Home Office Research Study 245 Improving Public Attitudes on the Criminal Justice System 2002

Home Office Research Study 245 Improving Public Attitudes on the Criminal Justice System 2002

• Topics: Prisoner Privileges, Release and Reentry, Sentencing, Collateral consequences of prison

Download original document:
Brief thumbnail

This text is machine-read, and may contain errors. Check the original document to verify accuracy.

Home Office Research Study 245

Improving public attitudes to the
Criminal Justice System:
The impact of information
Becca Chapman, Catriona Mirrlees-Black and Claire Brawn

The views expressed in this report are those of the authors, not necessarily
those of the Home Office (nor do they reflect Government policy).

Home Office Research, Development and Statistics Directorate
July 2002

Improving public attitudes to the Criminal Justice System: The impact of information

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. Other publications produced by the Research, Development and Statistics
Directorate include Findings, Statistical Bulletins and Statistical Papers.

The Research, Development and Statistics Directorate
RDS is part of the Home Office. The Home Office's purpose is to build a safe, just and tolerant
society in which the rights and responsibilities of individuals, families and communities are
properly balanced and the protection and security of the public are maintained.
RDS is also part of National Statistics (NS). One of the aims of NS is to inform Parliament and
the citizen about the state of the nation and provide a window on the work and performance
of government, allowing the impact of government policies and actions to be assessed.
Therefore –
Research Development and Statistics Directorate exists to improve policy making, decision
taking and practice in support of the Home Office purpose and aims, to provide the public and
Parliament with information necessary for informed debate and to publish information for
future use.

First published 2002
Application for reproduction should be made to the Communication Development Unit,
Room 201, Home Office, 50 Queen Anne’s Gate, London SW1H 9AT.
© Crown copyright 2002
ISBN 1 84082 851 X
ISSN 0072 6435

Foreword

Successive sweeps of the British Crime Survey (BCS) and other surveys have shown how
poorly informed the public are about crime, the operation of the criminal justice system (CJS)
and sentencing in particular. Furthermore, those who are worst informed hold the most
negative views about the CJS, and those who are best informed tend to have greater
confidence in most aspects of the system.
This previous work demonstrates a correlational relationship between knowledge and
attitudes, but not a causal one. The aim of the research reported here was to determine
whether providing information would have an effect on levels of knowledge; whether any
improvement in knowledge would have a direct impact on attitudes; and which of three
methods of presentation would be the most efficient and effective at imparting information to
the public.
The indications of this small-scale study are that public knowledge of the CJS can be
improved by a range of methods. Each mode of communication had differential effects on
attitudes, but each tended to improve satisfaction with at least some aspects of the system.
Of those modes of communication trialed, a specially designed booklet was as effective a
method as the others and considerably cheaper. On the available evidence it also
appeared to reach a wider audience. The booklet has now been revised in the light of the
research and has been made widely available.
David Moxon
Crime and Criminal Justice Unit

Improving public attitudes to the Criminal Justice System: The impact of information

Acknowledgements

Firstly, thanks are due to those members of the public who gave up their time to participate
in this research.
The research was organised and the interviews conducted by Taylor Nelson Sofres.
Particular thanks go to Neil Russell, Katherine Davis and Cynthia Pinto and to Professor Rod
Morgan who provided consultancy support and chaired the seminars.
The Prison Service presentations at the seminars were given by Gareth Davies, Louise
Taylor, John Thomas and Danny McAlister. Mark Harris, Stewart McPhillips, Granville Brunt
and Matthew Kelley were the presenters from the Probation Service. We are very grateful
for the time and thought they put into the presentations and the question and answer
sessions.
The booklet was produced by Eaglevision and Appetite and printed by Printflow.
The video was developed and produced by Eaglevision.
This research was commissioned by the Sentencing Review team. We are grateful to John
Halliday and Cecilia French for their input and interest in its development and findings.
A number of Home Office colleagues contributed to the development of this work. Particular
thanks go to Pat Dowdeswell and Victoria Richardson.
Julian Roberts of the University of Ottawa and Mark Ormerod provided very helpful reviews
in the final stages of writing this report.
Becca Chapman
Catriona Mirrlees-Black
Claire Brawn

Contents

Foreword

Acknowledgements

Summary

Introduction
Knowledge, attitudes about crime and sentencing and confidence in the CJS
Can information have an impact on attitudes?
What sort of information might have an influence on knowledge and attitudes?
Structure of the report

1
2
3
4
5

The information formats
The booklet
The seminar
The video

7
8
8
8

Improving knowledge
Who knows most?
Changes in knowledge
Whose knowledge improved?
Conclusion

9
11
11
15
15

Beliefs about sentencing aims and practice
Main aim of sentencing
How effective is sentencing at reducing crime?
How effective are specific sentences?
How well do sentences serve other aims?
Circumstances affecting sentencing
Sentencing practice in specific cases
Conclusion

17
17
19
20
22
22
24
26

iii

Improving public attitudes to the Criminal Justice System: The impact of information

Changing attitudes and confidence
Worry about victimisation
Is sentencing too lenient?
Confidence in the CJS
Conclusion

27
27
30
33
35

Evaluation of the information formats
Who took part?
Discussion of the information with others
Ease of understanding, general impressions, usefulness and enjoyment
Accuracy of the information
Learning and impact
How could the information sources be improved?
Conclusion

37
37
41
42
45
45
48
48

Discussion and conclusions
Using information to improve knowledge
Using information to change attitudes
The format of information
Next steps

49
49
50
50
51

Appendix A Methodology and statistical significance
(Questionnaire only available on website)

Appendix B

Booklet
(Only available on website)

Appendix C Additional tables

References

Publications

List of tables

3.1 Percentage of correct answers to the knowledge questions (a) that would be
expected by chance (b) for the general population sample (GPS) and (c) for
the project participants

3.2 Variations in mean knowledge score in the general population sample

3.3 Overall percentage improvement in proportion answering the questions correctly,
by information source

3.4 Indexed change in mean knowledge score, with positive scores indicating an
above average improvement, and negative scores a below average improvement

4.1 Percentage support for different ‘main aims’ of sentencing before and after
receiving information

4.2 Effectiveness of different sentences at reducing crime, before and after
receiving information

4.3 Percentage of people thinking different sentences very or fairly likely to
meet different aims of sentencing for a burglar and a robber

5.1 Change in punitiveness after receiving information

6.1 Variation in participation in the information stage of the research – percentage
of those asked to take part who were reinterviewed, by demographic group

6.2 Participant ratings of how easy the information sources were to understand

Improving public attitudes to the Criminal Justice System: The impact of information

List of figures

3.1

Change in mean knowledge score, by information source

4.1

Percentage of respondents saying sentencing has quite a lot or a great deal
of an impact on crime rates, before and after receiving information

Factors that should have a great deal or some influence on a sentence
(General Public Sample N=1022)

Percentage giving different sentences for a first time and repeat burglar,
before and after receiving information

Change in worry about being a victim by information format for those
answering the question both before and after

5.2

Change in worry about being a victim by original level of worry

5.3

Change in worry about being a victim of crime by change in knowledge

5.4

Change in opinion about sentencing of the courts, by original opinion

5.5

Percentage saying sentencing ‘about right’ before and after receiving
information, by change in knowledge

Percentage very or fairly confident in different aspects of the CJS, before
and after receiving information

Change in confidence that the CJS brings people who commit crimes to
justice, by information type

Change in confidence that the CJS is effective in bringing people who
commit crimes to justice, by initial level of confidence

4.2

4.3

5.1

5.6

5.7

5.8

5.9

Change in confidence that the CJS is effective in bringing people who
commit crimes to justice, by change in knowledge

6.1

Participation at different stages for the different information formats

6.2

Discussion of the information formats with others

6.3

Mean ratings of the three information formats on various dimensions

6.4

Parts of the information found most interesting for the different formats

6.5

Percentage who were or were not surprised by what they learnt who said
they learnt a lot, by information type

Percentage of participants who said they changed their views as a result
of what they learnt

6.6

Appendix tables

C4.1 Take up rates for the three information formats

C5.1 Percentage of respondents reporting reduced worry after receiving information
(not including those originally ‘not at all worried’)

C5.2 Percentage becoming less punitive after receiving information (only those
thinking sentencing a little or much too lenient in first survey)

C5.3 Percentage becoming more confident that the CJS brings people who commit
crimes to justice (not including those very confident in first survey)

vii

Improving public attitudes to the Criminal Justice System: The impact of information

viii

Summary

Building on previous research evidence that attitudes towards the criminal justice system
(CJS) are related to knowledge about the system, the aim of this work was to determine:
●
●

●

whether providing information has an effect on levels of public knowledge;
whether any improvements in knowledge that result have an impact on attitudes
and confidence in the CJS;
which of three methods of presentation of information would be the most efficient
and effective for imparting information to the public.

The research surveyed a nationally representative sample of 1022 people to assess levels of
knowledge about crime, sentencing and the CJS; attitudes to sentencing; and confidence in
the CJS. Of these, 220 then participated in an experiment to test the impact of providing
information.
Three formats were tested, all containing the same key facts about crime and the criminal
justice system:
●

●

109 people were given a 24-page booklet designed to be visually attractive and
easy to understand;
37 attended seminars involving presentations by experts, and question and
answer sessions; and
74 watched a video combining footage of one of the seminars with other visual
material.

The 220 were then reinterviewed to assess what impact participation had had on their
levels of knowledge and attitudes to the CJS.

Knowledge of crime and the criminal justice system
●

Overall knowledge about crime and the criminal justice system amongst the
general public is poor. There is particularly poor knowledge about crime trends
and current sentencing practice. In line with previous research, most people
believed crime was going up (over a period when it was falling), and the use of
custody in sentencing was considerably underestimated.
ix

Improving public attitudes to the Criminal Justice System: The impact of information

●

There was little systematic variation in knowledge by demographic characteristics,
although men and people of working age tend to be slightly more
knowledgeable. People who have had contact with the criminal justice system,
who are interested in law and order issues and those with some educational
qualifications also tend to be better informed.
After receiving information, participants increased the number of questions they
answered correctly. All three formats significantly improved knowledge scores.
There was some evidence that information that is surprising, or repeated, is more
likely to be recalled.
Those who had the lowest initial levels of knowledge were the most likely to
increase the number of questions they answered correctly after receiving
information. People with a demonstrated ability to learn and gain educational
qualifications performed better than average.

Sentencing aims and practices
●

●

Crime reduction was most frequently mentioned as the ‘most important’ aim of
sentencing. When asked to choose between seven aims of sentencing, ‘changing
behaviour to prevent re-offending’ was chosen by the largest number of people.
There were few changes in preferences for the main aim of sentencing after
receiving information. The largest shift was in the proportion of those who had
watched the video who thought punishment should be the main aim of sentencing,
but this was not statistically significant.
There was a widespread belief in the effectiveness of prison at reducing crime,
and in achieving other aims of sentencing. After receiving information, there were
some increases in support for alternative disposals, but the ratings for prison also
increased, and by a greater amount.
When sentencing a typical case, respondents’ use of custody for both burglars
and robbers was considerably lower than the rates of custody given by the courts
for these types of cases. Sentencing did not change very much as a result of
giving participants information.

Fear of crime, attitudes and confidence in the criminal justice system
●

A reduction in worry about victimisation was found for all information formats.
This effect was similar for most demographic groups. Those who were originally

Summary

●

‘very worried’ were the most likely to report a lower level of worry after receiving
information, as were those who had educational qualifications.
Providing information also had an effect on opinions about sentencing.
Respondents were less likely to think sentencing was too lenient after receiving
information. This effect was seen across all socio-demographic groups. Those who
had initially thought sentences were ‘much too lenient’ were most likely to change.
The shift in views was at least partly related to an improvement in knowledge.
When asked why they had changed their views, many participants said that the
information had prompted greater consideration of the issues, but others
attributed it to events in the media and personal experience.
There was also some impact on confidence in the criminal justice system. For
confidence that the criminal justice system brings people who commit crimes to
justice, those with poor confidence initially were most likely to improve in
confidence. Those who had initially said they were very interested in law and
order issues changed less than others.

