Hough 1998

Home Office Research Study 179
Attitudes to
punishment: findings
from the British Crime
Survey
by
Michael Hough
Julian Roberts
A Research and Statistics Directorate Report
Home Office
Research and
Statistics
Directorate
PURL: https://www.legal-tools.org/doc/d55410/
London: Home Office
Attitudes to punishment: findings from the British Crime Survey
Home Office Research Studies
The Home Office Research Studies are reports on research undertaken by or

on behalf of the Home Office. They cover the range of subjects for which
the Home Secretary has responsibility. Titles in the series are listed at the
back of this report (copies are available from the address on the back
c over). Other publications produced by the Research and Statistics
Directorate include Research Findings, the Research Bulletin, Statistical
Bulletins and Statistical Papers.
The Research and Statistics Dir ectorate

The Dire c t o rate consists of three Units which deal with re s e a rch and
statistics on Crime and Criminal Justice, Offe n d e rs and Corre c t i o n s ,
Immigration and General Matters; the Programme Development Unit; the
Economics Unit; and the Operational Research Unit.
The Research and Statistics Directorate is an integral part of the

Home Office, serving the Ministers and the department itself,
its services, Parliament and the public through re s e a rch,
development and statistics. Information and knowledge from
t he se sou rces in fo r m s p oli c y d eve lop m ent and the
m a n agement of pro grammes; their dissemination improve s
wider public understanding of matters of Home Office concern.
First published 1998
Application for reproduction should be made to the Information and Publications

Group, Room 201, Home Office, 50 Queen Anne’s Gate, London SW1H 9AT.
©Crown copyright 1998 ISBN 1 84082 017 9

ISSN 0072 6435
ii
Foreword
Foreword
Those responsible for formulating criminal policy need to have an accurate

appreciation of the impact of crime upon society. An important dimension
in assessing this is public attitudes to sentencing. Policy needs to command
at least a degree of popular consensus; and where this consensus is lacking,
it is essential to understand the reasons for it. The British Crime Survey has
included questions on attitudes to sentencing since its inception in 1982,
and has made a significant contribution to our understanding of the issues.
The 1996 sweep cove red attitudes to punishment in particular detail,
allowing for an in-depth analysis of the topic. This report presents the results
of this analysis.
DAVID MOXON
Head of Crime and Criminal Justice Unit
Research and Statistics Directorate
iii
Acknowledgements
We should like to thank Pat Mayhew and Catriona Mirrlees-Black for their
consistently helpful advice and suggestions about analysing the British Crime
Survey, and for their comments on earlier drafts of this report. We are also
grateful to the Home Office for funding this study.
MICHAEL HOUGH
JULIAN ROBERTS
iv
Contents
Inhalt
Page
Foreword iii
Acknowledgements iv
Summary vii
Knowledge of crime and sentencing vii
Assessments of sentencers and sentences viii
The best ways of tackling crime ix
Victims’ sentencing preferences ix
Policy implications x
1 Introduction 1
The British Crime Survey 4
Outline of the report 4
2 Knowledge of crime and criminal justice 7

Perceptions of violent crime 9
Murders in England and Wales 10
Crime cleared up by the Police 10
Percentage of adult males with criminal records 11
Changes in courts’ use of imprisonment 12
Knowledge of sentencing options 12
The use of imprisonment for specific offences 13
Estimates of average time served in prison 16
Summary 16
3 Opinion about sentencers and sentencing 17

Assessments of sentencers’ performance 17
Public ratings of sentencers and public misperceptions 19
Who underestimates sentence severity? 26
Sentencing preferences in specific cases 27
Summary 30
v
4 Strategies for controlling crime 31

The role of sentencing in causing crime 31
The most effective way of preventing crime 33
Strategies for reducing prison overcrowding 35
Impact of imprisonment on offending 35
Summary 37
5 The views of victims 39

Summary 42
6 Conclusions 43
Appendix A:Organisation and design of the

British Crime Survey 47
Sampling 47
Fieldwork 48
Structure of the questionnaire 48
Weighting 49
Sampling error and design effects 49
Appendix B:Extract from 1996 BCS questionnaire 51

Attitudes to sentencing 51
References 61
Publications 65
vi
Summary
Summary
The British Crime Survey (BCS) has been an important source of information
about attitudes to punishment in England and Wales since it was set up in
1982. This report presents findings mainly from the most recent sweep,
carried out in 1996. This had a nationally representative core sample of
16,348 respondents aged 16 or over, with a response rate of 83 per cent.
One part of the interview was devoted specifically to sentencing issues.
Roughly half the sample were asked a series of questions covering:
• their knowledge of crime and sentencing
• their assessments of sentencers and sentences
• the sentence which they thought should be passed in a specific case

of burglary
• their views on the best ways of tackling crime
Those identified by the survey as victims have been asked since 1984 what
punishment they thought their offender should get. Results for victims of
burglary and car theft are presented here.
Knowledge of crime and sentencing
The 1996 BCS found widespread ignorance amongst the public in England
and Wales about crime and criminal justice statistics. Misperceptions were
systematic rather than random, in that majorities overestimated the gravity of
crime problems, and underestimated the severity of the criminal justice
system. Findings of particular interest are:
• the mistaken belief amongst the majority that recorded crime had
rapidly increased
• substantial ove restimates of the pro p o rtion of re c o rded cri m e

involving violence
vii
• a tendency to underestimate the proportion of the population with

criminal records
• large minorities being unaware of the upward trend in the use of

imprisonment
• widespread ignorance of sentences available to the court
• very substantial underestimates of the courts’ use of imprisonment

for three types of crime – rape, mugging and burglary.
Assessments of sentencers and sentences
The survey shows that the public in England and Wales take a jaundiced
view of sentencers and sentencing. Eighty–two per cent of the sample
thought that judges we re out of touch with the public; the fi g u re fo r
magistrates was 63 per cent. Four-fifths of people think that sentences are
too lenient, half saying that they are much too lenient. Judges were thought
to be doing the worst job amongst criminal justice professionals. The order
of the ratings of these groups, from top to bottom, was:
• the police service
• the Prison Service
• magistrates
• the Crown Prosecution Service
• the probation service
• judges.
The BCS has demonstrated equally clearly, however, that at least in part,
public dissatisfaction is grounded in ignorance of current practice, and in
ignorance of current crime trends. Those who were most dissatisfied were
most likely to overestimate the growth in crime and the degree to which
c rime is violent, underestimate the courts’ use of imprisonment and
underestimate the clear-up rate.
T ho se w ho we re mo st like ly to underestimate the c our ts’ use of

imprisonment have lower educational attainment than others, were likely to
be above average age and were more likely to read tabloid newspapers.
Women were more likely than men to underestimate the proportion of
viii
Summary
convicted rapists sent to prison, and owner occupiers more likely than
others to underestimate the use of imprisonment for burglars.
When people were asked about a real case of burglary, their sentencing
p re s c riptions we re, on balance, well in line with current sentencing
practice. They were told that the burglary was committed in daytime by
someone with previous convictions, and involved the theft of a video and
television set from the home of an elderly man who was out at the time.
Fifty-four per cent of the sample wanted a prison sentence, with sentence
lengths averaging less than the two years that the burglar actually got. The
remainder of the sample proposed community service orders (26%), a fine
(21%), a suspended sentence (18%), tagging (11%) or probation (9%). A
l a rge minority (44%) suggested compensation, either by itself or in
combination with imprisonment or another penalty. Those who had been
victims of crime were no more punitive than others; this held true for
victims of burglary as well as for victims of other types of crime.
The survey included an ex p e riment to see if people’s pre fe rence fo r

imprisonment was a function of their ignorance of the alternatives. Whilst
most of the sample selected their preferred sentences from a ‘menu’ on a
showcard, a sub-sample were denied this, having to make ‘top of the head’
choices instead. This group was much less inclined to select community
penalties and compensation, and more inclined to select imprisonment – 67
per cent against 54 per cent. The finding underscores how sentencing
preferences are shaped by the level of information available to respondents.
The best ways of tackling crime
Most of the sample thought that many different factors underlie current
levels of crime. They believed that sentencing levels were an important
determinant of crime trends. However, they tended to see changes in
parenting and in unemployment levels as more promising ways of reducing
crime. Their attitudes towards greater use of imprisonment was at least
ambivalent, with a widespread belief that imprisonment can stimulate as
well as prevent further crime. Far more people expressed a preference for
tougher community penalties (56%) than for building new prisons (18%) as a
means of tackling prison overcrowding.
Victims’ sentencing preferences
The BCS shows that there was a marked increase over the period from 1984
to 1996 in victims’ preference for tough sentencing, at least in relation to
two types of crime, burglary and car theft. There was no evidence to
ix
suggest that this trend was a function simply of increasing seriousness of the
average crime of this sort. Nor was there any evidence that the experience
of victimisation fuels a desire for tougher penalties. Victims’ preferences did
not seem, on balance, to be substantially out of line with current sentencing
practice.
Policy implications
The 1996 BCS suggests that there is a crisis of confidence in sentencers

which needs tackling with some urgency. People think that sentencers are
out of touch, and that their sentences are far too soft.
A criminal policy of ‘playing to the gallery’ and extending the use of

imprisonment further is not appropriate. The BCS suggests an ingrained
belief in lenient sentences whatever the reality – the same way that they
probably assume prices are rising, regardless of the actual rate of inflation.
The most likely reason for this is that people receive information about
sentencing largely from the media, and media news values militate against
balanced coverage. Erratic court sentences make news, and sensible ones do
not. As a result large segments of the population are exposed to a steady
stream of unrepresentative stories about sentencing incompetence.
Correcting public misperception about sentencing trends in this country

should promote greater public confidence in judges and magistrates. And,
since the judiciary occupy such a critical place in the criminal justice
system, increasing confidence in the courts should promote confidence in
the administration of justice.
Improving public understanding of sentencing and crime is easier said than

done. A great deal has alre a dy been done to improve the quality and
accessibility of crime and sentencing statistics. However there remains
obvious room for improvement. The most challenging demands are in
identifying effective ways of interrupting the processes which feed public
cynicism. To date very limited use has been made of the communication
techniques of the late twentieth century in letting the public know about
c u rrent sentencing practice. A successful stra t e gy for tackling publ i c
misperceptions will almost certainly have to resort to these techniques. It
will have to identify key audiences, such as opinion fo rm e rs, victims,
potential offenders and people at risk of offending, and convey appropriately
to each audience an accurate portrayal of current sentencing practice.
x
Introduction
1 Introduction
Of all criminal justice issues, sentencing is perhaps the most controversial,

and the one which attracts the greatest degree of public concern. The
public tend to be more critical of sentencers than any other decision-makers
in the criminal justice process (cf Walker and Hough, 1988; Roberts, 1992).
Research in several countries has shown that public support for the police is
highest, while the heaviest criticisms are reserved for judges.1 Whilst it is
obviously important to maintain public confidence throughout the criminal
process, it is, therefore, a particular issue in relation to the judiciary.
The need to sustain public confidence means that public opinion plays an
important, albeit indirect role in sentencing policy and practice. Most
recently, the Lord Chief Justice, Lord Bingham (1997) observed that he did
not “consider it would be right, even if it were possible, for judges to ignore
the opinion of the public”. He continued by noting that “the increase in the
prison population is not explained by any recent increase in sentencing
powers, and I have no doubt that it is related to the pressure of public
opinion”. Politicians, too, have frequently referred to the need to reflect or
incorporate the views of the public in the debate surrounding sentencing
policy (see Ashworth and Hough, 1996, for further discussion).
Whatever the precise nature of the linkage between opinion and sentencing
p ractice, there fo re, it is obvious that those who are re s p o n s i ble fo r
sentencing policy require a good understanding of what people know and
think about sentencing. The difficulty is that public opinion manifests itself
to sentencers and policymakers in many different ways. Most obviously, the
media - and the tabloid press in particular – simultaneously reflect and shape
knowledge and opinion. There are politicians’ postbags, and the activities of
formal and informal pressure groups. However, these conduits of public
opinion can provide a distorted image of public views. The only truly valid
measure of opinion is a representative survey. Even with such a survey,
however, problems remain. Researchers must ensure that the questions
posed to the public are adequate to capture the complexities of the
sentencing process. As we shall demonstrate below, simply asking the
public whether they think sentences are too harsh or too lenient is an
inadequate and indeed misleading way of measuring public opinion.
1 Huang and Vaughn (1996) summarise research in America which shows very positive attitudes to the police. For
example, a recent American sur vey found that three-quarters of respondents rated the police as “excellent”. Similar
results emerge in Canada (Roberts and Stalans, 1997).
1
C e rt a i n ly, simple survey questions of this sort paint a clear picture of a

punitive public. For example, questions along the line “Are sentences too
harsh, too lenient or about right?” find large majorities saying the sentences
are not harsh enough.2 A poll carried out for the Daily Mail (1 April 1996)
found 92 per cent of a random sample supporting “tougher sentences for
criminals, especially persistent criminals”.3 Findings of this sort have done
mu ch to consolidate the view amongst politicians and criminal justice
p ro fessionals that the public wants tougher justice. However, this is a
serious oversimplification of public attitudes to sentencing.
Survey research conducted in the 1980s (Hough and Moxon, 1985; Walker
and Hough, 1988) suggests a more qualified interpretation of public opinion
about sentencing. This work indicated that the public were less punitive
th an was ge n e ra l ly supposed . When asked about the appro p ri a t e
punishments for individual crimes, many people’s preferences tended to
reflect the range of sentences actually imposed by the courts. This finding
has been replicated in several other countries such as Australia, Canada, The
Netherlands and the United States (see Diamond and Stalans, 1989; Walker
and Hough, 1988; Roberts and Stalans, 1997; Walker, Collins and Wilson,
1988).
Clearly though, the public are dissatisfied with sentencing practice, or what
t h ey perc e i ve sentencing practice to be. What is re s p o n s i ble for this
d i s s a t i s faction? One explanation is that people simply do not have an
accurate perception of the sentencing process. Recent qualitative work
e m p l oying focus groups (see Hough, 1996) has uncove red systematic
ignorance of current sentencing patterns, and has demonstrated that this is a
factor fuelling public dissatisfaction with the courts.4
Information, it would appear, is critical to public attitudes to sentencing. As

