major hurdles,
as David has just said, in how one does that and the complexity
of that in large organisations, the aims of which are to serve
other public goods, as it were, and have specific notions about
how precisely to achieve that. So I think there are large gaps
there, and there are lots of specific gaps, which are too numerous
to mention, here of things that we do not know about implementingthe
better way to monitor and deliver domestic violence interventions,
for example: that is just one of the many dozens of questions
that could be askedso I think there is an underinvestment
in research in this area. You would expect an academic to say
that, would you not, but I think it is quite major if you compare
it with some of the other areas of inquiry that inform government
Q103 Dr Whitehead: In terms of your
view of what efforts are being made to put together the two issues
on evidence-based and effective practice within the Department,
what is your view on what efforts are being made, or do you think
there are not any efforts?
Professor McGuire: I cannot claim
to be very familiar with such efforts as are taking place at the
moment. I think there is a large emphasis on econometric models
for modelling the aggregate offending in the community using that
effort to predict aggregate needs for certain kinds of services.
I think it is somewhat atheoretical. We do not really understand
why, what are the mechanisms, the functions, the relationships
between the different ingredients that contribute to the dependent
variable, the amount of crime, the number of prison places that
we will need. It seems to me that an over-emphasis on an econometric
approach has been done at the expense of looking in more detail
at the motivations, the factors that influence different patterns
of offending in different places at different times. Perhaps the
localisation of feedback information that has been discussed by
Dr Chitty will be a great advantage in trying to enable local
decision-makers to think about that, guided by government policy,
Q104 Mr Heath: I want to carry on
the theme of effectiveness and how we measure it. Let me mark
my own card, as it were. I am convinced that the purpose of this
policy should be to stop crime, to be effective in reducing recidivism,
and I am also not a great fan of the Criminal Justice Act 2003
which I tried very hard to prevent reaching the statute book,
but it set out various purposes for sentencing and I am just wondering,
both of you, what your views are on how you can reconcile those
quite different purposes into a model, a measurement of effectiveness
which very often will take a different complexion over what purpose
you are using at any given time?
David Faulkner: I think with difficulty
is the first answer. Some of them are instrumental aims about
reducing reoffending, and then there is punishment, which is declaratory
with a moral background, and it is a different kind of thing.
I think if there was an attempt seriously to monitor the effectiveness
of sentencing against those aims, you would have to identify the
aim which was being sought in the particular sentence and then
decide on the criteria you were going to use for that aim, but
to put them altogether in one list and one basket, I do not think
you can measure the effectiveness without disaggregating it in
Q105 Mr Heath: Would you agree, Professor
Professor McGuire: Yes, I think
there are lots of different ways to measure the impact of an intervention.
The police do collect many kinds of data about how crime moves
around and the pattern of it over time, and some projects have
been evaluated in that way rather than by looking at the reoffending
rates of individual offenders; but we have to do that too, and
we should also look at severity, density, the spacing of that
over time and the changing nature of different kinds of offending
and how that relates to the individual life course. Obviously,
that would involve quite a large amount of investment in additional
ways of collecting information, and we know that, unfortunately,
when large-scale new methodologies are introduced into services
for collecting information, as has happened in the Probation Service,
the quality and constancy of the information is not all that high.
In research that I have been involved in we found that there was
lots of missing data. So the change of policy on that front needs
to be accompanied by adequate resourcing to ensure that the information
that you are asking for would actually be there at the end of
Q106 Mr Heath: But even with that
investment, is it possible to reconcile those totally disparate
objectives into a single measure of success?
Professor McGuire: In the end,
I agree with the measure that you are suggesting, which is trying
to reduce crime. The specific ways in which we can measure that
are quite various, but in the end they all do come to the same
thing. I think it is perfectly possible to reconcile those things.
Q107 Mr Heath: Let me throw in another
potential measure which in other areas of government policy would
be an important driver, and that is the cost-effectiveness of
the measure that is in place. Should that be a driver for criminal
justice policy, or is it an irrelevancy in comparison with the
other more important aspects which you have just described?
Professor McGuire: No, I think
it should be a driver. I accept that there is, as it were, a minimal
amount of investment that we cannot go below if we are to keep
people in secure accommodation and restrain them from causing
harm to others, and in the end there is a moral imperative to
that which society probably has not got a choice about, but there
is a copious amount of information, just as there is on the outcomes
of so-called "what works" type policies, on benefit
and cost analysis too. Again, unfortunately, much of it is North
American, but it shows very large returns on investment of good
intervention services, per capita savings of many thousands of
US dollars for a range of things for both young offenders and
for adults and it is entirely possible to get good returns if
you look at it in the best benefit cost way. One of the implications
of that would be transferring net resources from custody to the
community where you would probably see better returns on criminal
Q108 Mr Heath: Is that North American
research material relevant, do you think, to the British experience?