Evaluation of the information sources
●

●

Some types of people were more likely to participate in the research. For
instance, those from minority ethnic communities and people with particularly
poor levels of knowledge were less likely to participate in any of the three
information groups. Those who attended the seminars were particularly
unrepresentative of the general population: the better qualified and those very
interested in law and order issues were most likely to attend. There was least
evidence of any bias in participation for the booklet group.
Generally, ratings of all the information sources were positive. All of the
information formats were thought to be informative, interesting, enjoyable and
helpful. The seminar was most positively rated on many of the evaluation
questions.
The booklet and video were both thought to be ‘modern’, and ‘different’ from
other government publications. Generally, the sources were thought to be credible
and accurate, although this was less so for the video. Respondents expressed
surprise at many of the facts presented, and these were the facts that were
particularly well recalled.
Many respondents from all the groups reported that they had changed their views
and that they were now more confident about the system, although the seminar
had less impact here.
xi

Improving public attitudes to the Criminal Justice System: The impact of information

Conclusion
●

●

xii

Providing simple factual information improved knowledge about crime and
sentencing, and also had an impact on attitudes to and confidence in the CJS,
although the evidence for a direct relationship between improvements in
knowledge and attitude change is less clear. All three of the information formats
tested improved knowledge and had some influence on attitudes.
Previous work to improve satisfaction and confidence in the criminal justice system
has tended to concentrate on targeting victims, witnesses and those in contact
with the system. The current research indicates that the widespread dissatisfaction
and misunderstanding of the system amongst the general public can also be
addressed.
The booklet was the most cost-effective format tested and reached the widest
cross-section of people. The booklet has therefore been updated and redesigned,
taking into account the comments from the participants in the research. It is likely
to form part of the strategy to improve confidence in the criminal justice system in
England and Wales.

Introduction

The research reported here was commissioned by the Home Office team who conducted the
2000/2001 review of the sentencing framework. In the main the purpose of the work was
to examine public attitudes to the current sentencing framework and to test out options for
change. Because previous work had shown the public to be very badly informed about
current sentencing practice, the review team was concerned about the value of public
opinion as a means of informing change, if that opinion was based on misperceptions. For
this reason it was decided to investigate the feasibility of improving public knowledge and
monitoring the impact this had on their views. Three methods of improving public
knowledge were tested: a booklet, a series of seminars, and a video. A full report of the
general public’s views on many aspects of the sentencing framework and options for
change, was published in Appendix 5 of ‘Making Punishments Work’, the report of the
review of the sentencing framework.1
A further impetus to this work was the government’s objective of an improvement by 2004
in the level of public confidence in the criminal justice system (CJS). Confidence in the
system is clearly important for maintaining public support for the rule of law. It is also key to
ensuring the public play their role in the process as witnesses or jurors and are willing to
participate as volunteers, for example as magistrates or on youth offending panels.
This report explores in greater depth the relationship between the format of information,
knowledge and attitude change. Specifically:
●
●

●

The analysis in ‘Making Punishments Work’ mainly compares the 1022 General Public Sample (GPS) and an
‘Informed Public Sample’ (IPS, N=92). The latter were those who answered seven or more questions correctly
in the second survey. No findings from the IPS are given in this report.

Improving public attitudes to the Criminal Justice System: The impact of information

Knowledge, attitudes about crime and sentencing, and confidence in the criminal
justice system
Public opinion about some aspects of the CJS is poor. Although a majority of people are
confident the CJS respects the rights of people accused of crime, less than half believe it is
effective in bringing offenders to justice, dealing with cases efficiently or meeting the needs
of victims. The courts and sentencers receive particularly low ratings (Mirrlees-Black, 2001).
This is at least partly because most people think that sentencing is too lenient.2 3
However this is not to say the public are more punitive than the judiciary. When asked to
select a sentence for a specific offence and offender, they tend to choose sentences which
are broadly in line with, if not more lenient than, those used by sentencers (Hough and
Roberts, 1998).4 It appears that the perception of lenience results from a lack of awareness
of the severity of the sentences given in court, and generally poor knowledge about criminal
justice system procedures.5 The general lack of confidence in the CJS also seems to be
affected by misperceptions of the crime problem. People tend to think crime is rising a lot
(even during periods when it is actually declining) and overestimate the proportion of crime
that is violent.
People who are better informed about crime and sentencing tend to rate the CJS more
highly. Hough and Roberts (1998) found that poor knowledge about crime and punishment
was associated with more negative ratings of courts and sentencers. Mattinson and MirrleesBlack (2000) confirmed a similar picture for perceptions of the youth justice system: those
with low levels of knowledge were more likely to think that sentencing in the youth justice
system was too lenient. And Mirrlees-Black (2001) found that misperceptions about crime
trends and sentencing practice were related to lower levels of confidence in the CJS.
In theory, therefore, improving public knowledge about crime, sentencing and the CJS might
be expected to result in more positive attitudes towards the CJS. Improvements in ratings of
2.

3.
4.

In 1996, 51 per cent of those interviewed in the British Crime Survey thought sentencing was much too lenient,
and 28 per cent a little too lenient. In 1998, the figures were 49 per cent and 30 per cent respectively.
The British Social Attitudes Survey has measured punitiveness since 1986. In this time, the proportion of
respondents agreeing strongly with the statement ‘People who break the law should be given stiffer sentences’
has varied, being generally between 27 per cent and 33 per cent in the late 80s, falling to 20 per cent in
1991, and rising to 41 per cent in 1993, before returning to similar levels in 1994 to 1996 as seen in 1986
to 1990.
A survey in Scotland produced similar findings: Anderson, Ingram & Hutton (2002).
This has been found for burglary by Hough & Roberts (1998) and Russell & Morgan (2001). Hough (1996)
found court of appeal judgements to be acceptable to the majority of groups of members of the public for
seven out of nine offences including burglary, actual bodily harm and shoplifting.
Lack of knowledge about sentences given by the courts has been found by a number of studies, including
Hough & Moxon (1985), Hough & Roberts (1998) and Mattinson and Mirrlees-Black (2000).

Introduction

the system should be achievable where current opinion is based on overly negative beliefs.
There is, however, little known about whether it is possible to influence knowledge, attitudes
and confidence in the CJS by providing information.

Can information have an impact on attitudes?
There is little research looking at whether information campaigns change public knowledge
and attitudes in the field of criminal justice. Most of what does exist relates to crime
prevention campaigns, which have a particular behavioural aim. However, some of these
also evaluate changes in knowledge and attitudes.
In the US, the National Crime Prevention Council evaluated their long running ‘McGruff’
crime prevention campaign. The NCPC found that a quarter of those interviewed reported
having learnt something new from the crime prevention campaign (O’Keefe et al, 1996). A
further half reported being reminded of things they had forgotten. Riley and Mayhew
(1980) found a high level of recognition of the slogans associated with press and television
campaigns aimed at promoting household security in the UK.
For such campaigns to change behaviour, the intended audience must be exposed to the
message and pay attention to it. Poor or passive attention paid to information will
undermine its effect (Parrott, 1995). Interest in and involvement with the subject matter is an
important factor in maintaining ‘active attention’ and processing of the information.
Parrott (1995) suggests that to promote active processing:
●
●

●

the presentation of the information should be unusual or unfamiliar;
there should be a discrepancy between expectations of the information and the
reality;
the information should request the audience to do something, as this forces an
increase in the level of attention.

There is also evidence that campaigns are more effective if they relate to subjects associated
with strong public opinion and concern, and topics on which there is generally a consensus
of opinion (O’Keefe et al, 1996).
The research reported here drew on some of these findings. The information formats were
designed to be attention grabbing and many of the facts presented were known to be
3

Improving public attitudes to the Criminal Justice System: The impact of information

surprising. There is no doubt that there is a widespread negative consensus of views on
sentencing in particular, and that interest in law and order issues tends to be quite high. In
the initial survey 37 per cent of the public said they were very interested, and a further 53
per cent were fairly interested in law and order issues.6

What sort of information might have an influence on knowledge and attitudes?
This research compared three different information formats: printed material in the form of a
booklet, a seminar event, and a video using footage of one of the seminars combined with
other images.
Audio-visual and printed sources are most commonly described as the main source of
information about crime and the criminal justice system. Three-quarters of people mention
the television and radio news. Television documentaries, local and tabloid newspapers are
mentioned by about half of respondents. Broad-sheet newspapers are an important source
for about a third (Mirrlees-Black, 2001).
Some previous work using a technique called deliberative polling was filmed and made into
a television programme by Channel 4. This suggested that face-to-face discussion was
influential in changing attitudes about crime. However, critics suggest that some of this
success may be due to demand characteristics. That is, where participants know what the
research is trying to achieve they may consciously or unconsciously comply with this (Ladd,
1996). This may be a particular issue where the event is to be televised, as was recognised
by the organisers.7 This is a criticism that could also be levelled at the research reported
here, although perhaps to a lesser extent as the research was not televised, and the purpose
was not made explicit. Other criticisms directed at similar deliberative polling exercises in
the US have included the questionnaire designs and the low response rates at follow-up
stages (Kay, 1996).
In the US, members of the public from Alabama and Delaware were more likely to advocate
alternatives to custody for non-violent offenders after seeing and discussing a video detailing
the problem of prison overcrowding and alternative programmes. The information provided
was specifically targeted at changing opinion on this particular issue (Doble, 1997; Doble
and Immerwahr, 1997).
6.
7.

General population sample N = 1022.
Fishkin refers to this effect in describing the UK criminal justice deliberative poll, "Knowing that they would be
on national television, they began discussing the topic with family and friends, they began to read newspapers
and listen to the media with more care." Fishkin (1996).

Introduction

There have been no previous attempts to evaluate the differential effects of alternative
information formats on attitudes to criminal justice. Psychological research has found most
support for the hypothesis that information in a printed form leads to better recall than that
in audio or audio-visual formats. Furnham, Gunter and Green (1990) review the literature
covering experiments that have used different types of information, including news and
advertisements. They report that the finding holds even when the amount of time spent
looking at or listening to the information is controlled for. It also holds whatever the mode of
questioning (written or verbal).

Structure of the report
The development of the information sources and the methodology of the research are
described in chapter 2. Chapter 3 discusses the improvements in knowledge found for the
three information formats. Chapter 4 looks at beliefs about sentencing aims and practice.
The impacts on attitudes and confidence in the system are described in chapter 5. Chapter
6 looks at the respondents’ evaluation of the information formats. Chapter 7 draws together
the findings and implications for policy in this area.

Improving public attitudes to the Criminal Justice System: The impact of information

The information formats

A randomly selected sample of 1022 members of the general public answered a
questionnaire about knowledge of crime and sentencing (see Appendix A). They were also
asked a range of questions to assess their attitudes to the CJS and specifically to
sentencing.8 Some of those interviewed were asked to take part in the further stages of the
research, and either received a booklet about crime and sentencing (109 people), attended
a seminar (37 people), or watched a video (74 people).9 Response rates for the different
stages are discussed in chapter 4. Low take up was an issue, particularly for the seminar.
This affects the ‘representativeness’ of each group and makes it more difficult to determine
whether measured differences between the groups over the course of the research can be
attributed to the relative impact of the information formats, rather than the initial profile of
the groups.
All the information formats included the same key items of information. These included the
correct answers to questions about crime and sentencing in the first stage questionnaire. The
facts were drawn from sources most widely accepted as reliable, including recorded crime
statistics, court proceeding statistics, and the British Crime Survey. After receiving the
information, the participants were re-interviewed.10 This design allowed an examination of
changes in knowledge and attitudes, and the relationship between these, if any. It also
allowed close control of the design of the information source materials and content. It was
important that the source materials had similar content so that differences between them
could be more confidently attributed to the format of the information, rather than differences
in content.

10.

The aim in the first stage was to obtain a representative sample of the general public aged 16 or over. The
sample was clustered because of the need for seminar participants to be within a reasonable distance of the
venue. Sampling points were selected systematically, stratified by region and social class. Respondents were
selected randomly, using systematic selection of sampling points stratified by region and social class. Postcode
sectors were then selected within the sampling points. The data were weighted to the population by region for
analysis.
Quotas were used to ensure that the groups receiving the different information sources were well balanced.
The participants received financial incentives to take part. Those in the seminar groups received £40, plus
travelling expenses. Those who read the booklet or watched the video received £20 each.
The interviews were conducted by a research company using Computer Assisted Personal Interviewing where
the questionnaire responses are entered directly on to a portable computer by the interviewer. See Appendix A
for a copy of the questionnaire.