a general rule, the less information that people have about any specific case,
the more likely they are to advocate a punitive response to it. This was
demonstrated by a Canadian study which divided a sample of respondents in
h a l f. One half we re gi ven a re l a t i ve ly brief newspaper account of the
sentencing decision in an assault case. Most people who read this account
of the case thought that the sentence was too lenient. The other group were
given a complete summary of all the court documents. In other words, they
had the same amount of information at their disposal as the judge in the
case. Only a small minority (19%) of this group belie ved that the sentence
was too lenient (Doob and Roberts, 1988). This experiment demonstrated
2 Similar results emerge with general questions relating to capital punishment. Three-quarters of the public favour re-
introduction of the death penalty (see Jowell et al., 1994).
3 One can speculate that questions using less loaded terms – ‘repeat offenders’ or ‘law-breaker’ rather than ‘persistent
criminals’ – might have generated less consensus.
4 Research in America, Australia, and Canada has also found that the public have little idea of sentencing patterns or
statutory maxima (see Williams, Gibbs and Erickson, 1980; Indermaur, 1987; Canadian Sentencing Commission,
1987).
2
Introduction
that the amount of information about a case is critical in determining public

reaction. Unfortunately, most newspaper descriptions of a case provide very
little information. As well, the cases chosen for coverage by newspapers
tend to be ones that resulted in what appears to be a lenient sentence. Both
these trends contribute to encouraging a public perception that the system
is very lenient, and that judges are thoroughly out of touch with the views of
the community.
One weakness of Doob and Roberts’ experiment (and with other studies that
have adopted this research strategy) is that the subjects were people who
had been recruited for a social science study. This kind of sample, it can be
argued, is not representative of the general population. A survey carried out
in England for the Royal Commission on Criminal Justice (Zander and
Henderson, 1993) cannot be criticised on the same grounds. The survey
interviewed actual jurors and came to similar conclusions. Participants were
asked what they thought of the sentence that had been imposed in the case
in which they had served as jurors. Fewer than one juror in four thought
that the sentence imposed was less severe than they had expected.
Finally, the critical role of information in determining attitudes also emerges

from US research on attitudes towards the use of imprisonment (Doble and
Klein, 1989; English, Crouch and Pullen, 1989). In these studies, people
were asked to sentence an offender described in a brief scenario. Half the
sample were given no other information about the possible sentences that
could be imposed; in other words they correspond to the average person on
the street. The other half we re gi ven a complete list of the ra n ge of
community-based penalties available to, and used by, the courts. People
who were given the complete list of available penalties were less likely than
members of the other group to advocate imprisonment of the offender. The
lesson is clear: part of the reason why so many people support the use of
imprisonment is that they may simply be unaware that there are other ways
of punishing offenders. In other words, the public may not be implacably
opposed to alternative punishments – but simply be ill-informed about them.
While these studies call into question stereotypes of a public committed to

“hang’em and flog’em” sentencing, many were mounted abroad, and many
a re now quite old. The 1996 British Crime Survey offe red a timely
opportunity both to chart opinion in an authoritative way and to explore the
factors which shape this opinion.
3
The British Crime Survey
The British Crime Survey (BCS) has been an important source of information
about attitudes to punishment in England and Wales,5 since it was set up in
1982. This report presents findings mainly from the most recent sweep,
carried out in 1996. Details about BCS methodology are given in Appendix
A. The 1996 BCS had a nationally representative core sample of 16,348
respondents aged 16 or over. The response rate was 83 per cent.
The 1996 version of the BCS contained a module devoted specifically to

sentencing issues. Roughly half the sample (8,365 respondents) were asked
a series of questions about various aspects of the sentencing process. In
addition to the standard question about sentence severity (“Are the sentences
handed down by the courts too tough, about right or too lenient?”),
respondents were asked to provide a sentence for a specific case. Previous
re s e a rch in the United Kingdom and elsew h e re has shown that it is
i m p o rtant to provide a specific case scenario in order to ensure that
respondents are not thinking of an atypical case. When respondents are
asked to respond to the general question without a specific case, they tend
to think of the most serious crime committed by an offender with a long
criminal record (c.f. Doob and Roberts, 1988).
The 1996 BCS also contained a series of questions which eva l u a t e d

respondents’ knowledge of the sentencing and conditional release process.
For example, respondents were asked to estimate the incarceration rate for
three common offences. After stating the percentage of offenders convicted
of these crimes that they thought were imprisoned, respondents were asked
to give their opinion about the percentage that they believed should be
incarcerated. In this way, analyses can relate what the public think about
sentencing trends to public opinion about sentencing practices. As well,
respondents were asked a series of questions about related issues such as
crime prevention, prison conditions and prison overcrowding. Finally, they
were asked to evaluate a number of criminal justice professions, including
the police, the Crown Prosecution service, judges and magistrates.
Outline of the report
Chapter 2 summarises findings from questions which probed publ i c

knowledge of the criminal justice system. In Chapter 3 we focus on opinion
re g a rding the sentencing process. We fi rst examine ge n e ral ratings of
s e n t e n c e rs and sentencing; we then examine how the ratings (large ly
negative) correlate with misperceptions about the criminal process; and
finally, we examine the sentencing preferences expressed in response to
5 The 1996 Scottish Crime Survey (MVA, 1997) has also covered attitudes towards crime and the criminal process,
though not in a form which is comparable to the findings presented here.
4
Introduction
specific cases. Chapter 4 presents findings about somewhat wider questions

on crime control strategy, and Chapter 5 discusses attitudes to punishment
amongst victims, including findings on trends since 1984. Chapter 6
s u m m a rises the findings and draws out policy implications. Whereve r
possible, we compare and contrast the present findings with the results of
research from other countries, in order to view British attitudes in context.
As we shall see, many findings are consistent with research elsewhere.
Appendix A contains additional information on the methodology of the
British Crime Survey. That part of the questionnaire which dealt with
attitudes to punishment can be found in Appendix B.
5
6
Knowledge of crime and criminal justice
2 Knowledge of crime and

criminal justice
This chapter presents findings from the 1996 BCS on public knowledge
about crime and criminal justice. Respondents were asked about:
• national crime trends
• the proportion of recorded crime involving violence
• the annual number of murders in England and Wales
• the clear-up rate
• the proportion of the male population with criminal records
• changes in the use of imprisonment
• the range of sentences available to the court
• current sentencing practice in the use of imprisonment for three

crimes
• time served in prison.
National crime rates
Respondents were asked whether the recorded crime rate for the country as
a whole had changed over the previous two years (i.e., 1993–1995). The
number of recorded crimes in 1995 was eight per cent lower than in 1993
(Povey, Prime and Taylor, 1997). As this information had not been published
at the time of fieldwork, people could not have known it. However, crime
figures for England and Wales are published every six months, and the
previous three sets all showed significant falls; figures for the 12 months
ending June 1995 were ten per cent lower than two years earlier (Home
Office, 1995).
7
Figure 2.1 gives a breakdown of responses (excluding the one per cent who
did not express a view). It shows that three-quarters of the sample “got it
wrong”. Thus well-publicised statistics about falling crime had very little
impact on popular perceptions – a finding in keeping with surveys in other
western nations (see Doble, 1996; Roberts and Stalans, 1997, for a review).
Both this and previous sweeps of the BCS asked respondents about changes
in the crime rate in their area. Whilst three-quarters thought that there was
more crime nationally, just over half (54%) thought that crime in their area
had increased. This proportion was higher in 1994 at 64 per cent and
higher still at 67 per cent in 1992 (Mirrlees-Black et al., 1996).
Figure 2.1: Changes in recorded crime
Figure 2.2: How much crime is violent?
There are se veral possible reasons why people disbelieve or fail to attend to
published crime statistics. First, headlines such as “Crime rate soars” have
been such a staple of tabloid journalism for so long that it would probably
take years of falling crime rates before the change became embedded in
public consciousness. Second, large proportions of the population will
simply miss or forget relevant newspaper stories and broadcasts. Third, even
when national crime rates fall, local crime rates may have risen; and
respondents in such areas would reasonably extrapolate from the local
experience in answering questions about national crime rates. Finally the
police figures probably were an unreliable guide to trends over this period.
The BCS provides a better measure for crimes against individuals and their
p ri vate property; the 1996 sweep actually showed a four per cent ri s e
between 1993 and 1995 (Mirrlees-Black et al., 1996). But even if we use the
BCS as a yardstick, almost half (46%) of the sample were wrong in saying that
there was “a lot more crime” nationally than two years before.
Perceptions of violent crime
Research in other countries has shown that violence is central to public

conceptions of crime. One explanation for this is that the news media –
which constitute the public’s primary source of information about crime –
disproportionately report crimes of violence. The more lurid and horrific,
the more newswo rt hy the crime. In fact, around six per cent of crimes
recorded by the police in England and Wales at the time of the survey were
violent or sexual in nature (Barcl ay, 1995); if BCS fi g u res are used,
woundings and robbery again account for 6%; a further 15 per cent are
common assaults involving little or no injury. Figure 2.2 shows that the
p u blic substantially ove restimates the extent to which crimes invo l ve
violence. Excluding the three per cent who chose “don't know” as a
response,1 almost four out of five said that 30 per cent or more of crimes
were violent.2 The mean response was 50 per cent.
1 This in itself is interesting. It suggests that people are confident that their perceptions of criminal justice are
accurate.
2 Comparable trends emerge in other countries. When this same question was posed in Canada, three-quarters of
Canadians estimated the same statistic to lie between 30 and 100 per cent (Doob and Roberts, 1988). Indermaur
(1990) reports the same finding using a sample of Australian respondents.
3 Mitchell (forthcoming) has shown that the public is sensitive to the wide variation in gravity of offences of homicide
– but it is unlikely that public conceptions of what constitutes murder are fully consistent with legal ones. In
responding to this question, people may have a rather different, broader definition in their minds.
9
Murders in England and Wales
Respondents were asked to estimate the number of murders recorded in

England and Wales in 1995. Comparing public perceptions to reality is a
little complex. Police statistics classify killings as homicides, making no
distinction between murder and manslaughter. In the large number of cases
i nvolving prosecution and conviction, the court ’s ve rdict subsequently
enables a distinction to be made between murder and manslaughter. In
1995, the police recorded 754 homicides; and extrapolating from cases
which actually came to court, between 350 and 400 could be regarded as
murders.3
People had more difficulty with this question than any other. This is
reflected in the relatively large number of people who refused to respond or
who chose “don't know” as a response: over five per cent. When they did
respond, people chose a wide range of responses, many of which were
clearly guesses. Fully one-quarter of the sample estimated under 50, while 3
per cent estimated in excess of 3,000. The average response to this question
was 616. It would be stretching things to conclude that the British public
tends to over-estimate the number of murders – a finding which has been
established in North America (Roberts, 1992).
Crimes cleared up by the police
An important indicator of police performance is the percentage of crime

incidents that are cleared up.4 The most relevant statistics, covering 1995
show a clear-up rate of 26 per cent (Home Office, 1996). The clear-up rate
receives little publicity in the news media, at least relative to trends in
recorded crime. Accordingly, it would be unsurprising if most members of
the public had little accurate idea of this statistic – or even, in many cases,
an understanding of the concept. People were asked the following question:
“Crimes are recorded as cleared up by the police when the offender is
prosecuted or otherwise admits guilt. Out of every 100 crimes reported to
the police, what number do you think are ‘cleared up’?” Respondents’
estimates varied from zero to 100 per cent. Table 2.1 shows the results.
4 Crimes are regarded as cleared up if they result in one of the following: a charge; a summons; a caution; a request
by the offender that the offence be ‘taken into consideration’ by the courts when sentencing for another crime; or a
formal admission of guilt made whilst serving a prison sentence for another offence.
10
Table 2.1: Estimates of the clear-up rate
Percentage of the sample making .....
Underestimates (under 20% cleared up) 9
Accurate estimates (20–30% cleared up) 26
Overestimates (31% or more cleared up) 65
Total 100
One interpretation of this pattern of results suggests that most respondents

had little idea about the actual clearance rate, and were simply guessing.
Certainly, the response most frequently chosen (by one in five respondents)
was 50 per cent – the successful bluffer’s way of expressing ignorance.
However, people who overestimate clear-up rates also tend to rate the police
most favourably. Thus overestimating clear-up rates could equally well be
interpreted either as an expression of confidence in the police, or as one of
the factors which actually shapes this level of confidence. We shall return to
this issue later in the report.
Percentage of adult males with criminal records
By the time that they reach the age of 40, fully 40 per cent of men in Britain
have a criminal record for a non-motoring offence – a fact little known
among the ge n e ral public. When the y we re aske d to estimate the
percentage of 40-year-old men with a criminal record, almost two-thirds
under-estimated this statistic. The median response was 30 per cent. If we
consider a response between 35 per cent and 45 per cent to be correct, one
in seven (14%) got it right. Over half (56%) under-estimated the statistic,
while 30 per cent overestimated.
One interpretation of this pattern of results relates to public perceptions of

