124-Mayhew Et Al-Crime As Opportunity

HOME OFFICE RESEARCH STUDY NO 34
Crime As Opportunity
By P. Mayhew, R. V. G. Clarke, A. Sturman and
J. M. Hough
A HOME OFFICE
RESEARCH UNIT
REPORT
LONDON: HER MAJESTY'S STATIONERY OFFICE

HOME OFFICE RESEARCH STUDIES
'Home Office Research Studies' comprise reports on research undertaken in the
Home Office to assist in the exerqise of its administrative functions, and for
the information of the judicature, the services for which the Home Secretary
has responsibility (direct or indirect) and the general public.
On the last pages of this report are listed titles already published in this series,
and in the preceding series Studies in the causes of Delinquency and the Treatment
of Offenders.
ISBN 0 11 340674 6
Foreword
This report attempts to show that a closer study of opportunity (in the physical
sense) as one of the factors associated with offending might help to redress an
imbalance in criminology brought about by concentration on the social and
psychological characteristics of known offenders. It might also help to provide
a sounder basis for those crime prevention measures, also in the physical sense,
which seek to reduce opportunities for crime.
Such points are expanded in the first part of the report, which offers a theoretical
perspective for the two empirical studies which follow. Whilst acknowledging
that the concept of opportunity is a complex one, the authors propose a classifi-
cation of opportunities for crime which might guide criminological investigation.
They suggest that the opportunities offered by lack of surveillance or physical
security (as opposed to those presented by the abundance of stealable property or
by people's patterns of activity) are those which criminologists could most use-
fully examine in relation to crime prevention practices. They also recognise,
however, that any examination of the role opportunity plays in crime must deal
with the question of whether measures which physically restrict opportunities for
offending lead to the reduction of crime or simply to a 'displacement' of it
elsewhere.
The empirical studies reported examine two forms of offending in terms, respec-
tively, of security and surveillance. The first looks at the effect on patterns of car
stealing of the fitting of steering column locks to all new cars in this country since
1971. It shows on the one hand, that cars protected by such locks have a reduced
chance of being driven away illegally; but, on the other hand, that their increased
security has in all probability resulted in older vehicles being stolen. The second
study examines how supervision of passengers by the drivers and conductors of
buses affects the extent and location of bus vandalism and confirms that more
damage is committed in those parts of the bus not easily supervised by the crew.
Both these studies have implications for crime prevention which are drawn out in
the discussion and it is hoped that they illustrate the value of the approach
advocated in the report.
I. J. CROFT
Head of the Research Unit
October 1975
Acknowledgements
We are very grateful for the help and advice given us in the course of our work on
steering column locks by the Statistics and the Stolen Motor Vehicles Investiga-
tion Branches of the Metropolitan Police.
In relation to the bus vandalism study, thanks are due to officers of Greater
Manchester Transport for their co-operation and guidance and to P Wiles of
Sheffield University for helpful discussions. The study, which was carried out by
A Sturman, forms part of a broader programme of research into vandalism under
the direction of T F Marshall.
We should also like to thank Home Office colleagues for useful comments on the
draft, and in particular Dr Ian Sinclair, formerly a member of the Home Office
Research Unit, and Professor Peter Garabedian, formerly research consultant to
the Home Office.
R. V. G. CLARKE
J. M. HOUGH
P. MAYHEW
A. STURMAN
Contents
Foreword Page iii
Acknowledgements v
Chapter 1 'Physical' crime prevention: A criminological perspective 1
Chapter 2 Steering column locks and car theft 9
Chapter 3 Damage on buses: The effects of supervision 21
Chapter 4 Concluding remarks 29
References 31
vn
1 'Physical' crime prevention: A
criminological perspective
Criminological interest in crime prevention has been largely confined to 'social'

measures which seek to eliminate or compensate for the deprivations seen as
leading to crime; much less interest has been taken in 'physical' preventive
measures which aim to make crime more difficult to commit.1 This is consistent
with a tradition of 'liberal reformism' within criminology (Platt, 1974) insofar as
'social' measures may bring about improvements in community and welfare
services, while 'physical' measures can all too readily be identified with the
unattractive features of high walls, barbed wire and guard dogs. Moreover, the
view on which 'physical' prevention appears to rest, that crime is a mechanistic
response to objectively powerful situations, seems to diminish the specialist
contributions of many of the disciplines that interest themselves in criminal
motivation.
Despite this, there is now evidence of an emerging interest among criminologists
in physical preventive methods (Jeffery, 1971; Grenough, 1974; Nieburg, 1974).