Improving public attitudes to the Criminal Justice System: The impact of information

The booklet
The text of the booklet was designed to be simple and easy to understand, and was rather
more geared to the general public than most Home Office research and statistics
publications. Some charts and pictures were included. It was a square, CD size with 24
brightly coloured pages with a modern font. On many pages there were ‘call-outs’ where
short parts of the text were reproduced and enlarged. The information was presented in the
order of the criminal justice process: from crime levels, through policing to courts and
sentences. A list of further contacts for information was included at the back. The booklet
can be accessed at http://www.homeoffice.gov.uk/rds/horspubs1.html or copies can be
obtained from the Communication Development Unit, RDS, at the Home Office.

The seminar
Four seminars were held, one in each of London, Manchester, Cardiff and Birmingham.
Each started with a presentation by a senior academic. This covered the facts about crime
and sentencing included in the text of the booklet. Visual aids were used where
appropriate.11 The seminar continued with presentations from a local prison governor and
senior probation officer. These speakers detailed what an offender might experience in
prison and under probation service supervision respectively. All the speakers based their
presentations on a previously agreed script, to ensure reliability between the different
sessions. Those attending were then divided into groups to discuss the issues raised,
facilitated by researchers. They were asked to discuss the aims of sentencing, what they
found surprising in the presentations, and if anything said had changed their views. The
group then reconvened and questions were put to the academic expert and the prison and
probation representatives. Each seminar lasted about three hours.

The video
The London seminar was filmed, and this formed the basis of a 37-minute video, together
with footage of street scenes, police cars, courts, prisons and interviews with prisoners. A
professional voice-over introduced the topic and linked sections together. Parts of the
seminar question and answer session were also included.

11.

In most cases this was using OHPs, but in London, which was filmed for the video, handouts were used but
retrieved before the respondents left to avoid confusing the impact of the video and written material.

Improving knowledge

Knowledge of the criminal justice system was measured by 11 questions asked of the full
1022 sample and also in the follow-up questionnaire. Five of the questions asked about
criminal justice processes and the other six covered statistical information about crime and
sentencing. The format of questions varied from multiple choice to open response.
Table 3.1 lists the questions, the acceptable range of a correct answer, and the proportion
of both the general population and the project participants who answered correctly. Also
shown is the percentage that might be expected to get the question correct if answering by
just guessing, i.e. by chance. The first column gives an indication of levels of knowledge
amongst the general population. The second column shows that the 220 respondents who
participated in the information project had a similar pattern of knowledge but initially
scored slightly higher on most items.
Overall, knowledge of CJS processes and statistics is not good. The highest score achieved
was eight out of 11, and that by just two of the 1022 sample. The mean score was 3.6. The
expected average score if the questions had been answered entirely by guessing is 3.3.
Fifteen people failed to get any questions correct.
There was considerable variation in the proportion of correct answers for different
questions, but to some extent this reflects question format. For instance, although knowledge
of justice processes appears relatively good, with around a half of people answering these
questions correctly, their multiple choice format means there was between a 20 and 50 per
cent probability of getting them correct through chance.
Knowledge of crime trends and current sentencing practice is particularly poor, with only
about one in ten people being reasonably well informed in these areas. In line with previous
research (e.g. Hough and Roberts, 1998) the use of custody for burglary is considerably
under-estimated. Eight in ten respondents thought half or less of adult male burglars were
given custodial sentences, although the actual proportion for 1999 was 72 per cent. Over a
third of respondents thought the mandatory sentence for third time burglars was one year,
and one in five thought it was three months. This perceived lenience of the system extends
also to the length of custodial sentences with the vast majority of respondents underestimating the length of a sentence for an adult male rapist. Two-thirds of respondents
thought it would be five years or less, but in 1999 it averaged 8.5 years.
9

45
8

25
50
33.3
20
12.5
17
20

A half
TRUE
It makes no
difference
64 to 80
8 to 9 years
£21k to £31k
Less crime

6
11

47
44

64
47

n=1022

GPS

8
10

48
9

50
42

71
53

Participants
Before
n = 220

Percent correct

50
50

Expected
percent
correct
(due to chance)

Fine

FALSE
FALSE
Three year
prison
sentence

In a magistrates' court a jury decides whether someone is guilty or not.
In the Crown Court it is the jury who decide the sentence for an offender
What is the minimum sentence for an adult who has been convicted
three times of house burglary?

Which is the most common sentence given by the courts for all offences
(except motoring)?
Approximately how much of a prison sentence is spent in prison
(not including life sentences)?
Prisoners serving a 12-month or longer sentence will be supervised on release
An offender is least likely to get convicted again if he is given a prison
sentence, a community penalty, or does it makes no difference?
Out of 100 convicted adult male burglars (21 and over) how many go to prison?
An adult male (21 and over) convicted of rape will get an average
sentence length of?
Roughly how much does it cost to keep a prisoner in prison for a year?
What do you think has happened to the crime rate for the country
as a whole over the past two years?

Correct
answer

Percentage of correct answers to the knowledge questions (a) that would be expected by chance (b) for the general
population sample (GPS) (c) for the project participants

Question

Table 3.1:

Improving public attitudes to the Criminal Justice System: The impact of information

Improving knowledge

Who knows most?
There was little systematic variation in how poorly informed people were, though men did
slightly better than women and those under 65-years-old better than the over 65s.
Not surprisingly, those who have had contact with the criminal justice system have higher
levels of overall knowledge, as do those who are more interested in law and order issues.
People’s own assessment of their knowledge level was correlated to their actual scores, but
even those who described themselves as ‘very knowledgeable’ only had a mean score of 4.1.
Those with university level of education scored marginally higher, as did those in the AB
socio-economic bracket. There is of course considerable overlap between these two groups.
The statistical technique of logistic regression identifies those factors that are independently
related to having a better level of knowledge, taking account of these overlaps. This
indicates that contact with the CJS, being under 65 in age, of higher socio-economic class
and not being ‘very worried’ about crime are all independently related to scoring above
average.12 Level of education and interest in the subject matter are not.

Changes in knowledge
The information required to answer all the questions correctly was given in the booklet,
seminar and video, but even so, only one person got all 11 correct on the follow-up
questionnaire. However, there was a significant increase in correct answers for each of the
three groups (Figure 3.1). The video group showed the largest increase, with an average
improvement of 2.5 questions. This could be a memory effect as, on average, the video
group were re-interviewed within a shorter timeframe than those in the other two groups
(see chapter 6).

12.

The variables in Table 3.2 were used to model the likelihood of scoring four or more.

16 to 34
35 to 64
65 +
All

Higher
Further
GCSE / trade
No qualifications

AB
C1
C2
DE

Broad-sheet papers
Tabloids
Local papers
TV and radio

Female

Education

Class

News source

Source: General population sample: N=1022.

16 to 34
35 to 64
65 +
All

3.6
3.6
3.7
3.7

3.8
3.5
3.5
3.5

3.8
3.6
3.6
3.3

3.5
3.5
3.2
3.4

3.7
3.9
3.5
3.7

Mean score

3.6

3.3
3.6
3.7
3.5

Worry about being Very worried
the victim of crime Fairly worried
Not very worried
Not at all worried

ALL

4.1
3.7
3.5
3.0

3.7
3.6
2.9

How knowledgeable Very
are you?
Fairly
Not very
Not at all

Very interested
Fairly interested
Not interested

Attitudes to crime
and CJS

3.6
3.8
3.9
4.0

Mean score
Victim reporting crime
In court for any reason
In court as defendant
In prison or YOI for any reason

Contact with CJS

Variations in mean knowledge score in the general population sample

Male

Table 3.2:

Improving public attitudes to the Criminal Justice System: The impact of information

Improving knowledge

Figure 3.1:

Change in mean knowledge score, by information source

8
7
5.7

6
5
4

4.1

6.2

5.7
3.7

5.9
3.9

3.7

3
2
1
0

Booklet

Seminar
Before

Video

All (n=220)

After

Some types of information seemed to have more of an impact than others. There were
statistically significant increases in the proportion of people correctly answering questions
about crime trends, prison place costs, sentence length, and supervision of longer-term
prisoners on release (Table 3.3). There was less change in understanding of the role of the
jury, where knowledge was initially relatively good, or in the relative effect of prison and
community penalties on reconviction rates (the concept of ‘no difference’ apparently being
difficult to take in).13
There is some evidence that the more ‘surprising’ the information, the more it is
remembered. The annual cost of a prison place was frequently mentioned as surprising, and
this was the question that showed the most dramatic improvement.
Further evidence of what makes information memorable can be construed where the
booklet, video and seminar had a differential impact on increases in knowledge. For
instance, only the booklet group were more likely to identify the fine as the most common
sentence. This piece of information was given once in the video and (as far as we can tell)
the seminar, but was repeated three times within the text of the booklet. Repetition appears
to be effective then.
Only the seminar group showed much of an increase in knowledge about the proportion of a
custodial sentence spent in prison. Transcripts of seminar discussions were not kept, but it may
be that this was a point the presenters drew attention to, or was questioned by participants.
13.

When asked if there were any parts they did not understand, some of the booklet group indicated that they
had difficulty understanding the chart showing reconviction rates for community penalties and prison, because
the trends were so similar. This chart has not been included in the revised version of the booklet.

35
56
55
46
51
11
8
7

What is the minimum sentence for an adult male burglar convicted
three times for house burglary?

Which is the most common sentence given by the courts for all
offences (except motoring)?

Approximately how much of a prison sentence is spent in prison
(not including life sentences)?

Prisoners serving a 12-month or longer sentence will be supervised
on release.

An offender is least likely to get convicted again if he is given a
prison sentence or a community penalty?

Out of 100 convicted adult male burglars, how many go to prison?

An adult male (21 and over) convicted of rape will get an average
sentence length of?

Roughly how much does it cost to keep a prisoner in prison for a year?

39 *

50 *

29 *

26 *

64 *

68 *

Source: Participant sample. Booklet n = 109; Seminar = 37; Video = 74.
* = Statistically significant difference at the p<0.05 level. Other differences are not statistically significant.

In the Crown Court it is the jury who decide sentence for an offender.

What has happened to the crime rate for the country over the
past two years

Booklet
Before
After

30 *

59 *

24 *

57 *

81 *

49 *

Seminar
Before
After

41 *

76 *

55 *

28 *

70 *

64 *

81 *

Video
Before
After

Overall percentage improvement in proportion answering the question correctly, by information source

In a magistrates’ court a jury decides whether someone is guilty or not.

Table 3.3:

Improving public attitudes to the Criminal Justice System: The impact of information

Improving knowledge

Whose knowledge improved?
The largest increases in knowledge overall were amongst those who knew the least initially,
while the lowest were amongst those who knew the most initially (Table 3.4). This was
despite the fact that everyone had plenty of room for improvement, and could in principle
have improved their scores by as much as the poorest knowledge group.
Women improved slightly more than men, the young more than the older age groups.
Again, generally speaking those with educational qualifications, and thus experience of
learning (or scoring reasonably in exams at least) did best. The class variations will also
reflect this to some extent, as perhaps do the greater gains amongst the broadsheet
newspaper readership. Being interested in the topic didn’t seem to help information
retention, while worry about victimisation seems to act as a positive barrier.

Conclusion
Levels of knowledge are generally poor. This is widespread across all socio-demographic
groups and it is hard to identify any particular group that would particularly benefit from a
targeted information campaign. It is certainly important not to restrict information
dissemination to those who have contact with the criminal justice system.
Substantial proportions of all groups thought they had learnt a lot from participating in the
study. The seminar was rated most highly, with three-quarters saying they had learnt a lot,
followed by the video group. In fact, increased knowledge of crime and sentencing, or at
least recall of key facts, was achieved through all the information sources. The initial
variations in the composition of the three groups make it difficult to identify any real
differences between the formats in their effectiveness here. There were differences in the
type of information that was taken on board. In some cases this was probably due to the
way it was presented, but surprise is also clearly an important variable, with those facts that
were thought particularly surprising most likely to be recalled. Some types of people were
better at recalling the information they had been given. Perhaps not surprisingly, those with
academic qualifications were the most adept at this. But the greatest gains were amongst
those who scored lowest initially, whatever their level of education.