criminality. Other research has demonstrated that the public believe that
c rime is committed by a small, easily identifi able population of re p e a t
offenders. This belief leads the majority of people to under-estimate the
number of men with criminal records.
11
Changes in courts’ use of imprisonment
Respondents were asked if they thought that the proportion of offenders

sent to prison had increased, decreased or stayed the same over the previous
two years. The correct answer is that there was a very substantial increase
in the custodial population and a corresponding reduction in the community
penalties. The number of persons sentenced to custody rose from 58,400 in
1993 to 79,100 in 1995, an increase of 35 per cent (Home Office, 1996); the
proportion of convicted offenders sent to prison rose from 15 per cent to 20
per cent.
Just over half the sample (56%) were aware of this change. Roughly a
quarter (27%) believed there had been no change while 15 per cent believed
that there had been a fall (the remaining 2 per cent said they did not know
or refused to answer). Widespread news media coverage of the swelling
prison population must account for this finding.
Knowledge of sentencing options
One question explored public knowledge of sentencing alternatives. Taking

for granted that everyone is aware of imprisonment, respondents were asked
to list sentencing options other than imprisonment. The thinking behind
the question was that public enthusiasm for imprisonment might at least in
part reflect ignorance of the alternatives.
The results indicated widespread awareness of some, but by no means all,

non-custodial sentences. Over two-thirds (69%) of respondents identified
community service; this was the most widely-known sentencing option.
Over half (58%) identified a fine. Surprisingly perhaps, only about a third
(35%) of the sample identified probation. Even smaller percentages of
respondents were aware of the other alternatives: suspended sentence5
(30%); compensation (16%); conditional discharge (8%); electronic tagging
(7%). It is clear then, that although large percentages of the public are aware
of some community penalties (such as community service), others such as
probation are not at all salient in people’s minds when thinking about
sentencing.
5 This sentence now accounts for only one per cent of the total – though it continues to loom large in public
consciousness (see Hough, 1996).
12
The use of imprisonment for specific offences
A final set of questions testing knowledge of the criminal justice system

c o n c e rned sentencing for three familiar cr imes: rape, mu g ging and
residential burglary. For each of these offences, respondents were asked
what percentage of convictions for males, aged 21 and over, actually result in
custody. In our analysis of these data, we have classified respondents into
various categories, reflecting their degree of accuracy in estimating the
incarceration rate for these crimes. We have regarded estimates as correct if
they fell roughly within ten per cent of the right answer. Those whose
estimates were roughly 10–30 per cent too low were classified as “a bit too
low”. For example, for residential burglary, the “correct” answer to the
question of what percentage of offenders were imprisoned in 1995 is 61 per
cent; any respondent providing an estimate between 50 per cent and 69 per
cent was classified as being accurate,6 and those scoring between 31% and
49% were classified as “a bit too low”. Figure 2.3 and Table 2.2 summarise
the findings.
L a rge m ajo r ities of respon de nts se ttle d fo r estimates of c urre n t

imprisonment rates which were much too low for all three types of crime.
For rape, 97 per cent of adult males convicted in 1995 were sent to prison
(data supplied by Home Office RSD). Respondents’ median estimate was 50
per cent. Figure 2.3 classifies 18 per cent of respondents as accurate, in that
they estimated between 85 per cent and 100 per cent. Twenty-six per cent
were ‘a bit low’, settling for between 60 per cent and 84 per cent. The
remaining 57 per cent made estimates which were ‘much too low’.
Since there is no legal offence of mugging, the survey defined it as “theft in

the street by means of force or the threat of force”. The Home Office
cannot provide sentencing statistics for this sub-group of ro bb e ri e s .
However, they probably make up the bulk of convictions, with robberies in
banks, shops and other businesses being relatively rare. Almost all (92%) of
adult male offe n d e rs convicted in 1995 of any fo rm of ro bb e r y we re
i m p risoned. We have conserva t i ve ly estimated that the perc e n t age of
convicted adult muggers who get custodial sentences is in the region of 70
per cent. According to this yardstick, 12 per cent of respondents were
accurate, and five per cent over-estimated the imprisonment rate, providing
estimates between 80 per cent and 100 per cent. Once again however, the
vast majority of the sample (75%) under-estimated the severity of sentencing
trends. Twenty per cent made a small underestimate – between 45 per cent
and 59 per cent, and 62 per cent made larger underestimates. The median
estimate of the percentage of adult muggers sent to prison was 35 per cent.
6 Classification decisions were not made rigidly in accordance with these criteria. For example, we erred on the side
of caution by including as accurate the large numbers who said that 50 per cent of convicted burglars were
imprisoned. Strictly speaking our criteria imply that these were “a bit low”.
13
Figure 2.3: Knowledge of sentencing practice: estimates of

courts’ use of custody
For residential burglary, 61 per cent of convicted adult male house-burglars

were imprisoned in 1995. Twenty-two per cent of responses were classified
as “correct”, namely those falling between 50 and 69 per cent. A further
eight per cent provided an over-estimate. Once again, the majority of the
sample under-estimated the severity of the system; 15 per cent were ‘a bit
low’ (31–49%) and the remaining 55 per cent made grosser underestimates.
The median estimate was 30 per cent.
Not surprisingly, responses to the three questions were inter-correlated.

That is, people who under-estimated the proportion of rapists sent to prison
also under-estimated the percentage of muggers and burglars who were
imprisoned.
14
Table 2.2: Summary of public estimates of imprisonment rates
Rape Mugging Burglary
Over-estimate
Rape: not applicable
Mugging: 80–100% – 5% 8%
Burglary: 70–100%
Accurate
Rape: 85–100%
Mugging: 60–79% 18% 12% 22%
Burglary: 50–69%
Small under-estimate
Rape: 60–85%
Mugging: 45–59% 26% 20% 15%
Burglary: 31–49%
Large under-estimate
Rape: 0–59%
Mugging: 0–44% 57% 62% 55%
Burglary: 0–30%
Total 100% 100% 100%
Question: Out of every 100 men aged 21 or over who are convicted of
rape (mugging/ house burglary), how many do you think are sent to
prison?
After estimating actual imprisonment rates, respondents were asked to say

what proportion should be imprisoned. Responses were unambiguous with
respect to rapists. The mean was 94 per cent, and over four-fifths of the
sample said that all rapists should be imprisoned. For muggers, the mean
was 84 per cent, with 57 per cent saying that all mu g ge rs should be
imprisoned. For burglars the mean was 80 per cent, with less than half of
the sample (42%) saying that all adult burglars should go to prison. The
implications of these findings are discussed in Chapter 3.
15
Estimates of average time served in prison
Respondents were asked the following specific question: “If someone is

sentenced to serve 12 months [in prison], how long, on average do you
think they will actually spend in prison?”. The right answer is six months
(assuming that no significant amount of extra time was awarded as a result of
infractions of prison rules). In general, people were fairly accurate. Half
( 49% ) put the fi g u re at fi ve to seve n mont hs. One in six (17%)
underestimated, and a third overestimated. The median estimate of the time
served was actually six months.
Summary
These findings demonstrate very clearly that there is widespread ignorance

about crime and criminal justice statistics. Misperceptions seem systematic
rather than random, in that majorities overestimated the gravity of crime
problems, and underestimated the severity of the criminal justice system.
Findings of particular interest were:
• the mistaken belief amongst the majority that crime was rapidly
increasing
• substantial ove restimates of the pro p o rtion of re c o rded cri m e

involving violence
• a tendency to underestimate the proportion of the population with

criminal records
• large minorities being unaware of the upward trend in the use of

imprisonment
• widespread ignorance of sentences available to the court
• substantial underestimates of the use of imprisonment for three types

of crime.
16
Opinion about sentencers and sentencing
3 Opinion about sentencers

and sentencing
The first part of this chapter presents results on assessments of sentencers’

performance and abilities. As will emerge, the results confirm earlier work
suggesting widespread dissatisfaction (e.g., Walker and Hough, 1988; Hough,
1996). The middle part of the chapter explores the relationship between
this dissatisfaction and public misperceptions about crime and sentencing.
The chapter ends with an examination of the sorts of sentences which
people think ought to be passed in specific cases.
Assessments of sentencers’ performance
The 1996 BCS shows that the majority of the public have little regard for
s e n t e n c e rs’ perfo rmance. Respondents we re asked seve ral re l eva n t
questions:
• whether sentences were tough enough
• whether judges and magistrates were in touch with ordinary people
• whether judges and magistrates were doing a good job.
Figure 3.1 shows that four out of five respondents thought sentences were
too lenient to some degree (excluding the 2% who expressed no opinion).
Half said that sentences were “much too lenient”. This result is consistent
with similar research in other common law countries. 1 There is more public
consensus on this issue than any other in criminal justice, including the
death penalty.
As Figure 3.2 shows, people also thought that sentencers were out of touch.
More than four out of five thought judges were out of touch to some degree,
and 46 per cent thought they were very out of touch. Magistrates fared
better: only 21 per cent of the sample viewed them as being very out of
touch. Even so, almost two-thirds thought they were out of touch to some
degree.
1 For example, 80 per cent of the Canadian public and a similar percentage of Americans hold this view, and have
done so for over 30 years (e.g., Flanagan and Longmire, 1996).
17
Figure 3.1: Are sentences tough enough?
Figure 3.2: Are sentencers in touch?
18
Figure 3.3 compares the ratings of the job done by judges and magistrates
with those of other professionals. Judges emerge with the lowest ratings of
all seven groups: excluding “don’t knows”, 32 per cent thought they did a
poor job, 49 per cent a fair job, and 20 per cent thought they did a good job.
Magistrates do marginally better than the probation service and the CPS, but
not as well as the prison or police services.2
Ta ken toge t h e r, these findings suggest that sentencers, and judges in

particular, face a crisis of public confidence. Their sentences are widely
regarded as far too lenient; they are reckoned to be out of touch with
ordinary people; and they are thought to do a worse job than the police, the
CPS or prison services.
Public ratings of sentencers and public misperceptions
In devising a policy response to public dissatisfaction with sentencers, the

first step must be to identify its source. If public attitudes are grounded in
misperceptions, then the problem is one of communication. But if the
better-informed members of society are equally dissatisfied, this would point
to a more substantive mismatch, soluble either by persuading people of the
merits of current practice, or else by adjusting practice.
Figure 3.3: How good a job are they doing? Sentencers and other
justice agencies
2 The police findings are consistent with other questions in the BCS about the quality of local policing. Eighty-one per
cent of respondents in the 1996 BCS thought that the police in their area did a good job (Mirrlees-Black and Budd,
1997).
19
As a first step in our analysis, we checked that the view that sentencers were
out of touch and did a poor job was indeed associated with a belief that they
were too lenient. Tables 3.1 and 3.2 confirm this to be so for judges: the
more that people thought sentences too lenient, the more likely they were
to believe that judges were out of touch with society, and doing a poor job.
Very similar findings emerged in relation to magistrates. This suggests that
when people express the view that judges are out of touch, they specifically
mean with respect to the leniency of sentences that are imposed.
Table 3.1: Perceptions of whether judges are in touch and

attitudes towards sentence severity
Judges are:
In touch A bit out Very out

of touch of touch
Sentences are: % % %
Too tough 2 2 3
About right 42 22 7
Too lenient 57 76 90
100 100 100
Table 3.2: Perceptions of how good a job judges do and attitudes

towards sentence severity
The job that judges are doing is:
Good Fair Poor
Too tough 3 2 3
About right 36 20 6
100 100 100
20
Having established that the belief in sentencers’ leniency is implicated in

p u blic dissatisfaction, we then examined how the belief in leniency
c o rrelated with misperceptions about sentencing practice. Table 3.3
c ro s s t abul ate s belief a bout leniency w ith estimat es of the use of
imprisonment for rapists, muggers and burglars. For all three crime types,
the lower the estimated use of imprisonment, the greater the belief that
s e n t e n c e rs we re too lenient. For example, people who believed that
sentences we re too lenient ge n e rated a lower ave rage estimate of the
p e rc e n t age of rapists incarc e rated than did people who believed that
sentences were about right. This implies that people who are dissatisfied
with the severity of sentences are also those who are particularly inaccurate.
Table 3.3: Estimates of imprisonment rates and beliefs about

sentence severity
Average estimates of imprisonment rate:
Rape Mugging Burglary
Sentences are:
Much too lenient 50% 34% 31%