This, however, has centred more on questions of architectural planning and
environmental design raised by Jacobs (1961) and Newman (1972) than on the
traditional police concern with protective surveillance, adequate securing of
property and the 'hardening' of vulnerable targets through the use of technical
devices. The purpose of this volume is to suggest that criminologists, by under-
taking evaluative studies of crime prevention measures and by developing exist-
ing work on the situational determinants of crime, have a useful contribution to
make to physical crime prevention of both the traditional and newer varieties.
Moreover, increased attention to the relationship between crime and the
immediate environment in which it occurs might be valuable for criminology
itself in countering what may be seen as the undue stress that has usually been
placed on 'internal' predispositions and the interaction of these with social
factors in crime. As Gibbons (1971) has argued, there is ample scope for
criminologists to consider also the extent to which deviance may be a temporal
response to the provocations, attractions, and opportunities of the immediate
situation.
1
Although the terms 'social' and 'physical' prevention have not been generally used, others
have recognised the distinction. Wheeler et al. (1967) distinguish 'technological' prevention
from 'efforts to reduce delinquency by creating broad changes in the structure of commu-
nity life', while Morris and Hawkins (1970) make a distinction between 'tactical' and 'general'
prevention.
1
CRIME AS OPPORTUNITY
SlTUATIONAL INDUCEMENTS TO CRIMINALITY
Although underdeveloped within criminology, a situational view of crime has
respectable antecedents in the work of Hartshorne and May, carried out as early
as 1928. As part of a Character Education Enquiry, some 11,000 children were
given the opportunity to cheat, steal and lie in settings which included their
homes, parties and athletic competitions. The results showed that a supposedly
stable trait such as honesty was not consistent across situations but was influ-
enced by immediate situational factors such as the teacher in charge of the tests
or the amount of supervision given. Those who behaved dishonestly in one situa-
tion did not necessarily do so in situations which were even slightly different, and
only a very small minority of children behaved honestly all the time.
In more recent criminology, situational theories of crime have been stressed by
theorists attempting to explain the widespread nature of youthful deviance and
what Wheeler et al. (1967) have called 'the precipitous decline in delinquency and
crime rates as adolescents move into adulthood'. Thus, Matza (1964) has argued
against deep motivational commitment to deviance by describing 'drifting' into
misconduct and episodic delinquency. Briar and Piliavin (1965), similarly, have
stressed 'mundane' situational inducements and lack of commitment to con-
formity, while Yablonsky (1962) and Short and Strodtbeck (1965) have evidenced
the pressures to deviance conferred by working class gang membership.
The importance of situational factors in the institutional treatment of delin-
quents has been underlined in some recent work undertaken by the Home Office
Research Unit. Sinclair (1971) showed that whether or not boys absconded or
re-offended whilst resident in probation hostels depended more on the attitudes
and training of the warden than on factors in the boys' previous history, while
Clarke and Martin's (1971) results also pointed to the importance of the
immediate environment, rather than the personality or background of boys, in
explaining absconding from approved schools. Following on from such findings,
and leaning heavily on recent psychological theory (cf. Mischel, 1968) which has
stressed the importance of behavioural adaptation to the environment rather than
internal predispositions or personality, an 'environmental/learning' theory has
been used by Cornish and Clarke (1975) to explain the ineffectiveness of resid-
ential treatment. They argue that such improvements in behaviour and attitudes
as might be achieved within institutions are not carried over after release because
important differences between the institutional and post-institutional settings
impede the generalisation of behaviour modified in the former.
The basic emphasis of such an approach is on the stimuli presented by the
situation in which action occurs and on the individual's previous experience of
similar situations. More precisely, stimulus conditions, including opportunities
for action presented by the immediate environment, are seen to provide—in a
variety of ways—the inducements for criminality. These are modified by the
perceived risks involved in committing a criminal act; the anticipated con-
sequences of doing so; and—in a complex, interrelated way—the individual's
' P H Y S I C A L ' CRIME PREVENTION: A CRIMINOLOGICAL PERSPECTIVE
past experience of the stimulus conditions and of the rewards and costs involved.1
Considerable difficulties attach to such an explanation of crime and it is not clear
to what extent it could stand as an alternative to views which have stressed
motivational and social factors in aetiology, or to what extent it could merely
complement them. If the latter, it remains to be seen how much causal weight
attaches to situational variables. For the present, all that is being proposed is
that greater prominence might be given in criminological explanation to how the
inducements of the situation operate in different instances of criminality, and that
physical crime prevention (already premised on a situational approach) might be
placed on a sounder footing through more systematic study of situational
variables—especially those which mediate opportunities for crime.