16 to 34
35 to 64
65 +
All

Higher
Further
GCSE / trade
No qualifications

AB
C1
C2
DE

2
3
4
5
6

Female

Education

Class

Initial level of
knowledge

160
40
-10
-80
-120

50
0
-50
-40

40
30
0
-100

50
10
-80
10

30
-40
-10
-20
30
-20
-10
-10

Very interested
Fairly interested
Not interested

ALL

Worry about being Very worried
-70
the victim of crime Fairly worried
-20
Not very or not at all worried 30

-30
20

-10
-10
60

Victim reporting crime
10
In court for any reason
20
In court as defendant
-60
In prison or YOI for any reason -60

Broadsheets
Tabloids
Local papers
TV and radio

How knowledgeable Very or fairly
are you?
Not very or not at all

Attitudes to crime
and CJS

Contact with CJS

News source

Indexed

Notes: Source participant sample n = 220.
Logistic regression indicates that all the characteristics shown are independently related to a change in knowledge level, except for news source and self-rated knowledge.

or less correct
correct
correct
correct
or more correct

16 to 34
35 to 64
65 +
All

Indexed

Indexed change in mean knowledge score, with positive scores indicating an above average improvement, and
negative scores a below average improvement

Male

Table 3.4:

Improving public attitudes to the Criminal Justice System: The impact of information

Beliefs about sentencing aims and practice

Public opinion about sentencing will, of course, be influenced by the purpose they assign it,
and the extent to which current practice is perceived as delivering this. Beliefs about the
purpose of sentencing were measured in the survey by questions asking generally about
what the main aims of sentencing should be, and a task of ranking seven possible aims. A
range of disposals were then assessed on how well they achieved these aims. A sentencing
task was also completed for two types of offence and offender.

Main aim of sentencing
When asked unprompted what the main purpose of sentencing should be, the most frequent
responses given were that it should stop re-offending, reduce crime, or create a safer
community. Only around 20 per cent spontaneously mentioned punishment, and 11 per cent
incapacitation. When asked to pick one of seven stated aims as the most important,
‘changing the behaviour or attitudes of an offender to prevent them re-offending’ was
chosen most often.14 A utilitarian belief in sentencing as a measure to reduce crime is widely
held, whether realistic or not. It is likely that people believe reduced offending will be a
consequence of ‘tougher sentencing’ and that this is in fact what ‘punishment’ is intended to
achieve, rather than being an aim in itself. Doble (1997) found that people in Alabama felt
the top priority for the criminal justice system was to protect people from becoming victims
of crime, rather than any of the aims of sentencing such as deterrence, incapacitation or just
deserts which professionals tend to think in terms of.
The purpose or aims of sentencing were not explicitly covered in the information formats. It
was stated that the sentence of the court depends largely on the seriousness of the crime.
Information was included on the effectiveness of disposals in reducing re-offending. There
was no direct reference to ‘just deserts’. The choice of aims after receiving information
changed only slightly. The video group showed the biggest shift, away from changing
behaviour, to punishment, but this change was not statistically significant (Table 4.1).

14.

In the first stage, general public survey, 48 per cent of people chose this option.

10
6
5
2

Punish

Scare the offender

Restrict opportunities to re-offend

Deter others from committing the
same crime

Make amends to the victim for
harm done

Express society’s disapproval

Before
%

Booklet

Source: General Public Sample (N=1022) and Participant Sample (N=220)

Change behaviour/attitudes

GPS

After
%

Seminar
Before
After
%
%

Before
%

Percentage support for different ‘main aims’ of sentencing before and after receiving information

Main Aim of Sentencing

Table 4.1:
Video

After
%

Improving public attitudes to the Criminal Justice System: The impact of information

Beliefs about sentencing aims and practice

How effective is sentencing at reducing crime?
The public clearly believe, or want to believe, that sentencing has some impact on crime
reduction and community safety. The extent to which this is realistic is probably fairly limited
in practice (see Halliday 2001, Appendix 6 for a review of the evidence). Some might
argue, therefore, that it is unwise to raise public expectations too far in this regard. On the
other hand, improved confidence in the system does appear to be predicated on a belief
that sentencing does have an impact on crime. Most people believe that the sentences given
by the courts currently have an impact on crime, although for the majority this is only ‘a
little’ impact (60%) rather than ‘quite a lot’ or ‘a great deal’ (31%). The greater the
perceived impact, the more confidence that the system was effective at bringing offenders to
justice.15

Impact of information
Despite the information they were given on levels of re-conviction, the proportion of
participants thinking sentencing had ‘quite a lot’ or a ‘great deal’ of impact on crime
increased from 31 per cent to 42 per cent. The booklet and video groups’ views changed
more than those of the seminar group (Figure 4.1). Only the increase for the booklet group
is statistically significant.

Percentage saying a great deal
or quite a lot of impact

Figure 4.1:

Percentage of respondents saying sentencing has quite a lot or a great deal
of an impact on crime rates, before and after receiving information
60
49

50
34

20
10
0

Booklet

Seminar
Before

15.

Video

All

After

On the before survey, 51 per cent of those who thought it had a great deal or quite a lot of impact were
confident the system was effective at bringing offenders to justice, compared to 38 per cent of those who
thought it had a little impact, and 23 per cent of those who thought it had no impact.

Improving public attitudes to the Criminal Justice System: The impact of information

How effective are specific sentences?
Different types of disposal varied considerably in the extent to which they were seen to be
effective in reducing crime (Table 4.2).16 Prison was thought to be the most effective in this
respect, with 38 per cent of the general public thinking prison was ‘very effective’ at
reducing crime, and a further 47 per cent thinking it was ‘fairly effective’. A similar
proportion thought prison combined with supervision in the community was effective, but the
percentage thinking it was ‘very effective’ was lower.
The other established disposals, such as fines, probation orders and community service orders
were less well rated – between 35 per cent and 49 per cent thought these disposals were very
or fairly effective at reducing crime, but few of these thought they were ‘very effective’.17
The concept of offenders compensating victims also received high ratings, an encouraging
finding for restorative justice schemes and some of the recent youth justice reforms. The
belief in drug treatment schemes no doubt reflects the finding that a third of the general
public identify drugs as the main cause of crime (Mirrlees-Black 2001).
There is a considerable difference, then, between the way prison and other sentences are
perceived in relation to crime reduction. This is not really surprising given its long history in
the criminal justice system and as the pinnacle of the sentencing framework since the
abolition of the death penalty. Nevertheless, many respondents also see the benefits of
combining prison with supervision in the community. Some see this as preferable to a
sentence just of imprisonment, even if the total length is the same: 72 per cent of the general
public thought a sentence of three months custody followed by three months supervision
would be more constructive than six months in prison.

Impact of information
After receiving information, participants tended to rate sentences as more effective at
reducing crime (Table 4.2). However, for all disposals apart from prison and ‘prison and
supervision’, the proportion saying they were very effective did not change much at all: the
increases were in the proportion saying they were fairly effective. In contrast, for prison and
‘prison and supervision’ the percentage thinking they were very effective at reducing crime
rose: for prison from 34 per cent to 43 per cent, and for ‘prison and supervision’ from 25
per cent to 33 per cent. There was little change in the proportion saying they were fairly
effective, mainly because there were so few people thinking prison was not effective initially.
16.
17.

A sub-set of disposals were also rated in respect of some of the other aims of sentencing. See next section.
Probation Orders are now known as ‘Community Rehabilitation Orders’ and Community Service Orders as
‘Community Punishment Orders’.

%
47
60
44
49
45
38
45
36

%
38
25
24
19
10
5
4
4

%
34
25
23
21
7
5
4
2

%
49
62
41
46
48
37
40
35

Participants before
information
Very
Fairly
effective
effective
%
43
33
19
24
5
4
2
3

%
79
46
34

%
74
53
48

%
78
54
55

burglar
%
85
45
45

robber

Punish

%
87
74
40

%
87
69
32

Restrict
opportunities
to re-offend
burglar robber

%
77
47
44

burglar

%
81
43
38

robber

Deter
others

%
71
25
42

burglar

%
76
21
35

robber

Make amends
to the victim

Source: General Public Sample (N=1022). Respondents were asked to rate each of the three sentences as ‘very likely’, ‘fairly likely’, ‘not very likely’ or ‘not at all
likely’ to meet each aim.

robber

burglar

Change
behaviour/attitudes

Percentage of people thinking different sentences very or fairly likely to meet different aims of sentencing for a
burglar and a robber.

Imprisonment
Electronic tagging
Community Service

Table 4.3:

%
48
60
50
53
46
44
59
43

Participants after
information
Very
Fairly
effective
effective

Source: General Public Sample, N=1022 and Participant sample (includes only those who answered the question both before and after) Minimum N=173

Fairly
effective

Very
effective

General Public Survey

Effectiveness of different sentences at reducing crime, before and after receiving information.

Prison
Prison and supervision in the community
Offenders compensating and making amends
Drug Treatment Schemes
Electronic Tagging
Fines
Probation Orders
Community Service Orders

Table 4.2:

Beliefs about sentencing aims and practice

Improving public attitudes to the Criminal Justice System: The impact of information

How well do sentences serve other aims?
One criticism of asking the public to state their preferred aim of sentencing is that this is
likely to vary according to the nature of the offence and the offender. To address this, the
surveys included details of two offenders and offences. The first was John, a burglar, the
second, Mike, a robber. Respondents were first asked to select an appropriate sentence
assuming that John or Mike were first-time adult offenders, with multiple choice allowed. This
exercise was then repeated on the assumption that they were now repeat offenders. They
were then asked how likely they thought it was that each of prison, a community service
order and electronic tagging, would meet different aims of sentencing for this particular
offender (as a recidivist).18
The belief in prison as an effective method for reducing crime is also seen here. Prison was
judged as very likely to punish, deter others, make amends to the victim, restrict re-offending
and change behaviour, much more so than community service or electronic tagging.
Support for electronic tagging and community service is around 40 to 50 per cent as a
method of punishing, deterring others and changing behaviour in the case of burglary, and
generally slightly lower for robbery (Table 4.3). Neither type of disposal was seen as effective
in making amends to the victim, particularly electronic tagging, with only 25 per cent believing
tagging could achieve this in the case of the burglar. Tagging receives particularly high
ratings, though, as a method of restricting offenders’ opportunities to re-offend: 74 per cent
thought it would be effective in this respect for a burglar and 69 per cent for a robber.

Impact of information
As the sample was split for these questions, the small numbers involved in some of the
information groups are too small to look at change in response on these questions after
receiving information.

Circumstances affecting sentencing
When considering factors to be taken into account when sentencing other than offence type
or seriousness, the public felt that previous convictions and the likelihood of re-offending
were very important: 99 per cent thought previous convictions should have some influence,
with 86 per cent saying it should have a great deal of influence.
18.

Although it wasn’t explicitly stated that the offender was a recidivist, the questions followed on from this scenario.

Beliefs about sentencing aims and practice

Other factors to do with the offence and offending were also thought to be important. 54
per cent of respondents thought a great deal of notice should be taken of the effect of the
crime on the victim, although the number thinking a great deal of notice should be taken of
the wishes of the victim was lower (25%).
Personal circumstances, such as domestic responsibilities, age and employment status were
not generally felt to be as important. Only 20 per cent thought a great deal of notice should
be taken of domestic responsibilities.

Figure 4.2:

Factors that should have a great deal or some influence on a sentence
(General Public Sample N=1022)19

The number of times the
offender has offended before
The likelihood that the
offender will reoffend

Whether the crime was planned
The effect on the victim
Whether the offender was mentally
ill at the time of the crime
Whether the particular type of crime is
occurring frequently within the local area

Whether the offender has made
amends to the victim

The wishes of the victim

The cost of the sentence
to the tax payer

Public opinion about the offence

The offenders' domestic
responsibilities

The age of an adult offender
Whether the offender is
employed or not
0

87
84
71

46
31

44
35

11
20

93
95

61
47

100

120

Percentage
Great deal of notice

19.

Some notice

The individual values for a great deal and some influence may not add to the total given due to rounding.

Improving public attitudes to the Criminal Justice System: The impact of information

Impact of information
For the booklet and seminar groups, there were no significant changes in the importance
respondents attached to these factors. After seeing the video, this group were more likely to
think that the likelihood of re-offending, the number of times the offender has offended
before and the age of the offender should be taken into account in sentencing. However
they were less likely to think that sentencing should take notice of the wishes of the victim.