About right 54% 38% 35%
Too lenient 60% 43% 39%
Averaging the estimates of imprisonment rates for the three crimes shows
the contrast between respondents who believe sentences are much too
le nient and the rest of the sample. Ave raged ac ross the offe n c e s ,
respondents who felt sentences are much too lenient believed that 38 per
cent of offenders were incarcerated. The average for those who felt that
sentences were a little too lenient was 42 per cent, and those who thought
that sentences were about right or too tough generated an average of 47 per
cent. This suggests that ignorance about current practice is one source of
public dissatisfaction with sentencing.
If beliefs about leniency correlated with misperceptions about sentencing

practice, they also were related to misperceptions of the crime rate, as Table
3.4 shows. People who thought that crime was steeply on the increase were
more likely than others to think that sentences were too lenient.
21
Table 3.4: Perceptions of crime rate and attitudes towards

sentence severity
National crime trends:
Rising Rising Stable or

a lot a bit falling
Too tough 2 2 4
About right 11 21 29
100 100 100
We can pursue this issue more directly by relating evaluations of judges to

the question in which respondents were asked whether they thought that
sentencing has played a role in the increasing crime rate. Not surprisingly,
people who thought that sentencing had been a significant cause of changes
in the crime rate were significantly more negative in their evaluations of
judges. Thus of those respondents who felt that sentencing has been the
most important cause of crime, almost half also felt that judges had been
doing a poor or very poor job. In contrast, of those who saw no relationship
between sentencing patterns and changes in the crime rate, only 24 per cent
felt that judges were doing a poor or very poor job (see Table 3.5).
Table 3.5: Ratings of judges and views about impact of

sentencing on crime
How good a job are judges doing:
Sentencing has Excellent/ Fair Poor/very poor

been.... good
The most important

cause 12% 41% 46%
A major cause 17% 49% 34%
Only a minor cause 25% 53% 22%
Not a cause 28% 49% 24%
....of changes in the crime rate
22
These trends also support the interpretation that the public regard judges as
playing an important role in crime control. People tend to think that varying
the severity of penalties will have an impact on crime rates; more lenient
sentences will lead to higher crime rates, harsher to a fall in crime. Here
too, there are international parallels.3
We carried out a logistic regression analysis to assess which aspects of public

misperception were most closely associated with a belief that sentences
were far too soft. Seven variables were included in the analysis. One, the
estimated pro p o rtion of convicted rapists who are sent to prison, wa s
rejected as a statistically non-significant predictor. The remaining six were
all identified as statistically signifiant predictors, and are listed below, in
order of predictiveness:4
1. Changes in national crime rate

(Those saying “a lot more crime” were most likely to think sentences
far too soft.)
2. Changes in use of imprisonment

(Those saying prison use “the same/down” were most likely to say
sentences far too soft.)
3. Estimated number of convicted muggers who were sent to prison

(Under-estimators were most likely to say sentences were far too
soft.)
4. The proportion of recorded crime involving violence

(Over-estimators were most likely to say sentences were far too soft.)
5. Estimated number of convicted burglars who were sent to prison

soft.)
6. Estimates of the clear-up rate

soft.)
This analysis provides strong evidence that people’s dissatisfaction with

perceived sentencing lenience stems at least in part from misperceptions
about crime and justice. It is striking that six out of the seven variables
should be independently predictive of perceived lenience. The excluded
va ri able, the estimated proportion of convicted rapists who are sent to
3 A recent poll in America found that when respondents were asked to explain increases in crime rates, almost half
the sample identified the courts (Maguire and Pastore, 1995). In Canada, over three-quarters of the polled public
agreed with the statement that “There is a great deal of crime because sentences are not severe enough” (Brillon,
Louis-Guerin and Lamarche, 1984).
4 Predictiveness was taken here from the order in which va ri ables were selected for inclusion in the regression
equation according to a forward stepwise procedure.
23
prison, was quite closely intercorrelated with the equivalent variable for
mugging (Pearsons r = +.4); if the latter is excluded from the analysis, the
former emerges as a significant predictor.
Table 3.6 illustrates the relationship between misperceptions and attitudes

using a different analytic approach. It contrasts the opinions of respondents
who were the best- and least-informed about crime and justice. We have
defined the best informed as the one in 20 who provided accurate answers
to four or more out of seven questions testing knowledge of crime and
justice. The least informed were the one in four who were wrong on all
seven counts.
Table 3.6: Opinions about sentencers, by level of knowledge

about crime and justice
Least informed Best informed Total sample

(n=2143) (n=376)
% % %
Sentences are:
Too tough 2 5 3
About right 12 37 19
Judges are:
In touch 14 29 18
A bit out of touch 34 40 36
Very out of touch 52 31 46
Magistrates are:
In touch 31 53 37
A bit out of touch 42 36 42
Very out of touch 26 11 21
Note: the seven “knowledge” questions were:

1. Trends in recorded crime over past two years (correct answer: same or less)
2. Percentage of recorded incidents that are violent (correct answer: 25%)
3. Percentage of recorded crimes cleared up (correct answer: 20% to 30%)
4. Percentage of men with criminal record by the age of 40 (correct answer: 35% to 45%)
5. Percentage of convicted rapists sent to prison (correct answer: 85% to 100%)
6. Percentage of convicted muggers sent to prison (correct answer: 60% to 79%)
7. Percentage of convicted burglars sent to prison (correct answer: 50% to 69%)
24
Table 3.7: Demographic breakdown of the proportion of the

sample making large underestimates of imprisonment rates
Percentage making large

underestimates of imprisonment rate for............
Burglary Mugging Rape
Educational attainment
A levels + 52% 59% 42%
Lower 56% 64% 63%
Newspaper preference
Tabloid 58% 66% 63%
Mail/Express 56% 62% 56%
Local 56% 63% 58%
Broadsheet 49% 58% 41%
None 53% 58% 59%
Social class
Non-manual 55% 62% 52%
Manual 56% 63% 61%
Sex
Males 56% 61% 46%
Females 54% 63% 66%
Age
16–29 47% 62% 53%
30–59 58% 63% 54%
60+ 54% 61% 65%
Housing
Owner 56% 62% 55%
Renter 50% 62% 61%
Household income
Under £15K 54% 62% 65%
£15K + 56% 62% 49%
Ethnic group
White 56% 62% 57%
Black 38% 58% 61%
Asian 46% 62% 54%
25
Who underestimates sentence severity?
The probability that misperceptions about crime and justice are fuelling
p u blic dissatisfaction makes it important to know which demogra p h i c
groups are especially misinformed about sentencing. Table 3.7 shows what
groups are most likely to under-estimate sentencing severity. Overall, those
who make large underestimates are likely to be poorly educated readers of
tabloid newspapers. There are few consistent patterns for age, sex, class,
income and race; however, women, those from manual households, older
people and the least affluent were more likely than others to underestimate
the use of imprisonment for rapists. Whites and owner occupiers were more
likely than others to do so in relation to burglars.
To assess which demographic va ri ables best predicted the tendency to

underestimate the use of imprisonment, we carried out a stepwise logistic
regression using as predictor variables all those shown in Table 3.7. The aim
of the analysis was to identify the variables which best predicted ‘under-
e s t i m a t o rs’ . We defined this group as all those wh o made large
underestimates (defined in Table 2.2) of the use of imprisonment for all
t h ree offence catego ries. Exactly a third of the sampl e we re under-
estimators. Poor educational attainment was the best predictor, followed by
sex (being female), reading tabloid newspapers, being over 50 and finally,
being an owner occupier. The fact that sex is a predictor reflects the fact
that women are more likely than men to underestimate rape sentences; that
owner-occupation is a predictor reflects the fact that owners underestimate
the severity of sentences for burglary.5
5 The analysis was carried out using weighted data. Similar findings emerged for the unweighted data-set, except that
the order of entry of the last two variables was reversed.
26
Sentencing preferences in specific cases
The implications of the pr eced ing analysis are clear: it wo uld be

i n a p p ro p riate for sentencers to respond to public dissatisfaction by
toughening up sentencing policy. At least in part, public dissatisfaction
stems from public ignorance of the system. In a sentencing climate in which
public misperceptions about crime and sentencing are pervasive, the only
safe way of assessing the acceptability of current practice is to elicit people’s
sentencing preferences for particular categories of crime, and to compare
their preferences to practice. Even then, it is clear that the category of
crime needs to be specified in some detail. When people are asked general
questions, they will answer with the worst case in mind. Previous research
has shown that when people respond to a question of this kind, most are
thinking of a violent offender with several, related previous convictions
(e.g., Doob and Roberts, 1988). This explains, in part at least, the punitive
response. The public is not “sentencing” the average offender coming
before the courts, but rather of the worst case scenario.
The approach followed in the 1996 BCS was to describe a real case, which
went to the Court of Appeal, and to get respondents to select a sentence (or
sentences, as multiple choices were permitted). The details of the case were
presented on a show-card as follows:
A man aged 23 pleaded guilty to the burg l a ry of a cottage

belonging to an elderly man whilst he was out during the day. The
offender, who had previous convictions for burglary, took a video
worth £150 and a television which he left damaged near the scene
of the crime.
The offender had been given a three-year sentence in the Crown Court,
which was reduced on appeal to two years. Had such a case appeared
before magistrates, it would almost certainly have attracted a custodial
sentence, and probably would have attracted the maximum of six months.
Crown Court sentences for similar cases might range from six months to two
years.
In getting respondents to “sentence” the offender, we varied the method of

asking the question. Two-thirds of the sample were first provided with a list
of sentencing alternatives,6 and were then asked to choose one or more
punishments. However, the remaining third, selected to constitute half the
sample of non-victims, were denied the “menu” of sentencing options.
(Victims had to be excluded from the sub-sample, as they had already been
gi ven the menu earlier in the interv i ew, when asked about ways of
punishing “their” offender.) The hypothesis being tested was that the top-of-
the-head reaction of most people is to think fi rst and fo remost ab o u t
imprisonment.
27
Table 3.8 presents findings for the majority of the sample, excluding the sub-
set of non-victims who “sentenced” without a menu of options. It shows
that imprisonment was the most favoured option. However, only slightly
more than half of respondents favoured imprisonment; a fifth favoured a
fine, and around a third favoured community penalties other than a fine.
These responses are on balance more lenient than either the Court of Appeal
judgement or the Magistrates Association guidelines which suggest that the
‘entry point’ sentence for a domestic burglary of this sort is a short prison
sentence.
It is also noteworthy that in the series of questions asking for estimates of

the pro p o rtion of convicted adult offe n d e rs who a re and ought to be
imprisoned, respondents on average said that 80 per cent of burglars should
go to prison. By implication, this – very typical – example of burglary fell in
the eyes of the public at the less serious end of the spectrum. This
underscores how useless for policy it is to provide survey findings pitched at
a general level. If the general public overestimates the seriousness of the
average burglary, as appears to be the case here, those responsible for
sentencing policy can derive little of value from the finding that, on average,
people think that 80 per cent of burglars should be locked up.
Table 3.8: Sentencing preferences in Court of Appeal burglary

(excluding sub-sample who sentenced without “menu”)
Percentage of
respondents choosing
Sentencing option %
Imprisonment 54
Suspended sentence 18
Fine 21
Probation 9
Community service 26
Tagging 11
Compensation 44
Discharge 1
Note: columns exceed 100% due to multiple selections by respondents
Respondents who chose sentences using the sentencing menu fell into
6 The alternatives were: imprisonment; suspended prison sentence; fine; probation; community service order;
electronic tagging; compensation; conditional discharge.
28
ro u g h ly equal groups of non-victims and victims (i.e., those who had

reported any victimisation as occurring within 1995). Common sense might
suggest that victims would be more punitive than non-victims. In the event,
this was not the case: 55 per cent of the victim sample favo u re d
imprisonment, compared with 53 per cent of non-victims. It might be
thought that this merely reflects the large number of victims of relatively
trivial incidents. However, when victims of burglary (whether attempted or
successful) are analysed separately, again a similar proportion (53%) favour
imprisonment for the Court of Appeal offender. The view that victims are
more punitive than non-victims, then, was not supported by these findings.
As we had expected, the sub-sample who had to choose a sentence without

the benefit of a sentencing menu opted more often for prison. Table 3.9
shows the sentencing choices of those who sentenced with and without the
menu of options.
It can be seen that just over two - t h i rds of the “non-menu” sub-sample
favoured a term of custody, whereas only half of the other group endorsed
i m p risonment as a sanction. This is a highly significant diffe re n c e .
Respondents provided with a list of options were more likely to favour
imposition of a suspended sentence, probation and community service.
Support for compensation was also higher when respondents were aware
that it was an option: almost half (44%) of the “menu” group ch o s e
compensation, compared with 22 per cent of the “non-menu” group.
Table 3.9: Sentencing preferences as a function of awareness of

options
Sentencing options No sentencing options

provided to respondent provided
Percentage of respondents
choosing sentencing options % %
Imprisonment 54 67
Suspended sentence 18 8
Fine 21 19
Probation 9 5
Community service 26 20
Tagging 11 4
Compensation 44 22
Discharge 1 1
Note: columns exceed 100% due to multiple selections by respondents
29
As for sentence length, the median7 term of imprisonment favoured by

respondents was 12 months. This result will also surprise those who believe
that the British public are highly punitive. After all, the case involved a
vulnerable victim, loss of property, and most important of all, the offender
had several previous convictions for the same offence.8 It is hard, on the
basis of these findings, to argue that the public are consistently more severe
than the courts. After all, the Court of Appeal sentenced the offender to two
years in prison, while almost half the public favoured a community-based
s an ction. And, of t hose member s o f the public who d id ch o o s e
imprisonment, only a quarter exceeded the Court of Appeal sentence.
Summary
This chapter has shown in unusually stark terms that the public in England
and Wales take a jaundiced view of sentencers and sentencing. Judges in
particular are regarded as out of touch with the public, and four-fifths of
people think that sentences are too lenient. The BCS has demonstrated
e q u a l ly cl e a r ly, howeve r, that at least in part, public dissatisfaction is
grounded in ignorance of current practice, and in ignorance of current crime
trends. Those who are most dissatisfied are most likely to overestimate the
growth in crime and the degree to which crime is violent, underestimate the
courts’ use of imprisonment and underestimate the clear-up rate. Those
who are most likely to underestimate the courts’ use of imprisonment have
lower educational attainment than others, and are more likely to read tabloid
newspapers. When people are asked about a specific case, their sentencing
prescriptions are, on balance, well in line with current sentencing practice.
7 The mean length of imprisonment was higher, but this was an instance in which the average was sk ewed by some
very extreme scores: 1% of the sample chose sentence lengths in excess of 20 years. For this reason, we feel that
the median is the most appropriate measure of central tendency.
8 Research has shown that like the courts, members of the public become far more punitive when the offender being
sentenced has several related previous convictions (see Roberts, 1997).
30
Strategies for controlling crime
4 Strategies for controlling