THE OPPORTUNITY FACTOR

The term opportunity has been familiarised in criminology in relation to the
anomic thesis (Merton, 1957; Cloward and Ohlin, 1961) that the restricted socio-
economic opportunities open to working class youths encourage illegitimate
solutions to the problems of acquiring wealth and status. In contrast, there has
been little discussion about the relationship between crime and opportunity,
either in the sense it has of providing the immediate environmental opportunity
for deviance or the inducement for it. Even those theorists mentioned who have
stressed the importance of situational rather than motivational factors in
explaining deviance, have paid little attention to opportunity specifically as a
situational contingency of crime—though again exceptions are to be found in a
variety of contexts.
In the first place, the relationship between crime and opportunity has been
debated in general terms in relation to the proposition that the social and
economic diversity of modern society has increased its criminogenic potential.
Morris and Hawkins (1970), for instance, have argued that affluence stretches
criminal initiative ('as you expand the bounds of potentiality for economic and
social activity, you equally expand the bounds of potentiality for non-conformity,
delinquency and crime'), while Radzinowicz (1966) has drawn attention to the
way in which with increasing affluence, the greater a\ailability of property will in
itself lead to more opportunities for crime ('the sheer frequency with which
opportunities present themselves will make some both tempting and easy'). The
point has been examined more closely in relation to autocrime, both by Wilkins
(1964) and by Gould and his associates (Gould, 1969; Mansfield et al., 1974).
Wilkins used figures showing the parallel relationship between levels of theft in
connection with vehicles and the number of vehicles registered in this country
between 1938 and 1961 to argue that varying levels of autocrime can be explained
in terms of the opportunities presented by variations in the number of vehicles on
1
Thus, to give what is perhaps a rather oversimplified example, a boy might be prompted to
steal a car if it is 3am on a rainy night, there are no buses running, and there is an unlocked
vehicle nearby; he may be especially likely to do so if he has successfully stolen a car in
similar circumstances before.
the road. Gould hats suggested a more complex situation in which crime rates are
affected not only by the supply of 'stealable' goods, but also by demand for them
and by public attitudes to their protection. In other contexts, de Alarcon (1973)
has suggested that the availability of drugs (as well as the presence of 'socialising'
types of drug users) was a crucial factor in an epidemic of heroin and methedrine
abuse in oae English town, while Smart (1974) has recently looked at a steep
reduction in drunkenness offences in this country between 1914-1918 in terms
of the licencing restrictions of the period and the limited availability of alcohol.
Variations in the opportunities for different kinds of crime have been used as an
explanation for differences in their distribution, at a general level, by Walker
(1965) in discussing urban/rural differences and, in more detail, by Boggs (1965)
and Baldwin and Bottoms (1975) in analysing differences within particular cities.
Opportunity has also been used to explain seasonal variations in crime—by Burt
(1944) who observed that the higher rate of property offending in winter months
is facilitated by the long dark nights, and by Clarke and Martin (1971) who
showed that the opportunities to abscond afforded by such conditions explain the
fact that nearly three times as many boys absconded in November as in June from
the approved school they studied. Finally, in a rather different way again, oppor-
tunities for crime have been related to the absence of supervision and security.
Wade (1967) has argued that abandoned houses and buildings under construc-
tion, for example, provide important opportunities for damage, while Newman
(1972) has shown how crime and vandalism on public housing estates can be
explained by the opportunities for crime presented by isolated areas not under
the eye of tenants and caretakers; in relation to security, Baldwin (1974) has
shown how carelessness on the part of victims can help explain the patterns of
housebreaking and vehicle theft.
From our point of view, however, such applications of the concept of opportunity
in the study of crime have done no more than skim the surface; for the most part
opportunity has been acknowledged in passing rather than taken as the main
object of empirical scrutiny. The potential value of studying opportunity more
directly is illustrated by reference to a study of suicide in Birmingham (Hassell
and Trethowan, 1972). During the period 1963-1969, the suicide rate in Birm-
ingham fell by 45 %, nearly double the national figure, from 122 to 67 per million
population. The decline was almost entirely due to a dramatic fall in coal-gas
deaths from 87 in 1962 to 12 in 1970, following a very substantial reduction in the
toxic content of domestic coal-gas. This finding is especially important to us.
First, it raises the question of whether there are similarly important environ-
mental factors in relation to particular categories of crime, which once identified,
could be successfully manipulated. (The recent memorandum on crime preven-
tion of the Scottish Council on Crime accepts, for instance, that stricter control
over the availability of dangerous weapons in Scotland might significantly reduce
the incidence of violent crime). Second, the research shows that although suicide
is commonly seen as behaviour that is determined by strong internal motivations,
a simple change in the opportunity for killing oneself can have a marked effect on
the suicide rate. Finally, it provides a comment on the extent to which blocking
the opportunities to behave in a particular way is likely either to prevent the
behaviour or change its form. In this case, the fact that violent forms of suicide
did not increase, suggests that most of those who would have chosen gas to kill
themselves were not willing to use instead a more unpleasant alternative or
perhaps one that would more certainly result in death. On the other hand, no
doubt, some potential suicides may have tried instead to kill themselves with
drugs—another relatively passive method—even if such 'displacement' was not
reflected in the suicide rate. This could have been, as Hassell and Trethowan
suggest, because of the increasing efficiency of hospital resuscitation services.