Sentencing practice in specific cases
Despite the confidence in prison to meet a wide range of sentencing aims, there was
evidence that people would rather reserve it for those they consider to be the most serious
criminals. When sentencing the two offenders in the scenarios as first-time offenders, use of
prison was very low.20 John the burglar was given prison as part of his sentence by only 9
per cent of respondents. 21 per cent of people said they would have sent Mike the robber
to prison. Even when they became repeat offenders, only 56 per cent opted to give John a
prison sentence, although 74 per cent now felt this was appropriate for Mike.21 The rates of
use of custody are much lower than would be received by offenders committing these sorts
of crimes.22
Other than the financial cost of prison – a fact that many participants expressed surprise at –
other costs are also recognised. There is awareness, for instance, of the potentially
damaging effects of prison, and this may partly explain the restraint in the use of prison
shown above. Two-thirds (68%) of the general public strongly or moderately agreed that
‘offenders come out of prison worse than they go in’. It may be that when answering this
question, they are reminded of the common perception that prison is a ‘university of crime’
but it could also be that they are expressing concern about perceived easy conditions or
regimes in prisons.

Impact of information
Certainly the effect of giving participants information about prison regimes was an
improvement in confidence in the efficacy of prison. The proportion agreeing that ‘offenders
come out of prison worse than they go in’ fell from 72 per cent to 59 per cent. Participants
20.

21.
22.

Respondents were not required to recall what sentences were available; rather they were asked, for each
sentence, whether they would use it for the offender. These questions followed other discussions about which
sentences respondents were aware of, both unprompted and prompted, so by this stage of the interview they
had some familiarity with the sentences available.
Findings from General Public Survey of 1022 people.
75 per cent of burglars are given custodial sentences, as are 91 per cent of those convicted of robbery.

Beliefs about sentencing aims and practice

generally became more supportive and understanding of prison procedures. For instance,
they were more likely to agree that there is a need for incentives for good behaviour and
that prisoners need support in preparing them for release.
As far as the choice of sentences was concerned, the use of prison for John and Mike
remained about the same (Figure 4.3).23 There were also few changes in the use of other
sentences. Fines were used a little more often for the burglar. Previous research such as
Hough and Roberts (1999) and Doble (1997) has found that increasing information about
the availability of alternatives leads to a greater use of alternatives to custody. Given that in
the current research participants sentencing preferences were already more lenient, if
anything, than judicial practice, and that the respondents had already been reminded of
available sentences, this lack of change is not surprising.

Figure 4.3:

Percentage giving different sentences for a first-time and repeat burglar,
before and after receiving information.24

Percentage using sentence

50
41
32

30
20
10
0

23.
24.

40
28

1718

21
17

9
3

First-time burglar - Repeat burglar - First-time burglar before information before information after information

Repeat burglar after information

Conditional discharge

Fine

Probation

Community service order

Imprisonment

Electronic tagging

The patterns of use of sentences for those taking part in the second stage, before receiving information, are
very similar to those reported above for the general public survey.
Participants could choose as many different sentences in combination as they wanted. This included
conditional discharge and some participants chose this as well as other sentences. Therefore the number
shown here choosing conditional discharge should not be interpreted as the number giving this and no other
penalty.

Improving public attitudes to the Criminal Justice System: The impact of information

Conclusion
Receiving information did not change participants’ views about sentencing aims and
practices very much. Crime reduction and changing behaviour were identified as the most
important aims of sentencing both before and after receiving information.
Prison was widely believed to be effective at reducing crime and in achieving other aims of
sentencing. Receiving information led to an increase in support for alternatives, but also to a
greater increase in support for prison.
Use of custody by respondents in particular cases was considerably lower than the rates
given by the courts for these types of cases. Providing information did not change this.

Changes in attitudes and confidence

This chapter evaluates the effect of the booklet, seminar and video on three key attitude
measures: worry about victimisation, perceived lenience of sentencing, and confidence in
the criminal justice system. For each of the measures, the characteristics of those people who
were more likely to change their opinions are identified. As discussed in chapter 1, the
known relationship between levels of knowledge about crime and the CJS and attitudes to
the CJS suggested that giving people information would have a positive impact on their
attitudes. This chapter considers the extent to which changes in attitude can be attributed to
improved levels of knowledge.

Worry about victimisation
Worry about victimisation has been associated with poor knowledge, particularly in regard
to crime trends and risks of victimisation. All of the information formats included a statement
that according to police and BCS figures, there were now fewer crimes than in the early
nineties. Each source also gave the relative risks of being attacked by strangers for people
of different ages and gender. This showed the low risk to the elderly and women in
comparison with young men.

Overall effects on worry about being a victim
Overall, there was a net reduction in worry amongst participants. This effect was mixed,
though. Nearly a quarter of people reported reduced worry, but 15 per cent had an
increased level of worry about becoming a victim. 38 per cent of the seminar group
reported reduced worry, more than for the other information format groups (Figure 5.1).

Improving public attitudes to the Criminal Justice System: The impact of information

Figure 5.1:

Change in worry about being a victim by information format for those
answering the question both before and after
80
68

70
Percentage

40
30
20

10
0

Booklet

Seminar

Video

No change

More worry

All
Less worry

Initial level of worry was highly related to the likelihood of becoming less worried: 54 per
cent of those who were initially ‘very’ worried became less worried, compared with 28 per
cent of those who were ‘fairly’ worried and 9 per cent of those who were ‘not very’
worried. (Figure 5.2). Although over two-thirds of those who were ‘not at all worried’ before
receiving information increased their level of worry, this was only slightly, to ‘not very
worried’, and there were only seven people in this group.

Percentage
Percentage

Figure 5.2:

80
80
70
70
60
60
50
50
40
40
30
30
20
20
10
10
0
0

Change in worry about being a victim by original level of worry

46
46

54
54

28
28
8
8

71
71

66
66

64
64

29
29

25
25
9
9

very worried
fairly worried not very worried not at all worried
very worried
fairly worried not very worried not at all worried
(n=37)
(n=87)
(n=89)
(n=7)
(n=37)
(n=87)
(n=89)
(n=7)
No change
Less worry
More worry
No change
Less worry
More worry

Changes in attitudes and confidence

Who becomes less worried?
Of course the seven who initially said they were ‘not at all worried’ couldn’t become less
worried on the measure used. To explore the characteristics of those who became less
worried these people were therefore excluded from the analysis.
Logistic regression analysis confirms that initial level of worry is associated with a reduction
in worry about being a victim of crime, even when the impact of other characteristics, such
as age and sex are held constant. Also associated with a reduction in worry is having
educational qualifications.
The format in which information was received did not reach statistical significance in the
regression model. Those who were very interested in law and order issues were also more
likely to get less worried, but again this was not independent of other factors.

Knowledge and fear
Figure 5.3:

Change in worry about being a victim of crime by change in knowledge
80

Percentage

50
40

30
20

16
9

10
0

worse or
no change

one or two
more correct

More worry

three more
correct

No change

22
15

4+ more
correct

Less worry

There was no evidence that an increase in knowledge was associated with a decrease in
worry about victimisation (Figure 5.3). The logistic regression analysis found no
independent effect of an increase in knowledge on worry. The ‘very worried’ were the most
likely to become less worried but as shown in chapter 3, the initially very worried were
among those who showed least improvement in knowledge scores. Participating in the
project does seem to have had a positive effect on levels of worry, but it is not possible to
identify why this is. It could, perhaps, be the experience of being involved in the project or
an improved awareness of crime and the CJS that the questionnaires failed to detect.
29

Improving public attitudes to the Criminal Justice System: The impact of information

Is sentencing too lenient?
The question ‘in general would you say the sentences handed down by the courts are too
tough, about right or too lenient?’ has been used as an indicator of punitiveness, but is also
useful as a barometer of how appropriate sentencing is seen to be, in comparison with
expectations.

Overall effects on views of sentencing
The proportion thinking sentencing was about right increased from 21 per cent before
receiving information to 31 per cent afterwards. The increase was largest for the booklet
group. Very few people (3%) thought sentencing was too tough, either before or after
receiving information.
Overall, over a quarter of participants became less punitive after receiving information
(Table 5.1). The proportion for the booklet is a little higher than those for the seminar and
video groups, but the difference between the groups is not statistically significant.

Table 5.1:

Change in punitiveness after receiving information

More punitive
No change
Less punitive
N=

Booklet
%

Seminar
%

Video
%

All
%

14
55
31
105

23
51
26
35

17
58
26
66

17
55
28
206

Looking just at those who answered the question both before and after receiving
information, those who thought sentencing was much too lenient before receiving
information were more likely to change their opinion and become less punitive than those
who thought it was a little too lenient (Figure 5.4). About a third of those who thought
sentencing was ‘about right’ before became more punitive. However, for all but one this
was only a change to ‘a little too lenient’.

Changes in attitudes and confidence

Figure 5.4:

Change in opinion about sentencing of the courts, by original opinion.
70

Percentage

60
50

53
47

43
29

30
20

10
0

Much too
lenient (n=57)

A little too
lenient (n=98)

More punitive

About right
(n=44)

No change

Too tough
(n=7)

Less punitive

Participants who changed their opinion were asked why this was.25 The answers included
many explicit references to the information sources and how these had raised their
awareness. There were also specific mentions of sentence lengths and the percentages of
offenders given custodial sentences. The formats presented this information as three ‘case
studies’ of offenders, their offences and the sentence they would expect to receive – both
their likelihood of imprisonment and the average sentence for that offence.26 One of the
participants from the video group said: ‘they were much tougher than I thought’. One of the
booklet group said they had now given more thought to the question.
But not all those who had changed their opinion credited this to the information sources.
Some made reference to the news, saying they had seen particular cases which had
changed their views. Some participants thought the actual levels of sentencing had
improved since the first interview. Others cited personal experience, either as a victim, or as
an offender, or knowing an offender. Nevertheless there was some evidence of an indirect
influence, as some felt taking part in the research had affected the way they approached the
issues: ‘I’ve been reading and my general impression now is it’s about right’ and ‘Because I
have been thinking more deeply about the problem since the first interview’.

25.
26.

Second stage interviews were matched in the CAPI program to some responses from the first stage. If the
interviewee gave a different answer, they were asked: ‘you said x before, why has your opinion changed?’.
The case studies involved John, who was a burglar, Mike who committed a robbery, and Frank who was a rapist.

Improving public attitudes to the Criminal Justice System: The impact of information

Factors influencing views of sentencing
To identify the factors related to shifts in opinion towards believing sentencing practice is
‘about right’ logistic regression analysis was used, including just the 72 per cent (N=159) of
participants who initially said sentences were ‘a little’ or ‘much’ too lenient. The small
minority who initially thought it was ‘too tough’ were excluded as different factors are likely
to be relevant for this group. The regression identified initial strength of belief, having
educational qualifications, and improvement in knowledge after receiving information as
independently predictive of shifts in opinion.

Effects of knowledge
Looking more specifically at responses of ‘about right’ before and after receiving
information confirms that increased knowledge does seem to be a relevant factor (Figure
5.5). The proportion saying sentencing was ‘about right’ increased overall, but those
improving their knowledge by one or two correct answers, or by four or more, showed the
greatest increase.

Percentage saying
sentencing ‘about right’

Figure 5.5:

Percentage saying sentencing ‘about right’, before and after receiving
information, by change in knowledge
45
40
35
30
25
20
15
10
5
0

39
34
30
19
14

stayed the same
or less (n=58)

increased by
1 or 2 (n=65)
Before

increased by
3 (n=31)
After

increased by
4 or more (n=59)

Changes in attitudes and confidence

Confidence in the CJS
Overall changes in confidence
There were improvements in confidence for all three of the confidence indicators but that for
confidence that the CJS meets the needs of victims did not reach statistical significance.27 The
greatest gains were in the number of participants who were confident that the CJS brings
people who commit crimes to justice which increased after receiving information, from 38
per cent to 60 per cent (Figure 5.6).28

Percentage very
or fairly confident

Figure 5.6:

Percentage very or fairly confident in different aspects of the CJS, before
and after receiving information

100
90
80
70
60
50
40
30
20
10
0

86
78
60

CJS meets the needs
of victims of crime

CJS is effective in
CJS respects the rights
bringing people who of people accused of
commit crimes to justice committing a crime
and treats them fairly.
Before

After

Figure 5.7 shows the proportion who decreased their confidence in this respect, stayed the
same, or became more confident, by information type. Only the booklet and the video had
statistically significant positive effects on confidence on this measure (although this may be
due to the small sample size for the seminar). Very few people were less confident after
taking part, and in each group at least 30 per cent were more confident.

27.
28.

Comparisons for those who answered the questions both before and after receiving information.
The fourth aspect of the CJS confidence is measured on as part of the CJS objective, ‘deals with cases
promptly and efficiently’ was not included in the survey due to lack of space.