crime
This chapter presents the results of a set of questions designed to explore

what role people saw sentencing as having in crime control, what
alternatives they envisaged as effective, and the perceived costs and benefits
of imprisonment as a crime control strategy. The questions covered the
following topics:
• the role of sentencing in causing crime
• the most effective way of preventing crime
• strategies for reducing prison overcrowding
• the impact of imprisonment on offending.
The role of sentencing in causing crime
In the last chapter we discussed the relationship between perceptions of the

influence that sentencing has upon the crime rate and evaluations of the
judiciary. At this point we explore in greater depth public beliefs about the
relationship between sentencing patterns and crime rates. The public in
m a ny countries appear to believe that lenient sentencing is a cause of
i n c reasing crime rates. This issue was ex p l o red in this survey in the
following way. Those who stated that they believed the crime rate had
increased or decreased over the previous two years were asked: “You said
e a rlier that you thou ght that the re c o rd ed c rim e rate had
increased/decreased over the past two years. What role would you say
that le nient/ tou gh se nte nc ing by the c our ts has played in th is
increase/decrease?” They were provided with four response alternatives:
(a) the most important cause; (b) a major cause, but not the most important;
(c) only a minor cause; (d) not a cause of this change.
As discussed in Chapter 2, three-quarters of the sample thought that crime

had risen over the last two ye a rs; they we re asked whether l e n i e n t
sentencing had played a part. They saw a clear link. A quarter thought that
31
sentencing patterns were the most important cause, and 48 per cent saw it
as a major cause, of rising crime (see Table 4.1). If people believe that
lenient sentencing causes crime, then by infe rence they pro b ably also
believe that harsher sentencing would have a preventive effect, resulting in
lower crime rates. These results suggest that a majority of the British public
appear to subscribe to a perception of sentencing as a crime contro l
m e ch a n i s m . 1 Sentencing is accorded rather less weight when similar
questions are asked in different contexts. For example, ONS’s omnibus
survey carried a module of questions about drug-crime. In this, only 37 per
cent of respondents identified lenient sentencing as a ‘main cause’ of crime,
c o m p a re d with 65 pe r cent who opted for d rug misu se (C har les,
forthcoming).
The four per cent of respondents who thought – accurately – that recorded
crime rates were falling were asked whether tough sentencing had played a
part in this. Opinion was divided fairly evenly between those who thought
that sentencing was unrelated, a minor cause and a major cause.
Table 4.1: Relationship between perceptions of changes in the

crime rate and the cause of this change
Crime in the country as a whole
There is more There is less
Sentencing is:
Most important cause 26% 6%

A major cause 48% 25%
A minor cause 20% 35%
Not a cause 6% 33%
100% 100%
1 In this respect their perceptions are at odds with reality. There is general agreement among sentencing scholars that
changes in sentencing severity will have little impact on the overall crime rate. One reason for this is that such a
small percentage of offences are ever prosecuted. In England and Wales, some two per cent of of fences result in the
imposition of a sentence (Home Office, 1994). Increasing the average sentence, sa y, from six months to two years
will have no apprecia ble impact on the overall crime rate. Ashworth (1995) concludes that: “It should therefore be
clear that, if criminal justice policy expects sentencing to perform a major preventive function, it is looking in the
wrong direction” (p. 23).
32
The most effective way of preventing crime
Two questions addressed public opinion regarding the most effective way of
preventing crime. Respondents were first asked which of a series of ways to
prevent crime would be effective. This was not a free response question;
they were given a list of six strategies and were allowed to choose up to six
options. After having identified the crime prevention strategies which in
their view would be effective, respondents who had given more than one
response were then asked to choose the single strategy which in their
opinion would be the most effective. Since the same hierarchy of options
emerged from both questions, for ease of presentation, we shall present the
results from the latter question only (Table 4.2).
In response to the question about the single most effective strategy, the
following hierarchy emerged. “Increase discipline in the family” attracted
the largest number of respondents, 36 per cent of the sample.2 This was
followed by “reduce levels of unemployment” (25%); “make sentences
tougher” (20%); “increase the number of police officers” (9%); “increase
discipline in schools” (8%); and finally “increase the use of community-based
penalties such as fines and community service” (2%). Thus even though
most people believe that lenient sentencing is a major cause of increasing
c rime rates, the public has a broad, multidimensional view of cri m e
prevention, one which does not place exclusive or even primary emphasis
on harsher sentencing. The British public, like their counterparts in North
America3 and Australia, appear to believe that there are several causes of
crime, and that this must be reflected in any crime prevention strategy.
Table 4.2: Views about relative effectiveness of crime prevention

strategies
Which is the most effective in preventing crime?
1. Increase discipline in the home 36%

2. Reduce levels of unemployment 25%
3. Make sentences tougher 20%
4. Increase the number of police officers 9%
5. Increase discipline in schools 8%
6. Increase use of community-based
sentences such as fines, community service 2%
100%
2 The 1997 ONS Omnibus Survey discussed above also found that poor parental discipline was the most frequently
identified ‘main cause’ of crime.
3 When Canadians were given the same list, only 27 per cent chose “making sentences harsher”; (see Roberts and
Grossman, 1991).
33
The criminal justice alternatives (tougher sentencing; more police officers)

together attract less than one-third of the responses. Thus although earlier
questions have established that lenient sentencing is a concern to the British
public, four out of five people see the most effective solution to crime as
lying outside the criminal justice system, namely in the home, the schools
and the workplace.4 These trends are worth noting as they contradict the
view of the public as being exclusively oriented towards punishment. The
British public do want harsher sentences (or at least harsher than they
believe them to be), but more punitive sentencing is not seen as the panacea
for rising crime.
As with many other issues in this survey, responses to this question were
related to perceptions of sentencing severity. Table 4.3 shows that of those
who felt sentences were too lenient, only 21 per cent favoured reducing
unemployment as the most effective crime prevention strategy. Reducing
u n e m p l oyment was chosen by 49 per cent of those who thought that
sentences were too tough and 39 per cent of those who believed that
sentences were about right.
Table 4.3: Relationship between perceptions of sentencing

trends and choice of most effective crime prevention strategy
Sentences are Sentences are Sentences are

too tough about right too lenient
Most effective way

of preventing crime is:
Increase discipline
in the family 27% 33% 37%
Increase discipline
in schools 5% 8% 8%
Reduce unemployment 49% 39% 21%
Increase number of police 8% 7% 9%
Make sentences tougher 10% 8% 23%
Increase use of community
sanctions 2% 4% 2%
Total 100% 100% 100%
4 The reactions of the public in North America are comparable. In response to a similar question, less than a third of
Canadians regarded criminal justice solutions as the most effective w ay to prevent crime (Roberts and Grossman,
1991).
34
Strategies for reducing prison overcrowding
The ove rc rowded state of British prisons has also been the subject of
c o n s i d e rable news media cove rage, including widespread publ i c i t y
surrounding “prison ships”. Respondents were told that there is some
evidence that the prisons are overcrowded, and were then asked to choose
one of three ways of addressing the problem. One solution emerged with far
more support from the public than the others. “Find new ways to punish
offenders that are less expensive than prison but tougher than probation”
was chosen by over half the respondents (56%), while the next most popular
solution to overcrowding (supported by almost one-quarter of the sample)
was to “release some non-violent offenders from prison earlier than at
present with more probation supervision after release”.
The least favoured solution to overcrowding in prisons was the option to

“build more prisons and pay for them by raising taxes or cutting spending
in other areas”. This was endorsed by only 18 per cent of the respondents.
Clearly then, additional prison construction, however it is financed, is an
option that is supported by few members of the public. Here too there are
parallels with responses to this question in other countries, where support
for intermediate sanctions (those lying between imprisonment and
p robation) exceeds support for prison construction by a considerabl e
margin (c.f. Roberts, 1992; Roberts and Stalans, 1997).
If people’s attention had not been drawn to the cost of prison building, it is
probable that support for this strategy would have been stronger. It strikes
us as more sensible when canvassing opinion not to offer respondents a
“free lunch”; unless their attention is drawn to opportunity costs, the public
generally want more and better public services of every sort.
Impact of imprisonment on offending
The prison is central to a layperson’s conceptions of punishment. When

most peop le think abo ut s ente ncing, they thin k abou t len gths of
imprisonment. It is important therefore to know about attitudes towards,
and knowledge of, prison life. To explore this issue, respondents were asked
to agree or disagree with three statements. The first dealt with employment
t raining. Almost thre e - q u a rt e rs (71%) of respondents agreed with the
statement that “in prison offenders receive training for jobs”. Only 16 per
cent disagreed with the statement. (Twelve per cent neither agreed nor
disagreed with the statement.) Thus most of the public are aware of this
aspect of prison life. Less than half the sample (46%) agreed with the
following statement: “In prison, offenders are helped to become law-
abiding citizens.” This is confirmation of a widespread disenchantment
35
with the effe c t i veness of prisons in terms of promoting law - ab i d i n g

behaviour. Over one-third (36%) disagreed with the statement, while 17 per
cent neither agreed nor disagreed.
If prisons do not encourage law-abiding behaviour, do they at least punish

those serving time? The public appear to believe so. Just over one-half of
the sample agreed that “being put in prison punishes offenders”. Fewer
than one-third disagreed with the statement and 20 per cent neither agreed
nor disa greed. An interesting relationship emerged when responses to this
question we re cro s s - t abulated with responses to the question ab o u t
sentence seve ri t y. Respondents who believed that sentences were too
lenient were also less likely to subscribe to the view that prison punishes
offenders, as Table 4.4 shows.
Table 4.4: Relationship between perceptions of sentencing and

whether prison serves as a punishment
Sentences are:
Too lenient About right
Prison punishes 48% 65%
No opinion 20% 18%
Prison does not punish 32% 17%
100% 100%
This relationship suggests that one reason so many people think sentences
should be harsher is that they believe that imprisonment is not that aversive
an experience for inmates.
One criticism of prisons that has been around for many generations is that
t h e y may in fact promote crim inal behaviou r b y e nc ourag ing the
transmission of ways in which to offend. The public would appear to
subscribe strongly to this view. More than four out of five respondents
(82%) agreed with the statement that “In prison, offenders learn new ways
to commit crime”. Only six per cent of the sample disagreed with the
statement. (A further 11% neither agreed nor disagreed.)
Although many people have strong views about prison life, and how a prison
should be run, few people have any direct ex p e rience with custodial
settings. Respondents were asked “Have you ever been inside a prison, as
a visitor, or for any other reason?”. Four out of five respondents stated that
36
they had never been in a prison in any capacity. We carried out analysis to
see if direct experience with prison bore any relation to attitudes towards
prison.
Surprisingly, the only significant difference which we found was that those
with direct experience were more likely to agree with the statement that
prisons help inmates to become law-abiding citizens: 44 per cent, compared
to only 35% of respondents without experience of prison.
Summary
Findings presented in this chapter show that most people recognise that
many different factors underlie current levels of crime. They generally
believe that sentencing is an important determinant. However, they tend to
see changes in parenting and in unemployment levels as more promising
ways of reducing crime. Their attitudes towards greater use of imprisonment
is at least ambivalent, with a widespread belief in the negative potential of
i m p risonment in stimulating further crime, and a greater pre fe re n c e
expressed for tougher community penalties than for building new prisons.
37
38
The views of victims
5 The views of victims
In this section, we take a closer look at the responses of victims of crime.

Victims have been asked the same question in every BCS sweep since 1984,
asking what punishment “their” offender should receive. The format of the
question has remained essentially unchanged, allowing analysis of trends
over the 12-year period. We have focussed here on residential burglary
involving losses (i.e., excluding all forms of unsuccessful burglary) and theft
of cars, partly because these are relatively homogeneous categories of crime
and partly because almost all incidents are reported to the police.1 Figure
5.1 presents the results.
Figure 5.1 shows an increase between 1984 and 1996 from 33 per cent to 48
per cent in the proportion of burglary victims who want their offender
imprisoned. Unlike the findings presented above, these estimates are based
on small numbers (especially in earlier sweeps) and are subject to significant
sampling error.2 Nonetheless, it seems clear that victims of burglary have
become more punitive over the 12 years since 1984. Victims of car theft
show a similar pattern, with the proportion favouring imprisonment rising
from 17 per cent in 1984 to 44 per cent in 1996. It should be remembered
that concern about joyriders was of particular public and political concern
in 1991 and 19923 – though the difference between the 1992 and 1996
figures is not statistically significant.
1 Ninety-two per cent of burglaries with loss and 99 per cent of car thefts (including unauthorised taking) were
reported in the 1992 BCS. If unreported crimes are excluded from Figure 3.3, the proportion of burglary victims
favouring imprisonment increases three percentage points to 44 per cent; there is no change for victims of car
theft.
2 Assuming minimal design effects, the estimate of 33 per cent of burglary victims favouring imprisonment in 1984
may have a true value anywhere between 27 per cent and 39 per cent, and the corresponding figure of 49 per cent
for 1996 may have a true value anywhere between 45 per cent and 53 per cent (p<.05).
3 A series of serious road accidents involving stolen cars coupled with media accounts of joyriding in places such as
the Blackbird Lees Estate near Oxford led to the Aggravated Vehicle Taking Act, 1992.
39
Figure 5.1: Victims’ preferred sentence
Note: No BCS in1986 and 1990.
It might be argued that the increasing enthusiasm for imprisoning burglars