DISPLACEMENT AND CRIME PREVENTION

Insofar as human nature is inventive and the opportunities for crime unlimited,
the possibility that blocking the opportunities for action merely displaces it
elsewhere, or changes its form, can be seen as constituting a serious challenge to
physical crime prevention. Nevertheless, such evidence on displacement as
exists—and it has rarely been subject to empirical test—suggests that impeding
action by physical means is likely to discourage at least some less determined
individuals.1 An isolated study by Press (1971), on the effects of increasing police
activity in one precinct of New York, showed that intensified enforcement
shifted less crime to adjacent areas than it prevented locally. In 'Defensible
Space', Newman (1972) suggests that crime prevented on local housing estates by
maximising surveillance of the estates' public areas, was not totally displaced to
nearby districts. In relation to the study of suicide referred to above, the statistics
since 1963 suggest that many of those whose death was prevented by the reduc-
tion in gas toxicity did not choose to kill themselves in other ways.
In order to be able to comment further on the extent to which displacement may
occur, the argument for it needs to be more precisely formulated and different
forms of crime separately examined in relation to displacement. In particular,
it seems worth considering whether displacement occurs more often, or perhaps
only, within particular categories of crime (what might be called 'specific' dis-
placement), or whether it also operates across them ('general' displacement).
This distinction can be clarified by considering the example of an increase in
burglar alarms in an affluent suburb following extensive publicity given to house-
breakings in the area. As a result, some 'specific' displacement of the burglars'
activities may occur in that (i) they may change their modus operandi to encom-
pass techniques of neutralising electronic alarms, or (ii) they may avoid secured
houses and be more likely instead to break into the unprotected ones if these
were sufficiently numerous in the locality. Their criminal behaviour might be
much less susceptible to 'general' displacement to markedly different types of
1
Wheeler et al. (1967) cogently express the reason for this: 'Much of the rest of our lives is
governed by a kind of economy of effort, whereby desired activities can become so difficult to
complete that the effort is no longer made, and there seems no clear reason that criminal
activities should not be governed by analogous principles'.
offending. They would be unlikely, for example, to start raiding banks (which
requires team-work and has much greater risks^-and rewards—attached to it), or
to start robbing people in the street (which involves personal confrontation with
victims—something housebreakers are traditionally thought to avoid). As should
be clear, we would question the idea that there is a floating body of people with
anti-social tendencies which must be expressed in crime of whatever kind. On the
contrary, we believe that criminal behaviour consists of a number of discrete
activities which are heavily influenced by particular situational inducements and
by the balance of risks and rewards involved. Upsetting this balance through
measures which make it more difficult to act is unlikely to displace action to
crime which serves different ends and for which different internal and external
sanctions might apply. The fact, on the other hand, that 'specific' displacement
of behaviour is likely to occur suggests that such an effect should be anticipated
in crime prevention practice and the possibility minimised of behaviour being
displaced to forms of action which are functionally equivalent and which have
similar social and personal significance.
CLASSIFYING AND MEASURING OPPORTUNITIES FOR CRIME

While this discussion has underlined the power of opportunity in determining
behaviour, it has also begun to show the great variety of ways in which it can be
conceptualised. First of all, opportunities for crime can be divided into those
attaching to people and those relating to the objects involved in crime, and each
of these categories, for convenience, can be subdivided further. Thus, opportun-
ities for crime attach to people, firstly, in the sense (as much criminological
theory would recognise) that an individual's personal opportunities to commit
crime vary according at least to age, sex and general life style (eg housewives will
have numerous opportunities to steal from supermarkets), and, secondly, in the
sense that people themselves, as victims of crime, will differentially generate
opportunities for it.1 Thirdly, people's opportunities for crime can be affected by
the patterns of daily activity that follow from particular forms of social organisa-
tion. (Clifford (1974) has argued, for instance, with reference to recent experience
in the Phillipines, that conditions of curfew can dramatically cut crime).