Improving public attitudes to the Criminal Justice System: The impact of information

Percentage saying
Percentage
sentencing ‘about right’

Figure 5.7:

Change in confidence that the CJS brings people who commit crimes to
justice by information type
45
70
40
60
35
50
30
25
40
20
30
15
20
10
105
00

34
30

40
32
19
14

stayed
the same
Booklet
or less (n=58)

Decreased confidence

increased
Seminar by
1 or 2 (n=65)

increased
Video by
3 (n=31)

Stayed the same
After
Before

increased
by
All
4 or more (n=59)

Increased confidence

Those who were not at all confident before receiving information were most likely to become
more confident that the criminal justice system brings people who commit crimes to justice
(Figure 5.8).

Percentage saying
Percentage
sentencing ‘about right’

Figure 5.8:

Change in confidence that the CJS is effective in bringing people who
commit crimes to justice, by initial level of confidence
45
90
40
80
70
35
60
30
50
25
40
20
30
15
20
10
105
00

15
6

stayed
the same
by confident
increased by
increased
very
or fairly
confidentincreased
not very
not at all
confidentby
or less(n=84)
(n=58)
1 or 2 (n=65)
3
(n=31)
4
or
more (n=59)
(n=108)
(n=27)

Decreased confidence

Stayed the same
After
Before

Increased confidence

Factors related to becoming more confident
Improvement in confidence on this measure could only occur for the 98 per cent (N=216)
who were not ‘very confident’ in the first stage survey. Logistic regression analysis shows
that interest in law and order and original level of confidence are independently associated
34

Changes in attitudes and confidence

with change in confidence. Those who said they were very interested in law and order were
less likely to improve in confidence. Those who were not at all confident were the most likely
to improve in confidence.

Knowledge and confidence
Although participation in the project clearly had some impact on public confidence, there
was no clear relationship between improved scores on the knowledge questions and
improved levels of confidence (Figure 5.9).

Percentage
saying
Percentage
sentencing ‘about right’

Figure 5.9:

Change in confidence that the CJS is effective in bringing people who
commit crimes to justice, by change in knowledge

70
45
40
60
35
50
30
40
25
20
30
15
20
10
105
00

34
30
37
31
19

30
35

14
14

stayed
worsethe
orsame
or
(n=58)
no less
change

Decreased in confidence

increased
one
or two by
1 or 2
(n=65)
more
correct

increased
three by
3 (n=31)
more
correct

Stayed the same
After
Before

increased
by
4+
4more
or more
(n=59)
correct

Increased in confidence

Conclusion
The small sample sizes involved make detailed analysis of patterns of change in attitudes
and confidence difficult. However, it is possible to say that the information sources did have
positive effects on fear of crime, on perceptions of sentencing practice, and in confidence
that the criminal justice system brings people who commit crimes to justice.
There seems to be little systematic variation in improved attitudes according to sociodemographic characteristics. For all three attitudes, the most predictive factor is original level of
attitude: those with the most extreme attitudes are most likely to change their opinions. Interest in
law and order issues and levels of educational qualification also seem to be related to change in
some of the attitudes. Who is susceptible to changing attitudes when presented with information
may be partly dependent on other factors which were not measured, such as learning style.
35

Improving public attitudes to the Criminal Justice System: The impact of information

Of these three attitude measures, improved knowledge scores were only found to be related
to changes in judgements about the appropriateness of sentencing practice. However, the
changes in fear and confidence may also have been brought about by the act of being
given information, but not so directly related to improved recall of specific facts.

Evaluation of the information formats

Chapter 3 discussed the effectiveness of the three information formats in improving
knowledge and chapters 4 and 5 their impact on attitudes. This chapter considers the extent
to which each of the formats successfully reached its target audience, and how well they
were received by those that read, attended or watched them.

Who took part?
Although every effort was made to ensure representative samples were offered each
information type, there were some considerable differences in rates of participation. There
were two main points at which people had the opportunity not to participate: the first when
they were asked to take part in the second stage of the research, and the second at the
actual point of participation. Reinterview could also ‘lose’ participants, either because they
refused (which might be associated with not having looked at the information), or because
there was not enough time to contact them. The video group, for whom the window for
reinterview was shortest, lost the most participants at this final stage.29
This section looks at the change in participation rates between being asked to take part,
and being reinterviewed at the second survey stage. There are limitations to this, including
that it does not allow identification of loss of participants due to other circumstances, such
as the time available for reinterview, as discussed above. It also does not show any
differences that occurred at the initial selection stage (such as who refused to be interviewed
in the first survey).

29.

The time between receiving the information and the follow-up interview was one to two weeks for the video
group, compared with between two and six weeks for the booklet group and one and six weeks for the
seminar group.

Improving public attitudes to the Criminal Justice System: The impact of information

Figure 6.1:

Participation at different stages for the different information formats
0

100

Number of people
150
200

250

300

350

195
141

booklet

113
109
319

seminar

116
37
37
188

video

139
77
74

asked to take part

had read/watched/attended

agreed to take part

recontacted

Take-up rates were quite varied for the different information formats (Figure 6.1).30 The
booklet had an agreement rate of 72 per cent, and a recontact rate of 80 per cent. The
initial agreement rate for the seminar was much lower at 36 per cent, as was the
percentage recontacted at 32 per cent (this is mainly due to the low numbers attending the
sessions, rather than problems with recontact. Thirty-eight people attended the seminars and
37 were reinterviewed).31 The video group had a reasonably good agreement rate of 74
per cent, but only 55 per cent of these were then recontacted. This is a particularly low rate,
perhaps because the video was the last intervention to take place and the period available
for reinterview was therefore the shortest. Low take-up rates, particularly for the seminar,
affect the ‘representativeness’ of each group. This makes it difficult to determine whether the
differences in outcomes between the groups are due to the impact of each information
format, or the different make-up of the groups.
Table 6.1 shows the percentages of those who were asked to take part who were
reinterviewed, by demographic group, for each information type.

30.
31.

Table C4.1 in Appendix C shows the take up and participation rates in more detail.
Because the number in this group is quite small, this makes it less likely that changes in variables reach
statistical significance. It should be borne in mind, therefore, that more statistically significant changes may
have been found with a larger group.

10
14
8
13
4
19
12
9
9
12
12
18
11
14
6*
16 *
9

66
60
38

60
25*

66
52
54
58

67
54

60
52
68
48

40
31
48
41

41
41

38
38
47
43

44
0*

40
41
43

40
42

10
15
7
14
10
15
11
17
11
14
20 *
6
6*
16
9
14
11
12
15
8
5
14
12

49
62
64
51
57
58
80*
55
72
56
60
57
61
55
63
49
61
46
65*
50
38
63*
58

29
45*
41

30*
47
42
41
42
41
41
43
43
39

46
33
46
38
56
40
57
39

44
40

% of those asked in each
demographic group who participated
Booklet
Seminar
Video
Think sentencing is
Much too lenient
Little too lenient or less
Contact with the CJS
Victim of crime
Not a victim
Been in court
Not been in court
Been in court as accused
Not in court as accused
Been in a prison
Not been in a prison
News source
Broadsheet
Not broadsheet
Tabloid
Not tabloid
local paper
Not local paper
TV news
Not tv news
Has access to internet
No access to internet
Knowledge about the CJS
2 or less correct
3 or more correct
All

* Rate of participation significantly different than rate for other demographic groups at p<0.05 level

12
11

61
56

% of those asked in each
demographic group who participated
Booklet
Seminar
Video

Variations in participation in the information stage of the research – percentage of those asked to take part who
were reinterviewed, by demographic group

Sex
Male
Female
Age
16-34
35-64
65+
Ethnicity
White
Non-white
Class
AB
C1
C2
DE
Children in household
Yes
No
Qualifications
Degree or higher
A levels or diploma
GCSE or trade
Other or none
Interest in law and order
Very interested
Fairly interested or
not interested

Table 6.1:

Evaluation of the information formats

Improving public attitudes to the Criminal Justice System: The impact of information

Two effects can be seen across all three of the information formats. Those from minority
ethnic communities are less likely to take part, as are those with lower levels of existing
knowledge.32 The lower rates of participation for those from minority ethnic communities
reflect findings from other research. For example, the British Crime Survey 2000 had a
response rate for these groups of 58 per cent, compared with 74 per cent for whites. Lower
participation rates from those with lower existing knowledge is disappointing, if this means
that those who could most benefit from information do not receive it.
There were also some effects for the particular information formats. For the booklet, those
more likely to take part were those:
●
●

who have access to the internet; or
who had previously been in court accused of a crime.

Of the people asked to take part in the seminar exercise, those who:
●

●

had read a broadsheet newspaper in the last week or are very interested in law
and order issues were more likely to take part;
had read a tabloid in the past week were less likely to take part.

Of those asked to watch the video, those who had:
●

read a broadsheet newspaper in the past week were less likely to take part.

When asked why they would not be willing to take part, the largest number of people said
they were too busy, did not have enough time or cited other commitments. Others said they
were just not interested. Being away on holiday or moving out of the area were other
common reasons given for not taking part in the next stage. Some respondents said they
thought they were too old. Others said they thought they had made enough of a
contribution, or that the questionnaire was very long. A few said it was a waste of time as
they thought their views would not change anything.
Those who took part were told they would be given financial incentives which they would
receive after they were reinterviewed. It is likely that considerably fewer would have agreed
to take part if there was no financial incentive. Future work planned will involve giving the
booklet, with little explanation, to British Crime Survey respondents after their interview.
They will be recontacted by ‘phone to find out how many read it without the motivation of a
financial incentive or an awareness that they were to be reinterviewed about it.
32.

These results are not significant for the seminar group – it is likely this is a result of the small sample size for this group.

Evaluation of the information formats

Discussion of the information with others
The participants in the booklet group were asked in the follow-up interview what they had
done with the booklet after reading it. Only three people in the booklet group said they had
thrown it away and a further three had given the booklet to a friend, although it is likely that
respondents were keeping it until after the follow-up interview.
What respondents did with the materials and whether they discussed it with others may give
an indication of how interesting they found it. Discussing the information with others is also
an important method for further dissemination of the information, important for the
effectiveness of any information campaign.
Those aged 16 to 24 and over 65 were more likely to have discussed the information with
others than those in the other age groups. Also, those who had read a broadsheet
newspaper in the last week were more likely to have discussed the information with
someone else.
Of the information groups, the seminar participants discussed the information the most. Only
one person did not discuss the seminar with anyone else, and 84 per cent discussed it with
friends. Fewer people discussed the booklet and video. 61 per cent of the video group did
not discuss it with others, and 43 per cent of the booklet group did not talk to other people
about it (Figure 6.2). However, this leaves a large number of participants who did discuss
the information with other people – a majority for the booklet and seminar groups.

Percentage

Figure 6.2:

Discussion of the information formats with others
90
80
70
60
50
40
30
20
10
0

84
76
61
43

43
26
11

family

16
8

friends
Booklet

work colleagues

other

Seminar

Video

no-one

Improving public attitudes to the Criminal Justice System: The impact of information

Ease of understanding, general impressions, usefulness and enjoyment
Although the information formats were carefully designed to be straightforward and easy to
understand, inevitably some of the information presented was fairly complex. It is therefore
reassuring that the majority of participants in all groups found the information very or quite
easy to understand (Table 6.2). The seminar was thought to be slightly easier to understand
than the booklet or video. This is not particularly surprising as those who attended the
seminars had the opportunity to ask questions and clarify any points that were not clear at
the time.

Table 6.2:

Participant ratings of how easy the information sources were to
understand.
Booklet
%

Very easy
Quite easy
Quite difficult
Very difficult
N

62
35
3
0
109

Seminar
%

Video
%

73
27
0
0
74

57
43
0
0
37

Participants were asked to rate the information sources on a number of dimensions. The
mean rating (from 1 to 5) is shown in Figure 6.3. The higher the mean, the more positive the
rating. All three information sources were rated as very easy to understand and informative.
They were also all judged as reasonably interesting and helpful. The only dimension on
which any of the average ratings were on the ‘negative’ side of the scale was the
exciting/dull dimension – the video group having a mean ranking of 2.54 and the booklet
2.86. This is not really surprising as the information and subject matter is not particularly
exciting – it is factual and for the most part relates to processes. The more positive ratings
for ease of understanding and interesting/boring support this.