re f lects not a growing pu nitiveness but an increase in the ave rage
seriousness of cases. The BCS provides only partial support for this. In each
sweep, victims were asked to assign a score between 0 and 20 reflecting
their perceptions of the seriousness of their crime. The average score for
burglary increased from 8.0 in 1984 to 9.7 in 1992, then falling back to 9.2
in 1996. Interpretation of such a scale is complex. How victims manage to
locate their crime on this scale is a mysterious process. The scores may
generally reflect the objective level of harm done by the offender or the
seriousness with which victims felt they should be dealt. If the former, then
the increase in victims' desire for imprisonment over the period 1984–1992
might re f lect an increase in the ave rage seriousness of cases, but the
increase from 1992–1996 does not.
The 1992 BCS allows comparison between victims’ responses to the

question about their “own” offender and that about offenders in analogous
vignettes. Table 5.1 makes this comparison for victims of burglary and of car
theft. Numbers are even smaller – and imprecision even greater – than for
Figure 5.1, because only half of the sample completed that section of the
questionnaire involving the vignettes. Sentence pre fe rences for “own”
offender are also slightly different from those in Figure 5.1, for the same
reason. The table shows that victims we re more like ly to advo c a t e
i m p risonment for “their” offender than for the one in the vignette –
somewhat so in relation to burglary, markedly so for car theft. The most
likely reason for this is that on average, victims suffered more serious crimes
40
The views of victims
than those in the vignettes.4 It is also possible – but not empirically testable
in the 1992 BCS – that people judge that crimes in which they have been the
victim deserve heavier punishment than similar crimes committed against
others.
Table 5.1: Victims’ preferred sentences for ‘their’ offender and

analogous offender in vignette
Crime
Burglary Car theft

with loss
Own burglary Vignette Own car theft Vignette
Sentence % % % %
Prison 44 36 45 24
Fine 7 12 15 23
Community service 20 13 11 6
Probation, susp. sent 7 11 7 12
Compensation 6 13 16 15
Discharge or caution 6 9 4 18
Other/DK/depends 10 7 2 3
-- -- -- --
100 100 100 100
Notes: British Crime Survey, 1992. Weighted data. All burglaries involve entry to home and theft of property. Car thefts
include incidents which would be recorded by the police as unauthorised taking.
The 1992 BCS can provide some tentative clues as to whether the
experience of victimisation fuels punitive attitudes towards offenders in
general, as responses to the vignettes can be broken down according to
respondents’ status as victims of crime. Thus for example, 36 per cent of
the 175 burg l a ry victims who completed the sentencing questionnaire
advocated imprisonment for the burglar in the vignette, compared with 30
per cent of the overall sample. Corresponding figures for car theft are 24
per cent of victims, and 20 per cent of the overall sample. These figures may
seem sugge s t i ve of a causal link between victimisation and punitive
attitudes, but the differences do not approach statistical significance. This
finding is consistent with the results from the 1996 survey which we
4 Of those BCS respondents who were victims of burglary involving loss in 1991, 55% suf fered losses of at least £500,
and a further 14% had losses between £250 and £500. The burglary vignette specified the loss of a video-recorder -
whose average value in 1992 was probably lower than this. Similarly, the vignette for car theft specified that the car
was recovered undamaged; in over two thirds of BCS car thefts in 1991, the vehicle suffered at least £100 of
damage or was never recovered.
41
summarised in an earlier section of this report. Provisionally, and in line

with other work (Brillon, 1988; Hough and Moxon, 1985; van Dijk and
Steinmetz, 1988) the conclusion must be that experience as victim of crime
does not fuel punitiveness in any marked way.
Summary
Over time, there has been a marked increase in victims’ preference for tough
sentencing, at least in relation to two types of crime, burglary and car theft.
T h e re is no evidence to suggest that this trend is a function simply of
i n c reasing seve rity of the ave rage crime of this sort. Consistent with
findings reported in Chapter 2, there is no evidence that the experience of
victimisation fuels a desire for tougher penalties.
42
Conclusions
6 Conclusions
Several themes emerge from this, the most extensive analysis to date of
British attitudes toward sentencing and related criminal justice issues. First,
the general perception of leniency by the courts is widespread, and is clearly
related to evaluations of the judiciary, and to a lesser extent, magistrates.
While the perception persists that sentences are too lenient, criticism of
judges and magistrates will continue. Judges received the most negative
evaluations of all criminal justice professionals in this survey. The lesson
would appear clear: cl a rifying public misperception about sentencing
t rends in this country will promote public confidence in judges and
magistrates. And, since the judiciary occupy such a critical place in the
criminal justice system, increasing confidence in the courts will promote
confidence in the administration of justice. Since the perception of leniency
has been around for two decades now, it constitutes a priority in terms of
public education.
However, the public also need more accurate information about issues
related to sentencing. As we have seen from the BCS results reported here,
part of the public misperception about sentencing involves a link between
sentencing patterns and crime rates. In the light of this connection, it is
important to educate the public about trends in crime and the proportion of
crime that involves violence. Otherwise, people will continue to believe
that crime rates are increasing inex o rably (part i c u l a r ly rates of violent
c rime), and they will be inclined to attribute this increase to lenient
sentencing. In this way, the cycle of disappointment with sentencing and
criticism of the judiciary can only continue.
It would be wrong to characterise the British public as being highly punitive,

or as being consistently more punitive than judges. The public adhere to
what may be termed a multi-track approach to the punishment of offenders.
First, the public do not look exclusively to the courts to prevent crime.
Crime prevention is seen as a problem for society at large; it is not regarded
p ri m a ri ly as a criminal justice issue. Politicians may advocate tougher
sentencing in order to “do something about the crime problem”, but many
m e m b e rs of the public believe that reducing the crime rate is more a
question of changing the family and school environment, and increasing the
number of people with jobs. This is not to say that the public see no role for
sentencing in preventing crime; responses to the BCS indicate that lenient
43
sentencing is seen as being a cause of crime, but tougher sentencers are not
seen as the only or even the primary remedy.
While the public respond to polls by endorsing the view that sentences are
too lenient, this result must be seen in light of the findings reported in this
paper and elsewhere. When asked about sentencing in general, the public
think of the worst kinds of offenders (recidivists) and the most serious
crimes of violence, who represent a small minority of the total offender
population. When presented with a complete description of an actual case,
the public tend to be less punitive. As well, when gi ven adequate
information about the range of legal punishments available, the public are
less likely to endorse the use of imprisonment. Another way of stating this is
to say that when members of the public have a level of info rm a t i o n
comparable to that which is available to a judge in a court, the public
respond in a way that is fairly consistent with judicial practice. This result
has emerged from a wealth of re s e a rch in seve ral countries, and now
emerges from the 1992 and 1996 British Crime Surveys.1
It is worth reiterating the point that the British public are not alone in
holding a number of misperceptions about crime and justice. Throughout
this report we have noted parallels with public reactions to sentencing in
several other western nations. The British public are no less informed, or no
more critical than the public in these other countries.
On the basis of the International Crime Victimisation Survey (Mayhew and

van Dijk, 1997), the proportion of the public favouring tougher penalties
has risen rather than fallen over the period since 1992 when the use of
imprisonment has been growing rapidly. Given the findings presented in
Chapter 1, it is more plausible that perceptions of leniency are independent
of actual sentencing practices. People assume that sentences are lenient,
whatever the reality, the same way that they probably assume prices are
rising, re g a rdless of the actual rate of inf lation. Increasing the use of
imprisonment for some offence – say burglary – from 50 per cent to 90 per
cent would not assuage public concern over sentencing trends, since the
public would remain unaware of the shift in sentencing patterns. The
reason for this degree of misinformation of course is that people do not
receive information about sentencing practices in the news media, but
rather a steady stream of stories about sentencing malpractice, cases in
which a judge imposes what appears to be a very lenient sentence for a
serious crime of violence.
1 It is also consistent with findings derived from earlier sweeps of the BCS.
44
Conclusions
The news media make no attempt to explain the judicial re a s o n i n g

underlying the decision, or to place the sentence imposed in some statistical
context. In order to reconcile the public and the courts, a rational approach
must involve informing people about the nature of sentencing practices.
The most ch a l l e n ging demands are in identifying effe c t i ve ways of

interrupting the processes which feed public cynicism. To date very limited
use has been made of the communication techniques of the late twentieth
century in letting the public know about current sentencing practice. All
commercial operations of any significance market themselves, and most
public institutions now do so. The court system may not be entirely unique
in continuing with eighteenth century strategies of pomp and ritual to
sustain its authority. However, a successful strategy for tackling public
m i s p e rceptions will almost cert a i n ly have to re s o rt to more modern
techniques. It will have to identify key audiences, such as opinion formers,
victims, potential offenders and people at risk of offending, and convey in
media appro p riate to each audience an accurate port rayal of curre n t
sentencing practice.
45
46
Appendix A: Organisation and design of the British Crime Survey
Appendix A: Organisation
and design of the British
Crime Survey
The British Crime Survey (BCS) has been carried out six times, in 1982,
1984, 1987, 1992, 1994 and 1996. This appendix summarises the design of
the 1996 sweep. Previous sweeps differed only in minor detail. The 1996
British Crime Survey was carried out in early 1996 by Social and Community
Planning Research (SCPR) 1. Design of the survey was shared between staff
of the Home Office Research and Planning Unit and SCPR. Full details of the
survey’s methodology are to be found in SCPR’s technical report (Hales and
Stratford, 1997), and are summarised in Mirrlees-Black et al. (1996).
Sampling
The 1996 BCS comprised a ‘core’ sample and va rious supplementary

samples, none of which only are relevant to this analysis. (As in the three
previous sweeps, the 1996 survey carried an additional ‘booster’ sample of
bl a ck and Asian respondents. Howeve r, the questions on attitudes to
punishment we re not asked of this subsample.) The core sample wa s
designed to give, after appropriate weighting, a representative cross-section
both of private households in England and Wales and of individuals aged 16
and over in such households. As in 1992 and 1994, but not in earlier
sweeps, the Postcode Address File (PAF) was used in 1996 as the sampling
frame as it represents the fullest register of household addresses. (The PAF is
a listing of all postal delivery points in the country, almost all households
having one delivery point, or letter box.) As in previous years, inner city
areas were oversampled by a factor of about two. The 1996 definition of
inner cities approximates to that used in the three previous swe e p s ,
classifying postcode sectors on the basis of population density, level of
owner-occupied tenu re and pro p o rtion of households with heads of
households in ‘professional’ jobs.
1 SCPR did the fieldwork in the first survey; NOP Market Research in the second, and a consortium of both companies
in the third; a consortium of SCPR and BMRB did the fourth; and OPCS did the fifth.
47
The sample design required selection of 800 postcode sectors as primary

sampling units (PSUs). These were selected from a stratified list of sectors,
with stratifying factors being (i) inner city or not, (ii) standard region and
(iii) social class of household head. Once postcode sectors had been
selected, 27 addresses were randomly selected from each (30 in inner
cities). Then, where there were two or more dwelling units at an address
(2% of the total), interviewers randomly selected one. Finally, in dwelling
units with two or more person aged 16+, interviewers selected one, using a
randomising procedure.
Fieldwork
The 1996 core sample covered a nationally representative sample of 16,348

households in England and Wales. The response rate was 82.5% for the core,
higher than in the previous four years by five percentage points. One adult
(defined as 16 or older) in each household was interviewed. Computer
assisted personal interviewing (CAPI) procedures were used.
Structure of the questionnaire
These were several parts to the questionnaire: the Main Questionnaire;

Victim Forms (a maximum of six per respondent, with the fourth, fifth and
sixth fo rms being truncated ve rsions); one of two possible Fo l l ow - u p
Q u e s t i o n n a i re s, a furth er set of questions on h ouse ho ld fi res, the
Demographic Questionnaire and two Self-completion Questionnaires (which
covered questions about drugs for those aged 16-59 and questions about
ex p e rience of domestic violence for women aged 16-59). Of the two
versions of the Follow-up Questionnaire, Follow Up A focused on contact
with the police, neighbourhood watch, home security and fear of crime; and
Follow Up B covered attitudes to punishment. The rationale for introducing
different follow-up questionnaires was that the largely attitudinal questions
asked at this stage of the interview did not require the same precision as
those in other parts; it made sense, therefore, to split the sample into two,
extending topic coverage at little cost in terms of precision. The selected
adult respondent in each household completed the Main, Demographic and
one or other version of the Follow-up Questionnaires. Victim Forms were
completed only by those who said they had experienced a crime since the
beginning of 1993.
48
Appendix A: Organisation and design of the British Crime Survey
Weighting
Data from the survey we re weighted in a number of ways at the data

p rocessing stage. Weighting served two main purposes: to corre c t
imbalances introduced in sampling; and to correct imbalances created by the
design of the interview. Weights were applied to correct for:
• the over-representation of inner city residents
• cases where more than one household was covered by an entry on

the PAF file
• the fact that individuals living in larger households we re under-

represented (as the chance of an adult being selected for interview
was inversely related to the number of adults in that household).
Sampling error and design effects
The sampling design, particularly the stratification and degree of clustering

of addresses, has an ef fect on the statistical reliability of the results. A design
factor quantifies this effect on estimates, and is a measure of the expected
variability of estimates from repeated samples of the sample design, relative
to a simple random sample. Some examples of design effects for personal
crime rates (using the same weighting procedure as in this report) are to be
found in Hales and Stratford (1997). Design factors for the core sample
typically ranged from unity to 1.4.
49
50
Appendix B: Extract from the 1996 British Crime Survey questionnaire
Appendix B: Extract from

the 1996 British Crime
Survey questionnaire
This is a transcript of the questions presented on the computer screen to the

interviewer when running the 1998 BCS CAPI programme. It may, therefore,
differ in some respects from the CAPI interview.
E.1 Attitudes to sentencing
CrimUK [ASK ALL] I would like to ask whether you think that the recorded crime
rate for the country as a whole has changed over the past two years.
Would you say there is more crime, less crime or about the same amount
(since two years ago)? PROBE. Is that a lot or a little more/less?
1. A lot more crime