Opportunities that attach to the properties of objects involved in crime can be
seen in the first place to be related to the abundance of'goods in circulation or to
the supply of objects involved in particular criminal offences. (As more cars come
on the road, so opportunities for stealing them will increase; in a society where
every household has a gun, more neighbours and spouses will be shot during
disputes than in a society where guns are tightly controlled.) Secondly, there are
the kinds of environmental opportunities for crime which are~closely linked to
the physical security of the objectsinvolved in a class of offence. (Cars which are
lockedr^areTTnoTeilifiicult to steal, and goods in self-service shops are more
1
Recent work by Sparks et al. (in press) suggests t h a t ' . . . the age-specific risk of becoming a
victim one or more times a year roughly parallels that of being convicted of a crime: that is,
it is highest for adolescents and young adults, declining in the middle and later years of life'.
vulnerable to shoplifting than those in the traditional corner-shop.) Thirdly,

there are opportunities for crime which are mediated by levels of surveillance and
supervision. (Where pram sheds and garages are not easily visible from the
dwellings on a housing estate, they are more likely to be vandalised; the less shop
assistants attend to customers, the more goods will be stolen.)
Apart from the variety of categories under which opportunities for crime can be
classified, an additional conceptual problem is that opportunity, can be seen as
having two components, doubly providing the occasion for action (the objective,
material conditions necessary for an act to be committed), and the temptation for
it (the conditions subjectively perceived as favourable to action). And these are
not discrete since an unlocked car provides both the condition and the induce-
ment for theft. Reconciling the objectively important component of opportun-
istic situations with the subjectivist claim that, in the last resort, opportunities
are only perceived opportunities, is a problem that remains to be tackled (cf.
Baldwin and Bottom, 1975). Clearly, there are countless occasions when the
very considerable opportunities of the situation are not sufficient to provoke a
criminal response and any examination of the concept of opportunity must fully
recognise the fact.
In testing the role of opportunity further, most, if not all types of offence could
be examined in terms of each of the categories of opportunity defined earlier.
For instance, burglary could be looked at in terms of the greater freedom of men
to be away from their homes at night, the density of residential properties in
rural as against urban areas, the security of these properties, or the extent to
which routes of access to them are visible to neighbours and passers-by. The
demands of general criminological explanation, however, will differ from those
of an approach concentrating specifically on crime prevention. While it may be
valuable for criminologists concerned with explaining differences in rates or
changes in the levels of offending, to study how opportunities for crime are
related to individuals' personal opportunities for crime or the abundance of goods
in circulation, the pay-off for crime prevention may be greatest, not from studying
such largely intractable phenomena, but from studying the physical security and
surveillance aspects of opportunity. Looking at car theft, for instance, in terms of
the numbers of cars on the road, or the numbers of persons holding driving
licences, may help in discussing temporal trends in autocrime, but will not lead
to feasible policies for reducing it. If one looks instead at how the securing of cars
affects illegal users (as was done for the study reported in Chapter 2), the results
may be of considerably more practical use. To illustrate the point further: in the
study of bus vandalism reported in Chapter 3, the opportunity for damage could
have been measured by the number of buses on the road, or the types of passenger
carried—in addition to, as was the case, the degree to which passengers are super-
vised by the conductor and driver; choosing to relate vandalism to levels of
supervision by the bus crew meant that we were in a better position to make
practical suggestions for its reduction.
SUMMARY
This introductory discussion provides a framework for the two empirical studies
reported next. It has argued that criminologists should supplement their interest
in 'social' crime prevention by offering more support to 'physical' prevention
through a theoretical development of the subject and the undertaking of more
empirical research. It was suggested that criminological explanation should take
greater account of how, and under what circumstances, situational variables
combine With other relevant factors to produce crime and determine its pattern.
From the point of view of physical crime prevention, it was stressed that the
availability of opportunities for crime is a situational variable particularly well
worth examining, though the difficulties of defining and operationalising oppor-
tunity were acknowledged, as well as the need to study whether reducing
opportunities prevents or merely displaces crime. In classifying opportunities for
crime, it was suggested that those related to lack of surveillance and physical
security were most worth analysing in relation to crime prevention; analysis of
the opportunities afforded by the abundance of property or by people's patterns
of activity might be of greater value in explaining differences or changes in
crimerates.
The two studies reported in the following chapters—one concerned with the
effectiveness of steering column locks on cars, the other relating vandalism on
buses to levels of supervision of passengers by the crew—are intended to illus-
trate the feasibility of analysing crime in terms of the opportunities for it
presented by lack of security and surveillance, the explanatory potential of doing
so, and the lessons for crime prevention that can result.