Evaluation of the information formats

Figure 6.3:

Mean ratings of the three information formats on various dimensions
0

Mean Rating
2
3

Difficult

5
Easy

Not informative

Informative

Not enjoyable

Enjoyable

Not interesting

Interesting

Not helpful

Helpful

Dull
Like other govt
publications
Old fashioned

Exciting
Not like govt
publications
Modern
booklet

seminar

video

The seminar scored more positively than the other information sources on most dimensions,
in some cases significantly so. For example, the seminar was seen as more enjoyable,
interesting and exciting than the other information types. Two dimensions were only asked
of the booklet and video participants. These were about perceptions of style and tone. Both
the booklet and the video were rated positively – as being modern, and not like other
government publications. The booklet was rated more positively than the video on both
these dimensions.

Interesting topics
All three formats were rated as interesting by a majority of participants. The type of format
had an impact on the extent to which the various topics covered were thought to be
interesting. When asked which parts they found particularly interesting, the differences
between the formats was more noticeable than the difference between subject areas (Figure
6.4). The seminar group rated all topic areas as more interesting than the video and booklet
groups did.

Improving public attitudes to the Criminal Justice System: The impact of information

Figure 6.4:

Parts of the information found most interesting for the different formats
53

types of sentences

61
55

levels of convictions

48
49

types of offences/courts

48
43

amounts and trends in crime

47
42

who is at risk

43
20

reconviction

28
32

community penalties/probation

27
45

prisons

27
28

investigation and prosecution

Percentage
video

seminar

booklet

The topics found most interesting overall were types of offences and court procedures, types
of sentences, levels of convictions, and amounts and trends in crime. Of least interest was
the information given about investigation and prosecution.
The prison part of the booklet was rated as much less interesting than the corresponding
parts of the video or seminar. It is likely this was at least partly due to the video including
footage of a prison and some interviews with prisoners, and the seminar including a
presentation from a prison governor. Those attending the seminars were much more likely to
have found the information about community penalties and probation interesting than those
reading the booklet or watching the video, perhaps due to the presence of a probation
officer at the seminars. The seminar group was also much more likely to have found the
information about rates of reconviction interesting, perhaps because the presence of
probation officers tended to prompt discussion of this topic.
44

Evaluation of the information formats

Accuracy of the information
When rating the information sources on the biased/unbiased scale, the average rating for
the booklet and seminars was about 4 (ratings were between 1 and 5, with 5 being
‘unbiased’). The video was rated as being a little more biased at 3.5. Similarly, for ratings
on whether the formats were thought to be propaganda or factual, all three sources were
rated on average about 4 out of 5, where 5 was ‘factual’ and 1 ‘propaganda’. There was
some mention of bias in response to open questions, particularly from the seminar and video
groups, but not many people felt this.
Around a third of those in the seminar and video groups said there were things they did not
believe, or disagreed with in the information presented to them. This figure was lower for
the booklet at only 15 per cent. The most commonly mentioned things that participants said
they did not believe or disagreed with were firstly that the crime rate is falling, and secondly
the number of 40-year-old men with criminal convictions. All the speakers at the seminars
were thought to be highly accurate and convincing.

Learning and impact
Self-reported learning and information that was surprising
A majority of participants (75%) reported being surprised by what they learnt from the
information sources. People were consistently surprised across most socio-demographic
groups. Fewer of the booklet group reported being surprised by anything they learnt than
the other groups.
The information that participants in the booklet group said they were most surprised by
included the low rates of victimisation of older age groups, the fact crime is falling, the cost
of a prison place, and that a third of men have a conviction by the age of 40. Reactions to
these facts were mixed. Some respondents said they were surprised that some of the
sentences did not seem long enough, and that community sentences were lenient. However,
others were surprised that offenders were supervised so much by the probation service, and
that sentences were longer than they thought. One respondent made the link in their
response between the cost and effectiveness of sentences:
‘…that prison cost more than a community penalty – a lot more, but made no difference in
the re-offending rate.’

Improving public attitudes to the Criminal Justice System: The impact of information

Similar comments were made by the seminar and video groups. The seminar group also
mentioned the effort put into reducing re-offending by the probation service. The comments
from the video group included some on the footage of prisoners. Most of these were either
expressing surprise that the prisoners seemed ordinary and articulate, or that the prisoners
were criticising the prison regime.
All but two of the participants reported learning at least something from the information they
had received and 55 per cent said they learnt a lot. Those from social classes A and B were
more likely to say they had learnt a lot and those who had read a tabloid newspaper in the
last week were less likely to say so. The percentage saying they had learnt a lot was highest
for the seminar (73%) and lowest for the booklet (46%).
The booklet presented the information in a more concise way, and contained less
information overall (concentrating primarily on that required to answer the ‘quiz’ questions),
so it is unsurprising that the participants were not so surprised and felt they learnt less.
There appears to be a relationship between being surprised and the perceived amount
learnt. This is consistent with the theory of Parrott (1995) discussed in chapter 1 who
suggests that a discrepancy between expectations and reality is one way active processing
can be promoted. People who reported being surprised by what they learnt were
significantly more likely to say they learnt a lot than those who said they were not surprised.
Breaking down by information type, this difference is significant for the booklet and video
groups. The seminar group were all likely to say they had learnt a lot (Figure 6.5).

Figure 6.5:

Percentage who were or were not surprised by what they learnt who said
they learnt a lot, by information type.

Percentage learning a lot

75
68

70
60

50
40
30

20
10
0

Booklet

Seminar

Surprised by what learnt

Video
Not surprised

Evaluation of the information formats

This relationship seems to apply to individual types of information. Many of the items of
information which were said to be most surprising were also those where most improvement
in knowledge is seen (chapter 4), particularly crime trends and the cost of prison.

Changing views
44 per cent of respondents said they had changed their views as a result of the information
they had been given, with those with qualifications and those aged under 45 more likely to
say this. Participants who thought they had learnt a lot were also significantly more likely to
report having changed their opinions. Almost three-quarters of the seminar group said they
had changed their views, compared with 41 per cent of the booklet group and 32 per cent
of the video group (Figure 6.6).
On average participants in all groups reported that they were somewhat more confident in
the CJS as a result of the information they had received. This is one of the only questions
where the seminar was not rated the most highly. This is consistent with the findings reported
in chapter 5 for the confidence measures as the seminar had less of an impact on
confidence than the other information types.33

Figure 6.6:

Percentage of participants who said they changed their views as a result of
what they learnt.
80

Percentage

70
60
50
40

41
32

30
20
10
0

33.

Booklet

Seminar

Video

There were no significant changes in confidence on any of the measures for the seminar group.

Improving public attitudes to the Criminal Justice System: The impact of information

How could the information formats be improved?
Although all the information sources were rated highly by participants, there were a number
of areas identified for improvement.
Some participants in the booklet group thought that some of the colours were difficult to
read, and that sometimes the statistics were hard to understand. However, others thought
that it should go into more detail in a few of the topic areas. Those who saw the video
thought that it could be improved by breaking up some of the long segments of film of the
seminar and including more footage of prison and prisoners, as well as interviews with
victims. The main suggestion given for the seminar was that it should be made longer to
allow fuller discussion. Some of those who attended the seminar said they would have liked
a written summary.34

Conclusion
Generally, ratings of all the information sources were positive. All of the information formats
were thought to be informative, interesting, enjoyable and helpful, with the seminar the most
positively rated on nearly all of the evaluation questions. The booklet and video were both
judged to be ‘modern’, and ‘different’ from other government publications. All the formats
were generally thought to be credible and accurate, although this was less so for the video.
Participants expressed surprise at many of the facts presented, and this was associated with
greater self-reported learning. Respondents from all the groups said that they had changed
their views and many thought they were now more confident about the system.

34.

This was not provided at the time as it would have confused the impact of the seminar with that of the written
material. However, participants in both the seminar and video groups were provided with a copy of the
booklet after their follow-up interview.

Discussion and conclusions

Providing simple factual information can improve public knowledge about crime and
sentencing in the short term at least. Participating in the project also had an impact on
attitudes to and confidence in the CJS. After being given information about crime and
criminal justice system procedures and practices, participants were (on average): less
worried about being the victim of crime; less likely to say sentencing is currently too lenient;
and, had more confidence that it is effective in bringing people who commit crimes to
justice, and respects the rights of the accused. Each of the three information formats tested
(the booklet, seminar and video) improved knowledge and had some influence on attitudes.

Using information to improve knowledge
National Statistics, which has a code of practice that many Home Office Research
Development and Statistics Directorate publications are subject to, has a stated aim of
providing a window on the work and performance of government. Most National Statistics
publications are not, however, aimed at the general public. Even the more ‘user friendly’
formats that have been developed, such as the Digest of Information on the criminal justice
system in England and Wales (Barclay & Tavares (eds) 1999) and the Guide to the criminal
justice system in England and Wales still require quite a high level of literacy and/or
numeracy.
Previous research had shown that levels of knowledge about crime and the CJS in the
general population are very poor and that those with the poorest level of knowledge are
also likely to be those with the lowest levels of literacy and understanding.35 For this reason,
the information participants were provided with was deliberately kept to a minimum. The
complexities of the criminal justice and legal system were necessarily simplified to convey
general principles. There is of course no ‘right’ answer as to the level of detail to include.
Finding a reasonable compromise between keeping the message simple and not
misrepresenting the facts is itself a challenge. But, as even at this relatively simple level no
one who took part could correctly answer all the knowledge questions, we were reassured
that there was still further material that could have been absorbed.

35.

See Hough and Roberts (1998) and Mattinson and Mirrlees-Black (2000).

Improving public attitudes to the Criminal Justice System: The impact of information

Despite the simple style of the messages presented, and the range of methods (written, aural
and visual) used to convey the material, it was the better educated who were best at
absorbing and recalling the information in the survey context. However, even those with less
demonstrated experience of learning whose initial level of knowledge was particularly poor
did show a marked improvement. It is important to be aware, therefore, that some people
are more responsive to improving knowledge than others, particularly those with
educational qualifications. Alternative strategies may be necessary to improve the
knowledge of those without these advantages.

Using information to change attitudes
Although there were marked improvements in attitudes to the CJS, there was little evidence
that the shifts in attitudes found here could be attributed solely to improved knowledge about
crime and criminal justice. The sheer act of engaging people in this type of exercise
appears to be sufficient to bring about an improvement in attitudes. Despite our
preconceived belief that those with extreme views were likely to be the most intransigent, for
all three of these attitudes, this was the group who were the most likely to change their
minds. It may be that those with the most extreme views are those in most need of
reassurance that the state is acting in their best interests.
Participants’ beliefs about the aims of sentencing did not change much after receiving
information, but they were already focussed on crime reduction and arguably the way the
material was presented – measuring effectiveness of disposals in terms of reduction in reoffending rates for instance – was unlikely to have influenced this view. Choice of sentences
for offenders also did not change much, but again this is hardly surprising given the initial
preference for non-custodial sentences.

The format of information
All three formats were well received by those taking part in the research. All three produced
similar improvements in knowledge, and although there were different patterns on the
attitudinal measures, all showed some improvements. The key distinction between them,
though, was the extent to which they reached their potential audience.
Participation in the seminar was very low, despite the fact that participants were being paid to
attend. Many of those who initially agreed to attend did not turn up at the venues, and those
50

Discussion and conclusions

that did were not at all representative of the general population. They were more likely to have
a prior interest in the subject matter, to have educational qualifications and to read broadsheet
newspapers. Although the seminar received particularly positive ratings – perhaps reflecting
participants’ greater involvement – it may be that this group would have been equally as
enthusiastic about the other formats. We have to conclude that seminars are unlikely to be an
effective way of conveying this type of information to a wide cross-section of the general public.
The effectiveness of the video in this respect is more difficult to judge, because there was a
particularly poor response rate to the follow-up survey for the video group. We cannot rule
out the possibility that this was because respondents had not got around to watching it by
the time the interviewer called.
The booklet was read by the greatest proportion of those it was offered to. Reflecting this,
those who read it were fairly representative of the general public in composition.
The booklet also scores highly on cost-effectiveness. The seminars were the most expensive
because they incurred venue and travel costs. They also required a significant time input on
behalf of presenters. The booklet and video on the other hand initially incurred design costs,
and then production and distribution costs. But the per participant cost would decrease
significantly over time with the booklet becoming cheaper on a copy by copy basis than the
video format. However, technological advances may make the delivery of audio-visual
material, to personal computers for instance, a viable method in the future.