2. A little more crime
3. About the same
4. A little less crime
5 A lot less crime
Nvio [ASK ALL] The next few questions are about your perceptions of the level
of crime. If you don’t know an answer, please give us your best guess. We
are equally interested in what you think the answers might be. For several
of these questions we will be asking you to give an answer out of 100, for
example out of every 100 crimes how many are a particular type of crime.
Of every 100 crimes recorded by the police, roughly what number do you
think involve violence or the threat of violence? PROMPT if you don’t
know, please just guess.
0 ... 100
51
Nmurd [ASK ALL] How many recorded mu rd e rs do you think there were in
England and Wales last year? PROMPT if you don’t know, please just
guess.
0 ...99997
ClearUp [ASK ALL] Crimes are recorded as cleared up by the police when the
offender is prosecuted or otherwise admits guilt. Out of every 100 crimes
reported to the police, what number do you think are ‘cleared up`?
PROMPT if you don’t know, please just guess.
0 ....100
M40Rec [ASK ALL] When someone is convicted of a crime, they will have a
criminal record. By the time men reach the age of 40, how many out of
100 do you suppose have a criminal record?
0 ...100
SentSev [ASK ALL] The next few questions are about sentencing by the courts,
that is both the Crown Court and magistrates courts. In general, would
you say that sentences handed down by the courts are too tough, about
right, or too lenient? PROBE Is that a little too tough/lenient or much too
tough/lenient?
1. Much too tough

2. A little too tough
3. About right
4. A little too lenient
5. Much too lenient
SentCrim [ASK ALL] CARD B1 You said earlier that you thought that the recorded
crime rate had increased/decreased over the past two years. What role
would you say that lenient/tough (Sentsev) sentencing by the courts has
p l ayed in this incre a s e / d e c rease? Would you say that sentencing has
been......
1. ...the most important cause

2. ...a major cause, but not the most important one
3. ...only a minor cause
4. ...or not a cause of this change
52
NPrisCh [ASK ALL] Over the past two ye a rs do you think the pro p o rtion of
offenders sent to prison has increased, stayed the same or decreased?
1. Increased
2. Stayed the same
3. Decreased
TypSent [ASK ALL] CARD B2 This card shows a description of an actual criminal
case. READ OUT IF NECESSARY: A man aged 23 pleaded guilty to the
burglary of a cottage belonging to an elderly man whilst he was out
during the day. The offender, who had previous convictions for burglary,
took a video worth £150 and a television, which he left damaged near the
scene of the crime. ALLOW RESPONDENT TIME TO READ PROPERLY
THEN ASK:
TypeSentA [ASK IF Digit (SERIAL NUMBER) = 1 OR 5 OR 9] What type, or types, of

sentence do you think the offender should receive?
1. Imprisonment
2. Suspended prison sentence
3. Fine
4. Probation
5. Community service order
6. Electronic tagging
7. Have to pay compensation
8. Conditional discharge
9. Other
TypeSntAO [ASK IF TypeSentA = Other] INTERVIEWER RECORD OTHER
SentType The courts can impose a number of different types of sentences upon
people convicted of criminal offences. One of these is immediate
imprisonment. Which other types can you think of? PROBE. RECORD
EACH MENTIONED IN ORDER (UP TO 8 MENTIONS). CODE ALL THAT
APPLY

2. Fine
3. Probation
8. Other
53
SentTypO [ASK IF SentType = Other]. INTERVIEWER RECORD OTHER.
TypSentB [ASK I F Digit (SE RIAL NUMB ER) = 3 OR 7]. CARD B3. ALLOW
RESPONDENT TIME TO READ PRO P E R LY, THEN ASK: There are a
number of possible sentences which could be imposed in this case. What
type, or types, of sentence do you think the offender should receive?
1. Imprisonment
3. Fine
4. Probation
9. Other
TypSntBO [ASK IF TypSentB = Other]. INTERVIEWER RECORD OTHER
PrSent [ASK IF TypeSentA OR TypeSentB = Imprisonment]. How long do you

think the prison sentence should be?
1. In years only
2. In months only
3. In years and months
PrSentY [ASK IF PrSent = Year only OR Years and months]. CODE NUMBER OF
YEARS
0..30
PrSentM [ASK IF PrSent = Months only OR Years and months]. CODE NUMBER OF
MONTHS
0..30
NRapePr1 [ASK ALL] Now I would like to ask you about the kinds of sentences that
are imposed for rape, mugging and house burglary. First of all, out of
every 100 men aged 21 or over who are convicted of rape, how many do
you think are sent to prison?
0...100
54
NRapePr2 And how many do you think should be sent to prison?
0..100
NmuggPr1 [ASK ALL] Now turning to mugging which is theft in the street by means
of force or the threat of force, out of every 100 adults aged 21 or over
who are convicted of mu g ging, how many do you think are sent to
prison?
0...100
NmuggPr2 And how many do you think should be sent to prison?
0...100
NrBurgPr1 [ASK ALL] Now turning to house burglary out of every 100 adults aged 21
or over who are convicted of house burglary, how many do you think are
sent to prison?
0...100
NBurgPr2 And how many do you think should be sent to prison?
0...100
JudTouch [ASK ALL] I would now like to ask for your opinions of judges and
magistrates who decide what sentences to give. Firstly, Judges. Do you
think that judges are ge n e ra l ly in touch or out of touch with what
ordinary people think? IF OUT OF TOUCH: Is that a bit out of touch or
very out of touch?
1. In touch
2. A bit out of touch
3. Very out of touch
MagTouch [ASK ALL] Do you think that magistrates are generally in touch or out of
touch with what ordinary people think? IF OUT OF TOUCH: Is that a bit
out of touch or very out of touch?
1. In touch
2. A bit out of touch
3. Very out of touch
55
InPris [ASK ALL] Now I would like to ask you some questions about prisons.
Have you ever been inside a prison, as a visitor or for any other reason?
1. Yes
2. No
Serv1Y If someone is sentenced to serve 12 months, how long, on average, do

you think they will actually spend in prison? PROMPT if you don’t know,
please just guess. ENTER NUMBER OF MONTHS:
0..
PrSkill [ASK ALL]. CARD B4 I am going to read out some statements about
prison. For each one please choose a phrase from the card to say how
much you agree or disagree with it. READ OUT...
In prison, offenders receive training for jobs?
1. Agree strongly
2. Agree
3. Neither agree nor disagree
4. Disagree
5. Disagree strongly
PrLawAb CARD B4. In prison, offenders are helped to become law-abiding citizens?
1. Agree strongly
2. Agree
4. Disagree
PrPun CARD B4. Being put in prison punishes offenders?
1. Agree strongly
2. Agree
4. Disagree
56
PrCrim CARD B4. In prison, offenders learn new ways to commit crime?
1. Agree strongly
2. Agree
4. Disagree
OverCrow [ASK ALL]. CARD B5. There is some evidence that the prisons in this
country are overcrowded. Looking at this card which one of these do you
think would be the best way of reducing overcrowding?
1. Find new ways to punish offenders that are less expensive than
prison but tougher than probation
2. Release some non-violent offenders from prison earlier than at

present with more probation supervision after release
3. Build more prisons and pay for them by raising taxes or cutting
spending in other areas
PrevCr1 [ASK ALL]. CARD B6. Here is some possible ways of helping to prevent
crime in Britain. Which of these ways would in your view be effective in
preventing crime? (Enter at most six cases).
1. Increase discipline in the family

2. Increase discipline in schools
3. Reduce levels of unemployment
4. Increase the number of police officers
5. Make sentences tougher
6. Increase the use of community based penalties such a fines and
community service
PrevCr2 [ASK IF MORE THAN ONE ANSWER AT PREVCR1]. SHOW CARD B6 . And
which one way would in your view be most effective in preventing crime?
(Enter code). (SCREEN TO SHOW ONLY THOSE CODES RECORDED AT
PREVCR1)
1. Increase discipline in the family

2. Increase discipline in schools
3. Reduce levels of unemployment
4. Increase the number of police officers
5. Make sentences tougher
6. Increase the use of community based penalties such a fines and
community service
57
JobPol [ASK ALL]. CARD B7 . This card lists some different groups of people who
collectively form the criminal justice system. We would like to know how
good a job you think each off these groups of people are doing.
How good the Police are doing? Would that be an excellent, good, fair,
poor or very poor job? (Enter code) PROB
1. Excellent
2. Good
3. Fair
4. Poor
5. Very poor
JobCPS How good the Crown Prosecution Service, that is the body responsible for
making prosecutions, is doing? Would that be an excellent, good, fair,
poor or very poor job? (Enter code)
1. Excellent
2. Good
3. Fair
4. Poor
5. Very poor
JobJud How good judges are doing? Would that be an excellent, good, fair, poor
or very poor job? (Enter code)
1. Excellent
2. Good
3. Fair
4. Poor
5. Very poor
JobMag How good magistrates are doing? Would that be an excellent, good, fair,
1. Excellent
2. Good
3. Fair
4. Poor
5. Very poor
58
JobPri How good the prison services are doing? Would that be an excellent,
good, fair, poor or very poor job? (Enter code
1. Excellent
2. Good
3. Fair
4. Poor
5. Very poor
JobProb How good the probation services are doing? Would that be an excellent,
good, fair, poor or very poor job? (Enter code)
1. Excellent
2. Good
3. Fair
4. Poor
5. Very poor
MediaCov ALL] CARD B9. Now I would like you to think about coverage of crime in
the media, that is television, radio, newspapers and magazines. How good
a job you think the media does in providing you with accurate and
balanced information about crime? Would that be an excellent, good, fair,
1. Excellent
2. Good
3. Fair
4. Poor
5. Very poor
Newspap [ASK ALL] CARD B9. Which one of the following daily newspapers do you
read most often?(Enter code)
1. Daily Express
2. Daily Mail
3. Daily Mirror
4. Daily Star
5. Daily Telegraph
6. Financial Times
7. The Guardian
8. The Independent
9. The Sun
10. The Times
11. Local daily newspaper
12. Other daily newspaper
13. None
NewsPapO [ASK IF NewsPap = Other] INTERVIEWER RECORD OTHER
59
60
References
References
Ashworth, A. (1995) Sentencing and Criminal Justice. Second Edition.

London: Butterworths.
Ashworth, A. and Hough, M. (1996) ‘Sentencing and the Climate of

Opinion’. Criminal Law Review, 776–787.
Barclay, G . (ed.) (1995) Information on the Criminal Justice System in

England a nd Wa l e s . London: Home Office, Research and Statistics
Department.
Brillon, Y. (1988) ‘Punitiveness, status and ideology in three Canadian

provinces’, in N. Walker and M. Hough (eds.) Public Attitudes to Sentencing:
Surveys from Five Countries. Aldershot: Gower.
Brillon, Y., Louis-Guerin, C., and Lamarche, M.-C. (1984) Attitudes of

the Canadian Public Towa rd Crime Po l i c i e s . O t t awa: Ministry of the
Solicitor General.
Canadian Sentencing Commission. (1987) Sentencing Refo rm: A

Canadian Approach. Ottawa: Supply and Services Canada.
Charles, N. (forthcoming) Public perceptions of drug related crime in

1997. Home Office Research and Statistics Directorate Research Findings.
London: Home Office.
Diamond, S. and Stalans, L. (1989) ‘The Myth of Judicial Leniency in

Sentencing.’ Behavioral Sciences and the Law, 7: 73–89.
Doble, J. (1996) C rime and Corrections: The Vi ews of the people of

Vermont. Report available from the authors.
Doob, A. and Roberts, J.V. (1982) Crime: Some Views of the Canadian
Public. Ottawa: Department of Justice Canada.
Doob, A. and Roberts, J.V. (1988) ‘ P u blic punitiveness and publ i c

knowledge of the facts: some Canadian surveys.’ In (eds.) N. Walker and
Hough, M. Public Attitudes to Sentencing. Surveys from Five Countries.
Aldershot: Gower.
61
English, K., Crouch, J. and Pullen, S. (1989) Attitudes toward crime: A

s u rvey of Colorado Citizens and Criminal Justice Offi c i a l s . D e nve r :
Colorado Department of Public Safety.
Flanagan, T. and Longmire, D. (eds.) (1996) Americans View Crime and

Justice. A National Public Opinion Survey. Thousand Oaks: Sage.
Gibbons, D., Jones, J. and Garabedian, P. (1972) ‘ G a u ging Publ i c

Opinion about the Crime Problem.’ Crime and Delinquency, 18: 134-146.
Home Off i c e . (1995) C rim inal Statistics for England and Wa l e s .