2 Steering column locks and
car theft
Since January 1971, all new cars imported to and manufactured in this country
have been fitted with a steering column lock as standard equipment. These locks,
whioh are automatically brought into operation when the ignition key is removed,
were introduced in the face of increasing autocrime in preceding years to make it
more difficult for vehicles to be illegally driven away.1 The potential savings to be
offset against the cost of fitting new vehicles with anti-theft equipment (approxi-
mately £10 a vehicle at 1971 prices) were considerable. Autocrime involves a
great deal of police time and effort (it accounts for no less than 24 % of recorded
known indictable crime2), and there are losses to insurance companies which are
passed on to car owners through the premiums they are required to pay. It also
presents considerable hazards to road safety: according to a recent, unpublished
paper by the Federal Bureau of Investigation, a stolen vehicle is 200 times more
likely to be involved in a car accident than one which is not stolen. This may be
because many of those who take cars are young and inexperienced drivers: 76 %
of those caught for taking cars in England and Wales in 1973 were under the age
of 21, and of these almost half were under the minimum legal driving age of 17.
Although it was hoped that fitting new cars with steering column locks would
lead to a reduction in the overall level of vehicle theft and unauthorised taking,
this has already been confounded by a remarkable increase in these offences since
the beginning of 1971. In the Metropolitan Police District in 1974, for instance,
vehicle theft and unauthorised taking was some 80 % higher than in 1970. (Other
indictable crime rose by 22 % over the same period). This increase, however, does
not necessarily mean that steering column locks are ineffective, since published
statistics make no distinction between cars protected by locks and those not ;3 the
1
The agreement to fit steering column locks to cars (and vans derived from cars) was a voluntary
one negotiated by the Home Office with the Society of Motor Manufacturers and Traders.
Though the agreement allowed for alternative anti-theft devices, in practice most cars have
been fitted with steering column locks. Thus, for convenience, all devices covered by the 1971
agreement are referred to in this report as steering column locks.
2
Criminal Statistics: England and Wales, 1973. Autocrime here is unauthorised taking of
vehicles, vehicle theft and the theft of property from vehicles. Vehicles in this case comprise
commercial vehicles and two-wheelers as well as private cars and vans.
3
At the same time, given increasing scepticism about the validity and reliability of official
criminal statistics, it is worth making the point that statistics relating to cars which are
illegally driven away are unusually accurate, at least as far as the reporting of offences is
concerned. Because of insurers' requirements that the police be notified when a claim for theft
is made, and because of the owner's dependence on the police to help retrieve cars, failure
to report missing vehicles is rare (cf. Mansfield et al., 1974).
overall increase in theft and unauthorised taking may well have been accounted
for by offences involving unprotected vehicles.
The study reported below aimed to assess the extent to which the increased
security of vehicles manufactured since 1971 has prevented their unauthorised
use, and given the continuing increase in theft and unauthorised taking, to
examine the question of whether higher levels of protected cars will eventually
reduce the overall incidence of these offences. Thus, the study analyses car theft in
terms of one of the sources of opportunities for crime identified in Chapter 1 (see
pages 6-7), namely lack of physical security. At the same time, it also provided a
chance to study another aspect of opportunity through an examination of
Wilkins's (1964) hypothesis of a direct relationship between the abundance of
vehicles on the road and the frequency of their unauthorised use and, more
generally, to comment on the long-standing question of whether preventive
measures actually reduce crime or simply displace its pattern.
THE EFFECTIVENESS OF STEERING COLUMN LOCKS

The method employed in evaluating the effectiveness of steering column locks
was to see whether a smaller proportion of'new' cars were stolen or driven away
in 1973 (ie since the introduction of the locks) than in 1969 (before their intro-
duction). In both years 'new' cars were defined as those which, according to their
licence numbers were three years old or less1. In 1973 all 'new' cars would have
steering column locks, whereas in 1969 the great majority would not.
1969 rather than 1970 was taken to represent the 'before' situation since a small
number of new cars introduced in 1970 were fitted with anti-theft devices in
anticipation of the 1971 measure. Cars on the road in 1969 would have included a
number of foreign models some of which had anti-theft devices and, although for
strict accuracy some account should have been taken of these, the difficulties of
doing so were incommensurate. In any case, the numbers involved would have
been small; from information given to us by the Society of Motor Manufacturers
relating to new foreign cars registered in this country it can be estimated that in
1969, in the country as a whole, foreign cars accounted for about 5 % of the total
cars on the road. And not all of them were fitted with anti-theft devices.
Sample
The sample was drawn from the Metropolitan Police District's statistical
records. These maintain a distinction between theft of vehicles and unauthorised
1
The age of a vehicle cannot always be determined from its registration number. For example,
second-hand cars imported from abroad are registered by year of import rather than manu-
facture, and some owners of new cars, especially expensive ones, obtain personalised number
plates. These and other exceptions were rare enough to be discounted in the analysis.