Professor McGuire: I think it
is. We use some of the interventions that are costed within it
in some of our services in England and Wales.
Q109 Chairman: Do you think the prospects
for the measures such as you have just described are inhibited
by a system in which the process of deciding and, therefore, allocating
different resources is so divided between those who sentence,
those who build prisons and make decisions to build prisons and
those who are responsible for services in the community? Have
we got a decision-making structure which is incapable of deciding
whether or not that transfer should take place?
Professor McGuire: To some extent
that is true. I think it has become very disparate and there are
large discrepancies in knowledge and the practice gaps between
the agencies that are responsible for the different components
of that. I think it is very hard to hold it together, and perhaps
one of the tasks of NOMS is to look more closely at that, but,
if I can introduce a slightly different idea, the notion of therapeutic
jurisprudence, which many people in the legal profession are very
keen on, that would give courts a different relationship, both
with offenders and with the agencies that are providing services
and supervision to offenders, thereby, if you like, giving a feedback
of information that would enable magistrates and judges to be
better informed about the decisions they had made, the impact
of those decisions locally and on individuals that they had sentenced.
Q110 Alun Michael: It relates very
much to what you have just been talking about in terms of who
uses the evidence for what policy-making purposes. We have heard
a suggestion that the relationship between internal and external
research on crime policy has broken down. Do you think that is
the case? What should the relationship between internal and external
evidence be and how could its effectiveness perhaps be strengthened?
David Faulkner: I think the relationship
at present is not very good. Professor McGuire has made the distinction
between the econometric research, which is where most Government
effort is going at present either internally or directly funded,
and the research which is being done by some of my own colleagues
in Oxford on relationships and motivation where there is good
work. Again there are gaps and weaknesses in the evidence but
it is important work with some real pointers for policy which
are not at this stage being brought into the Government's approach
to dealing with offenders. Again, I am fairly optimistic that
at least there are people within the Department of Justice who
are looking at this and seeing the importance of that motivational
work alongside the econometric work that is already taking place.
Another thing which does worry me and does rather depress me is
that many academics are not interested in working with policy
or Government but are more interested in pursuing their theories
and developing notions of punishment or security or whatever.
That is important stuff but they do not really want to engage
with policy on that and they do not get much encouragement, and
I would like to see a better relationship in that sort of field,
and probably it applies to areas other than criminal justice.
Q111 Chairman: That is what Lord
Nuffield said when he founded Nuffield College.
Professor McGuire: I think there
have been problems in that area. I am aware of a number of conflicts
which I do not think can be discussed here but I know
Q112 Alun Michael: It is starting
to get interesting!
Professor McGuire: To some extent
my own experience was a sense of projects being micro-managed
and not enough credibility being given to researchers who had
experience and who could, as it were, do the work. I think the
group I was part of became curious as to why that was the case
and were frankly puzzled by it, because there was not anything
we were doing which was in any way going to embarrass anyone.
So there are problems. I am aware, having spoken to a number of
criminology colleagues, that there have also been some disagreements.
I think that could do with some healing, could do with some improvement.
Q113 Alun Michael: What is the answer
to that? Is it some sort of independent research unit informing
policy, or is it mediation between policymakers, who clearly want
information now but can help to inform policy, and researchers
who, as David Faulkner suggested, perhaps are taking a more philosophical
view of their research? How can we improve the quality of the
way in which policy is informed?
Professor McGuire: I think it
would be good if RDS was more independent but the Government does
need its own research unit and RDS produces fantastic research
and it is extremely highly regarded all over the world and is
the biggest unit of its kind almost anywhere, perhaps apart from
the National Institute of Justice in Washington. But because it
is policy driven I think sometimes that does create some friction
with independent academics.
David Faulkner: I think more opportunities
for people just to talk to each other and be encouraged to talk
to each other and not have their conferences in isolation with
nobody coming to each other's meetings and things like that, would
be a good start.
Alun Michael: Like the role of the Committee,
Chairman: Indeed. That is a positive
note to end on. Thank you very much indeed. The Committee will
continue in private session.