Supplementary Tables. Volume 2. London: Her Majesty’s Stationery Office.
Hough, M. (1996) People Talking About Punishment. The Howard Journal,

35: 191–214.
Hough, M. and Mayhew, P. (1985) Taking Account of Crime: Key

Findings from the Second British Crime Survey. Home Office Research
Study No. 85. London: Her Majesty’s Stationery Office.
Hough, M. and Moxon, D. (1985) ‘Dealing with Offe n d e rs: popular

opinion and the Views of Victims.’ The Howard Journal, 24: 160–175.
Huang, W. and Vaughn, M. (1996) ‘ S u p p o rt and Confidence: Publ i c

Attitudes Towa rd the Po l i c e .’ In: T. Flanagan and D. Longmire (eds.)
A m e ricans Vi ew Crime and Justice. A National Public Opinion Survey.
Thousand Oaks: Sage.
Indermaur, D. (1987) ‘Public Perception of Sentencing in Perth, Western

Australia.’ Australian and New Zealand Journal of Criminology, 20: 163–183.
Indermaur, D. (1990) Perceptions of Crime Seriousness and Sentencing: A

comparison of court practice and the perceptions of a sample of the public
and judges. Canberra: Criminology Research Council of Australia.
Jowell, R., Curtice, J., Brook, L. and Ahrend, D. (1994) British Social
Attitudes: 11th Report. Aldershot: Dartmouth.
L o rd Bingham (1997) The Sentence of the Court. Police Fo u n d a t i o n

Lecture, July 1997. London: Police Foundation.
Maguire, K. and Pastore, A. (eds.) (1995) Sourcebook of Criminal Justice

Statistics. U.S. Department of Justice, Bureau of Justice Statistics.
62
References
Mayhew, P. and van Dijk. J.J.M. (1997). C riminal Victimisation in

Eleven Industrialised Countries. The Hague: Research and Documentation
Centre.
Mitchell, B. ( fo rthcoming) ‘Attitudes to homicide’ B ritish Jo u rnal of

Criminology.
Mirrlees-Black, C., Mayhew, P. and Percy, A. (1996) The British Crime

Survey. Home Office Statistical Bulletin, Issue 19/96.
MVA. (1997). The 1996 Scottish Crime Survey: first results. Crime and
Criminal Justice Research Findings No. 16. Edinburgh: Scottish Office
Central Research Unit.
Povey, D., Prime, J. and Taylor, P. (1997) Notifiable Offences. England

and Wales, 1996. Home Office Statistical Bulletin, Issue 3, March 1997.
Roberts, J.V. (1992) ‘Public Opinion, Crime and Criminal Justice.’ In (ed.)
M. Tonry, Crime and Justice. A Review of Research. Volume 16. Chicago:
University of Chicago Press.
Roberts, J.V. (1997) ‘ The Role of Criminal Record in the Sentencing

P ro c e s s .’ In: M. To n ry (ed.) Crime and Ju s t i c e. A Review of Research.
Chicago: University of Chicago Press.
Roberts, J.V. and Grossman, M. (1991) ‘Crime Prevention and Public

Opinion.’ Canadian Journal of Criminology, 32: 75-90.
Roberts, J.V. and Stalans, L. (1997) Public Opinion, Crime and Criminal
Justice. Colorado: Westview Press.
van Dijk, J. and Steinmetz, C. (1988) ‘Pragmatism, ideology and crime

control: three Dutch surveys.’ In: Walker, N. and Hough, M. (eds.) Public
Attitudes to Sentencing: Surveys from Five Countries. Aldershot: Gower.
Wa l k e r, J., Collins, M. and Wilson, P. (1988) ‘ H ow the Public Sees

Sentencing: An Australian Survey.’ In: N. Walker and M. Hough (eds.) Public
Attitudes to Sentencing: Surveys from Five Countries. Aldershot: Gower.
Walker, N. and Hough, M. (1988) (eds.) Public Attitudes To Sentencing.

Surveys from Five Countries. Aldershot: Gower.
Williams, K., Gibbs, J. and Erickson, M. (1980) ‘Public Knowledge of

Statutory Penalties: The extent and basis of accurate perception.’ Pacific
Sociological Review, 23: 105-128.
63
Z a n d e r, M. and Henderson, P. (1993) C rown Court Study. Roya l

Commission on Criminal Justice. Study No. 19. London: Her Majesty’s
Stationery Office.
64
Publications
Publications
List of research publications
A list of research reports for the last three years is provided below. A full list
of publications is available on request from the Research and Statistics
Directorate Information and Publications Group.
Home Office Research Studies (HORS)
151. Drug misuse declared: results of the 1994 British Crime Survey.
Malcom Ramsay and Andrew Percy. 1996.
152. An Evaluation of the Introduction and Operation of the Youth

Court. David O’Mahony and Kevin Haines. 1996.
153. Fitting supervision to offenders: assessment and allocation

decisions in the Probation Service. 1996.
155. PACE: a review of the literature. The first ten years. David Br own.
1997.
156. Automatic Conditional Release: the first two years. Mike Maguire,
Brigitte Perroud and Peter Raynor. 1996.
157. Testing obscenity: an international comparison of laws and

controls relating to obscene material. Sharon Grace. 1996.
158. Enforcing community sentences: supervisors’ perspectives on

ensuring compliance and dealing with breach. Tom Ellis, Carol
Hedderman and Ed Mortimer. 1996.
160. Implementing crime prevention schemes in a multi-agency

setting: aspects of process in the Safer Cities programme. Mike
Sutton. 1996.
161. Reducing criminality among young people: a sample of relevant

programmes in the United Kingdom. David Utting. 1997.
65
162 Imprisoned women and mothers. Dianne Caddle and Debbie Crisp. 1996.
163. Curfew orders with electronic monitoring: an evaluation of the first twelve
months of the trials in Greater Manchester, Norfolk and Berkshire, 1995 -
1996. George Mair and Ed Mortimer. 1996..
164 Safer cities and domestic burglaries. Paul Ekblom, Ho Law, Mike Sutton, with
assistance from Paul Crisp and Richard Wiggins. 1996.
165. Enforcing financial penalties. Claire Whittaker and Alan Mackie. 1997.
166. Assessing offenders’ needs: assessment scales for the probation service.
Rosumund Aubrey and Michael Hough. 1997.
167. Offenders on probation. George Mair and Chris May. 1997.
168. Managing courts effectively: The reasons for adjournments in magistrates’

courts. Claire Whittaker, Alan Mackie, Ruth Lewis and Nicola Ponikiewski. 1997.
169. A d d ressing the literacy needs of offenders under pro b a t i o n

supervision. Gwynn Davis et al. 1997.
170. Under st an ding t he sente ncing of women. edited by Caro l

Hedderman and Lorraine Gelsthorpe. 1997.
171. Changing offenders’ attitudes and behaviour: what works? Julie

Vennard, Darren Sugg and Carol Hedderman 1997.
172 Drug misuse declared in 1996: latest results from the British
Crime Survey. Malcolm Ramsay and Josephine Spiller. 1997.
No. 159 is not published yet.
Research Findings
30. To scare straight or educate? The British experience of day visits

to prison for young people. Charles Lloyd. 1996.
31. The ADT drug treatment programme at HMP Downview – a

preliminary evaluation. Elaine Player and Carol Martin. 1996.
32. Wolds remand prison – an evaluation. Keith Bottomley, Adrian James,

Emma Clare and Alison Liebling. 1996.
66
Publications
33. Drug misuse declared: results of the 1994 British Crime Survey.
Malcolm Ramsay and Andrew Percy. 1996.
34. Crack cocaine and drugs-crime careers. Howard Parker and Tim
Bottomley. 1996.
35. Imprisonment for fine default. David Moxon and Claire Whittaker.
1996.
36. Fine impositions and enforcement following the Criminal Justice

Act 1993. Elizabeth Charman, Bryan Gibson, Terry Honess and Rod
Morgan. 1996.
37. Victimisation in prisons. Ian O’Donnell and Kimmett Edgar. 1996.
38 Mothers in prison. Dianne Caddle and Debbie Crisp. 1997.
39. Ethnic minorities, victimisation and racial harassment. M a ri a n

Fitzgerald and Chris Hale. 1996.
40. Evaluating joint performance management between the police and

the Crown Prosecution Service. Andrew Hooke, Jim Knox and David
Portas. 1996.
41. Public attitudes to drug-related crime. Sharon Grace. 1996.
42. Domestic burglary schemes in the safer cities programme. Paul

Ekblom, Ho Law and Mike Sutton. 1996.
43. Pakistani women’s experience of domestic violence in Great Britain.

Salma Choudry. 1996.
44. Witnesses with learning disabilities. Andrew Sanders, Jane Creaton,

Sophia Bird and Leanne Weber. 1997.
45. Does treating sex offenders reduce reoffending? Carol Hedderman and
Darren Sugg. 1996.
46. Re-education programmes for violent men - an evaluation. Russell

Dobash, Rebecca Emerson Dobash, Kate Cavanagh and Ruth Lewis. 1996.
47. Sentencing without a pre-sentence report. Nigel Charles, Claire Whittaker

and Caroline Ball. 1997.
48 Magistrates’ views of the probation service. Chris May. 1997.
67
49. PACE ten years on: a review of the research. David Brown. 1997.
50 Persistent drug–misusing offenders. Malcolm Ramsay. 1997.
51 C u rfew orders with electronic monitoring: The first twelve

months. Ed Mortimer and George Mair. 1997.
52 Police cautioning in the 1990s. Roger Evans and Rachel Ellis. 1997.
53. A reconviction study of HMP Grendon Therapeutic Community.

Peter Marshall. 1997.
54. Control in category c prisons. Simon Marshall. 1997.
55. The prevalence of convictions for sexual offending. Peter Marshall.

1997.
56 Drug misuse declared in 1996: key results from the British Crime
Survey. Malcolm Ramsay and Josephine Spiller. 1997.
57 The 1996 International Crime Victimisation Survey. Pat Mayhew

and Phillip White. 1997.
58 The sentencing of wome n: a se ction 95 publication. C a ro l

Hedderman and Lizanne Dowds. 1997.
Occasional Papers
Mental disorder in remand prisoners. Anthony Maden, Caecilia J. A.

Taylor, Deborah Brooke and John Gunn. 1996.
An evaluation of prison work and training. Frances Simon and Claire

Corbett. 1996.
The impact of the national lottery on the horse-race betting levy.

Simon Field. 1996.
Evaluation of a Home Office initiative to help offenders into

employment. Ken Roberts, Alana Barton, Julian Buchanan, and Barry
Goldson. 1997.
The impact of the national lottery on the horse-race betting levy.
Simon Field and James Dunmore. 1997.
68
Publications
Requests for Publications
Home Office Research Studies from 143 onwards, Research and Planning
Unit Papers, Research Findings and Research Bulletins can be requested,
subject to availability, from:
Research and Statistics Directorate

Information and Publications Group
Room 201, Home Office
50 Queen Anne’s Gate
London SW1H 9AT
Telephone: 0171-273 2084
Fascimile: 0171-222 0211
Internet: http://www.open.gov.uk/home off/rsd/rsdhome.htm
E-mail: rsd.ha apollo @ gtnet.gov.u.
Occasional Papers can be purchased from:

Home Office
Publications Unit
50 Queen Anne’s Gate
London SW1H 9AT
Telephone: 0171 273 2302
Home Office Research Studies prior to 143 can be purchased from:
HMSO Publications Centre
(Mail, fax and telephone orders only)

PO Box 276, London SW8 5DT
Telephone orders: 0171-873 9090
General enquiries: 0171-873 0011
(queuing system in operation for both numbers)
Fax orders: 0171-873 8200
And also from HMSO Bookshops
69

Hough 1998

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hough 1998

Uploaded by

Copyright:

Available Formats

Home Office Research Study 179

A Research and Statistics Directorate Report

Home Office Research Studies

The Home Office Research Studies are reports on research undertaken by or

The Research and Statistics Dir ectorate

The Research and Statistics Directorate is an integral part of the

First published 1998

Application for reproduction should be made to the Information and Publications

©Crown copyright 1998 ISBN 1 84082 017 9

Those responsible for formulating criminal policy need to have an accurate

2 Knowledge of crime and criminal justice 7

3 Opinion about sentencers and sentencing 17

4 Strategies for controlling crime 31

5 The views of victims 39

Appendix A:Organisation and design of the

Appendix B:Extract from 1996 BCS questionnaire 51

• their knowledge of crime and sentencing

• their assessments of sentencers and sentences

• the sentence which they thought should be passed in a specific case

• their views on the best ways of tackling crime

Knowledge of crime and sentencing

• substantial ove restimates of the pro p o rtion of re c o rded cri m e

• a tendency to underestimate the proportion of the population with

• large minorities being unaware of the upward trend in the use of

• widespread ignorance of sentences available to the court

• very substantial underestimates of the courts’ use of imprisonment

Assessments of sentencers and sentences

• the police service

• the Prison Service

• the Crown Prosecution Service

• the probation service

T ho se w ho we re mo st like ly to underestimate the c our ts’ use of

The survey included an ex p e riment to see if people’s pre fe rence fo r

The best ways of tackling crime

Victims’ sentencing preferences

The 1996 BCS suggests that there is a crisis of confidence in sentencers

A criminal policy of ‘playing to the gallery’ and extending the use of

Correcting public misperception about sentencing trends in this country

Improving public understanding of sentencing and crime is easier said than

Of all criminal justice issues, sentencing is perhaps the most controversial,

C e rt a i n ly, simple survey questions of this sort paint a clear picture of a

Information, it would appear, is critical to public attitudes to sentencing. As

that the amount of information about a case is critical in determining public

Finally, the critical role of information in determining attitudes also emerges

While these studies call into question stereotypes of a public committed to

The British Crime Survey

The 1996 version of the BCS contained a module devoted specifically to

The 1996 BCS also contained a series of questions which eva l u a t e d

Outline of the report

Chapter 2 summarises findings from questions which probed publ i c

specific cases. Chapter 4 presents findings about somewhat wider questions

2 Knowledge of crime and

• national crime trends

• the proportion of recorded crime involving violence

• the annual number of murders in England and Wales

• the clear-up rate

• the proportion of the male population with criminal records

• changes in the use of imprisonment

• the range of sentences available to the court

• current sentencing practice in the use of imprisonment for three

• time served in prison.

National crime rates

Figure 2.1: Changes in recorded crime