Because of the practice since 1967 of changing the suffix to licence numbers on August 1,
J registration cars were subdivided into those registered before February 1,1971 (assumed to
be without security protection) and those registered after that date (assumed to be fitted with
anti-theft devices"*.
10
STEERING COLUMN LOCKS AND CAR THEFT
taking: a vehicle is considered stolen if it is not recovered within 30 days, other-

wise it is recorded as having been taken without authority, The sample comprised,
in 1969 and 1973, the last 20 cars recorded as taken without authority and the
last 20 (or as many as were available) recorded as stolen in each of the 23 main
divisional stations of the MPD. Most of the cars in the sample were taken
between August and December; although in 1969 and 1973 the number of cars
taken at the very end of the year was rather high, there is no reason to think that
for our purposes bias has been introduced by sampling car theft and unauthorised
taking mainly from the second half of the year.
Results
Table 2:1 shows that in 1969 'new' cars represented 20.9% of all cars illegally
taken, whereas in 1973 the figure had dropped to 5.1 %, a difference we would
attribute to the protection afforded by anti-theft devices.1 Moreover, since in
1973 'new' cars represented a greater proportion of the total number of cars on
the road (an estimated 37%) than in 1969 (34%), the difference is a little more
accentuated than it appears.
The 47 'new' cars sampled which were stolen or taken without authority in 1973
despite being protected by steering column locks were not all, of course, neces-
sarily moved by tampering with the locks themselves. Although from the data
examined it was not possible to tell how many of the 'new' cars were left insecure,
it is quite likely that some of the cars would have been left with the keys either in
them or readily available. A United States President's Commission report (1967)
on crime suggested that 42% of cars stolen had unlocked ignitions, while in this
country Baldwin (1974) has shown, similarly, that a disproportionate number of
cars left insecure are taken and driven away, or have property stolen from them.
1
Two other categories of autocrime maintained in MPD statistics are theft of property from a
vehicle which is not moved and theft from a vehicle which is moved. Since 1969, theft from
vehicles which have been moved has continued to rise along with unauthorised taking and
theft, confirming that steering column locks have not improved the overall picture of car
theft. At the same time, it appears from some rather limited data we collected that moving a
car and taking property from it occurred very infrequently among vehicles fitted with steering
column locks, again suggesting that such locks are effective anti-theft devices.
11
Thus, if steering column locks are less effective than might be expected from their
technical design, some degree of carelessness on the part of car owners is likely to
provide part of the explanation.
Although Table %: 1 shows a greater decline in theft than in unauthorised taking,
this was not statistically significant (x2 = 3.6; 1 df). Steering column locks might
have been expected, in fact, to have had a greater impact on unauthorised taking
than on theft, given that 'stolen' cars are often regarded as being taken by
determined thieves interested in resale, but cars recovered within 30 days as being
taken by more opportunistic 'joy-riders', or those in need of transport. It is
becoming increasingly difficult, however, to maintain the distinction between
theft and unauthorised taking on the basis of a 30-day retrieval period, since more
than half the cars recorded as stolen in the MPD are eventually recovered. In
easier circumstances, they might have been recovered sooner and thereby have
been classified as taken without authority. Moreover, in the present sample, there
was no evidence, as one might have expected, that stolen cars were newer than
those taken for more casual purposes, since the distribution of cars of various
ages between the two categories of theft and unauthorised taking was statis-
tically indistinguishable. This again suggests that the present distinction between
the two categories is insufficiently sound to test whether steering column locks
have less of a deterrent effect on professional thieves than on more casual car-
takers.
The possibility that some factor other than increased vehicle security had inter-
vened since 1969 to reduce the vulnerability of 'new' cars in 1973 was dismissed
as remote given that the theft of'new' commercial vehicles and 'new' motorcycles
(neither of which had been covered by any comparable requirement for additional
security) had not dropped since 1969. On the contrary, a limited sample of these
vehicles was examined, and the proportion of 'new' models stolen was found to
have increased from 19 % in 1969 to 22.5 % in 1973.
It seems, then, that steering column locks are efficient in reducing the risk of cars
fitted with them being illegally driven away. In fact, extrapolating from our
sample, the risk of a 'new' car being stolen or taken without authority in the
MPD was about three times less in 1973 as in 1969. The risk to 'old' cars, on the
other hand, nearly doubled over the same period and it seems most likely that
part of this increase in risk reflects the greater security of'new' cars—protecting
these may well have re-directed some thieves to easier targets.