Next steps
Improving public confidence in the criminal justice system is one of the key aims of the three
government departments involved in criminal justice. Given the improvements in confidence
demonstrated here, providing information to the public might prove an important element of
this work alongside initiatives to improve the system itself. Initiatives have tended to
concentrate on improving satisfaction and confidence in the criminal justice system amongst
victims, witnesses and those in contact with the system. However, clearly there is
widespread dissatisfaction and misunderstanding of the system amongst the general public
that also needs to be addressed.
One of the conclusions reached by the review of the sentencing framework in the Making
Punishments Work report, using evidence from this research, was that there was no need to
change the levels of punishment purely to increase public confidence. Instead it suggested
51

Improving public attitudes to the Criminal Justice System: The impact of information

that better and more accessible information for the public was needed, and it recommended
that the Home Office should consider ways of increasing public knowledge about how
sentencing is intended to work and how it is working in practice.
For those groups in contact with the CJS, dissemination is fairly straightforward. For
instance, material can be left with courts and victim support organisations. The contact itself
is likely to be sufficient to prompt interest in and engagement with the information.
It is more difficult to identify methods for disseminating material to those with no contact, or
indeed, no particular interest in the CJS. To absorb this sort of information, interest in the
subject matter needs to be raised. Law and order issues seem to be a subject where a
relatively high level of interest already exists, so all that might be required is a way of
tapping into this existing interest. Social surveys, news articles, and prize quizzes are the
kinds of methods that are likely to be successful.
An information marketing campaign would therefore have a number of different strands –
targeting groups of witnesses, victims and jurors through existing contacts with the CJS,
alongside a proactive information campaign targeted at the general public.
The booklet is currently the most cost-effective of the formats tested and it also reached the
widest cross-section of people. For this reason the booklet has been updated and
redesigned, taking into account the comments from the participants in the research. If it is
well received it will be revised periodically and will form part of the strategy to improve
confidence in the criminal justice system in England and Wales.

Appendix A

Methodology and statistical significance

Sampling error and statistical significance
In any sample survey or research, the sample may produce results that differ from the figures
that would have been obtained if the whole population had taken part in the research. This is
called sampling error. Sampling error means that changes or differences found may have
occurred by chance. Tests of statistical significance are used to identify which changes are
unlikely to have occurred by chance. Where a test has found a result to be significant at the
5% level (p<0.05) there is only a one in 20 chance that this result is due to chance variation.

Weighted data
In this research, the raw data from the general public survey (GPS) have been adjusted to
correct for imbalances in sampling which affect the representativeness of the sample. The
participant data have not been weighted.

Interviewing method
The interviews were conducted using Computer Assisted Personal Interviewing (CAPI) where
the questionnaire responses are entered directly on to a portable computer by the
interviewer and took an average of 31 minutes in the first stage, and 44 minutes in the
second stage. The questionnaire was piloted before the first stage, with ten respondents in
two towns. This led to some minor revisions. Interviewers were provided with instructions for
the interview and the recruitment to the second stage. The interviews were carried out in the
afternoon and evening, and at weekends to help make sure the sample included people
who worked.

Sample design
The aim in the first stage was to obtain a representative sample of the general public aged
16 or over. The sample was clustered because of the need for seminar participants to be
within a reasonable distance of the venue. Sampling points were selected systematically,
53

Improving public attitudes to the Criminal Justice System: The impact of information

stratified by region and social class. Respondents were selected randomly, using systematic
selection of sampling points stratified by region and social class. Postcode sectors were then
selected within the sampling points. The data were weighted to the population by region for
analysis.
Quotas were used to ensure that the groups receiving the different information sources were
well balanced. The participants received financial incentives to take part. Those in the
seminar groups received £40, plus travelling expenses. Those who read the booklet or
watched the video received £20 each.

Design of the questionnaire
Two versions of the questionnaire were used – one for the before stage and general public
survey, and the second for after participants had received information in one of the three
formats. Much of the content of the two versions was the same and included open-ended as
well as closed questions and covered confidence in the criminal justice system, awareness
of sentences, sentencing scenarios, aims of sentencing, aggravating and mitigating factors
in sentencing, knowledge questions about the CJS and demographic questions. The
questionnaire found at http://www.homeoffice.gov.uk/rds/horspubs1.html combines both versions
to include all the questions asked. Most of the items were closed questions, but there were a
number of open-ended questions for which the interviewers used Pen CAPI to note the
responses.

Appendix B

Booklet

Catching up
with crime and
sentencing

The full version of this booklet can be found at http://www.homeoffice.gov.uk/rds/horspubs1.html

Improving public attitudes to the Criminal Justice System: The impact of information

Video

Seminar

Booklet

Those
Those
Those
Those
Those
Those
Those
Those
Those
Those
Those
Those

asked to take part
who agreed to participate
recontacted
who had read/watched/attended
asked to take part
who agreed to participate
recontacted
who had read/watched/attended
asked to take part
who agreed to participate
recontacted
who had read/watched/attended

Table C4.1: Take up rates for the three information formats

195
141
113
109
319
116
37
37
188
139
77
74

Number

72
100
36
100
74
100
-

58
80
100
12
32
100
41
55
100
-

% agreeing
%
to participate recontacted

56
77
96
100
12
32
100
100
39
53
96
100

% who had
read/
watched/
attended

Appendix C
Additional tables

Improving public attitudes to the Criminal Justice System: The impact of information

Table C5.1: Percentage of respondents reporting reduced worry after receiving
information (not including those originally ‘not at all worried’)
N
Total = 213
Sex
male
female
Age
16-34
35-64
65+
Class
AB
C1
C2
DE
Ethnicity
White
Not white
Qualifications
Some
None
Victim of crime
Has been a victim
Hasn’t been a victim
Original level of fear
Very worried
Fairly worried
Not very worried
Interest in law and order issues
Very interested in law and order
Fairly or less interested in law and order
Original level of knowledge
Less than 6 correct
6 or more correct
Information source
Booklet
Seminar
Video
Change in knowledge
Not increased knowledge
Increased knowledge

% reduced
worry

103
110

28
21

60
125
28

23
24
29

60
57
42
54

22
28
33
17

208
5

25
20

176
37

27
11

141
72

25
24

37
87
89

54
28
9

79
134

33
19

189
24

25
17

108
35
70

19
40
17

60
153

22
26

Sig at p<0.05
level

*
*
*

Appendix C

Table C5.2: Percentage becoming less punitive after receiving information (only those
thinking sentencing a little or much too lenient in first survey)
N
Total = 159
Sex
male
female
Age
16-34
35-64
65+
Class
AB
C1
C2
DE
Ethnicity
White
Not white
Qualifications
Some
None
Victim of crime
Has been a victim
Hasn’t been a victim
Original beliefs about sentencing
Much too lenient
A little too lenient
Interest in law and order issues
Very interested in law and order
Fairly or less interested in law and order
Original level of knowledge
Less than 6 correct
6 or more correct
Information source
Booklet
Seminar
Video
Change in knowledge
Not increased knowledge
Increased knowledge

% more
satisfied

75
80

31
40

35
96
24

37
32
46

40
39
36
40

38
31
36
38

152
3

36
33

124
31

39
23

55
100

36
35

98
57

29
47

60
95

37
35

134
21

36
33

79
24
52

38
38
31

48
107

25
40

Sig at p<0.05
level

Improving public attitudes to the Criminal Justice System: The impact of information

Table C5.3: Percentage becoming more confident that the CJS brings people who
commit crimes to justice (not including those very confident in first survey)
N
Total = 216
Sex
male
female
Age
16-34
35-64
65+
Class
AB
C1
C2
DE
Ethnicity
White
Not white
Qualifications
Some
None
Victim of crime
Has been a victim
Hasn’t been a victim
Original level of confidence
Fairly confident
Not very confident
Not at all confident
Interest in law and order issues
Very interested in law and order
Fairly or less interested in law and order
Original level of knowledge
Less than 6 correct
6 or more correct
Information source
Booklet
Seminar
Video
Change in knowledge
Not increased knowledge
Increased knowledge

% more
confident

105
110

34
36

59
127
29

37
33
38

61
57
41
56

38
32
34
36

210
5

35
40

177
38

33
42

141
74

34
37

80
108
27

6
44
85

78
137

33
36

190
25

37
20

106
37
72

33
30
40

57
158

32
36

Sig at p<0.05
level

*
*

References

Anderson, S., Ingram, D. and Hutton, N. (2002). Public attitudes towards sentencing and
alternatives to imprisonment. SP Paper 537. London: HMSO.
Barclay, G. and Tavares, C. (eds) (1999). Digest 4: Information on the criminal justice system
in England and Wales. London: Home Office.
Doble, J. (1997). Survey shows Alabamians Support Alternatives. In M. Tonry and K.
Hatlestad (eds) Sentencing Reform in Overcrowded Times: A comparative perspective.
Oxford: Oxford University Press.
Doble & Immerwahr (1997). Delawareans Favor Prison Alternatives. In M. Tonry & K.
Hatlestad (eds) Sentencing Reform in Overcrowded Times: A comparative perspective.
Oxford: Oxford University Press.
Fishkin, J.S. (1996). Bringing Deliberation to Democracy. The Public Perspective Vol. 7.
No.1. pp. 1-4.
Furnham, A. Gunter, B. and Green, A. (1990). Remembering Science: the Recall of Factual
Information as a Function of the Presentation Mode. Applied Cognitive Psychology, 4, pp.
203-212.
Halliday, J. (2001). Making Punishments Work: Report of a review of the Sentencing
Framework for England and Wales. London: Home Office.
Home Office (2000). A guide to the criminal justice system in England and Wales. London:
Home Office.
Hough, M. (1996). People talking about punishment. The Howard Journal Vol. 35 No.3. pp.
191-214.
Hough, M. and Moxon, D. (1985). Dealing with offenders: Popular opinion and the views of
victims, findings from the British Crime Survey. The Howard Journal Vol. 24. No.3. pp. 160-175.

Improving public attitudes to the Criminal Justice System: The impact of information

Hough, M. and Roberts, J. (1998). Attitudes to punishment: findings from the British Crime
Survey. Home Office Research Study No.179. London: Home Office.
Hough, M. and Roberts, J. (1999). Sentencing trends in Britain: Public knowledge and public
opinion. Punishment and Society: The International Journal of Penology Vol. 1. No.1. pp.
11-26.
Kay, A. F. (1996). How the questionnaire could have been improved. The Public Perspective
Vol. 7. No.3. pp. 20-22.
Ladd, E.V. (1996). Magic Town: Jimmy Stewart demonstrates the "Hawthorne Effect". The
Public Perspective Vol. 7. No.3. pp. 16-17.
Mattinson, J. and Mirrlees-Black, C. (2000). Attitudes to Crime and Criminal Justice: Findings
from the 1998 British Crime Survey. Home Office Research Study No. 200. London: Home
Office.
Mirrlees-Black, C. (2001). Confidence in the Criminal Justice System: Findings from the
2000 British Crime Survey. Home Office Research Findings No.137. London: Home Office.
O’Keefe, G.J., Rosenbaum, D.P., Lavrakas, P.J., Reid, K. and Botta, R.A. (1996). Taking a bite
out of crime: The impact of the National Citizens’ Crime Prevention Media Campaign.
Thousand Oaks, CA: Sage.
Parrott, R.L. (1995). Motivation to attend to Health Messages. In Maibach, E. and Parrot,
R.L. Designing Health Messages. London: Sage.
Riley, D. and Mayhew, P. (1980). Crime Prevention Publicity: an assessment. Home Office
Research Study No. 63. London: HMSO.
Russell, N and Morgan, R. (2000). Sentencing Domestic Burglary. Sentencing Advisory Panel
Research Report 1.
Tarling, R. and Dowds, L. (1997). ‘Crime and Punishment’. In Jowell et al (eds) British Social
Attitudes, the 14th Report. Aldershot: Ashgate.

Improving public attitudes to the Criminal Justice System: The impact of information

RDS Publications

Requests for Publications
Copies of our publications and a list of those currently available may be obtained from:
Home Office
Research, Development and Statistics Directorate
Communication Development Unit
Room 275, Home Office
50 Queen Anne’s Gate
London SW1H 9AT
Telephone: 020 7273 2084 (answerphone outside of office hours)
Facsimile:
020 7222 0211
E-mail:
publications.rds@homeoffice.gsi.gov.uk

alternatively

why not visit the RDS website at
Internet: http://www.homeoffice.gov.uk/rds/index.html
where many of our publications are available to be read on screen or downloaded for printing.