FUTURE LEVELS OF THEFT AND UNAUTHORISED TAKING

While, at present, steering column locks are not providing the police with any
overall benefit, their effectiveness in preventing the theft and unauthorised taking
of cars to which they are fitted suggests that as the proportion of protected cars
increases, the numbers of these offences might fall. For various reasons, however,
it is difficult to make any precise estimate of when steering column locks might
begin to have such an effect. In the first place, like other locks, steering column
12
locks become easier to 'break' as they become worn, so that the protection they
give may diminish with age. Again, as more cars have locks, it is arguable that
the need for the unauthorised user to 'break' them will increase and he may
become increasingly ingenious in his attempts. This might be especially applicable
to those who make a living from stealing cars (by resale of the vehicle, or its
parts) since the value of old, unprotected cars will diminish as the proportion of
cars with locks increases. Lastly, (and this is particularly likely if locks maintain
their effectiveness, or are supplemented by more elaborate security devices),
professional thieves may respond by changing their modus operand!. For instance,
they may increasingly acquire cars from locations such as garage forecourts
where the keys are likely to be available, or by fraudulent means from car hire
firms.
The central difficulty in making reliable predictions about future levels of auto-
crime, however, is that our findings show that the number of 'stealable' cars1 on
the road (ie those without steering column locks) does not clearly or directly
influence the level of theft and unauthorised taking. Figure 2:1 shows a progres-
sive increase in theft and unauthorised taking since 1961 (notably sharp in 1974)
which has not been affected by the reduction in the number of 'stealable' vehicles
following the 1971 measures. Although the number of 'stealable' vehicles was
even lower in 1974 than in 1961, the volume of car theft and unauthorised taking
was 160 % higher at the later date than at the earlier one. Theft and unauthorised
taking is apparently not dictated solely by the number of easy opportunities
available, and reducing opportunities by fitting an increasing proportion of cars
with anti-theft devices might not effect the level of theft in any predictable way.
The increase in theft and unauthorised taking shown in Figure 2:1 also calls into
question the hypothesis that autocrime is fairly closely related to the number of
vehicles registered, as Wilkins (1964) has argued was the case in England and
Wales between 1938 and 1961. If the number of cars without steering column
locks is taken as an index of opportunity to steal cars, his hypothesis is, as
indicated above, clearly untenable; if the total number of cars registered is taken
as this index, the data still disproves the hypothesis as it stands: for instance, in
1974 there was a 34 % increase over 1973 in the number of thefts and unauthorised
takings, but a slight decrease in the total number of cars registered.2
A more sophisticated model to explain levels of car theft has been proposed by
Gould and his associates (Gould, 1969; Mansfield et al., 1974) which takes into
1
'Stealable' cars are those assumed to be without steering column locks on the grounds that
they were first registered before 1971. The number of such vehicles on the road in subsequent
years was estimated by subtracting new registrations from the total number registered in each
year, allowing for a small amount of'wastage' of vehicles first registered after January 1971.
2
There are other problems with Wilkins's analysis. In relating the level of vehicle registration
to the volume of theft from motor vehicles, rather than to the theft of or to the unauthorised
taking of these vehicles, Wilkins appears to have chosen the index of autocrime which
supported his proposition best. Moreover, even this rather convenient measure of autocrime
has not related well since 1961 (when his analysis finished) to the increase in vehicle
registration.
13
STEERING COLUMN LOCKS AND GAR THEFT
account the changing *«lationship between the 'supply' of vehicles and the
'demand' for them from various sections of the population. Explaining varying
levels of autocrime at different periods and in different countries, they have
claimed that when vehicles are in short supply they are the preserve of the
professional thief, but when they are abundant they are stolen mainly by
amateurs (for instance by those who wish to keep a vehicle for their own use),
The model, however, does no adequately accommodate the pattern of auto-
crime in this country. For instance, while it is claimed that vehicle thefts peak and
begin to decline when there are about 160-200 cars per thousand of population*
there is no sign that vehicle thefts in this country are beginning to decline even
though vehicle registrations are now well beyond the level specified. Moreover^
the steep increase in theft and unauthorised taking that has occurred in this
country recently cannot be easily explained in terms of supply and demand since
the number of vehicles on the road has not greatly altered.
One shortcoming of the predictive models of both Wilkins and Gould would
appear to be that no weight is given to varying levels of vehicle security. More
precisely, their models may only be tenable when the abundance of similarly
insecure vehicles is the only changing factor over time. They fail to accommodate
situations in which the overall level of vehicle security is raised, as we explain
below has been the case in the German Federal Republic, and those in which a
proportion of the cars on the road are made more secure.
The importance of vehicle security has been confirmed by our findings that cars
protected with locks are much less likely to be taken or stolen than they would
otherwise have been. Other evidence (Bundeskriminalamt, 1973) of its importance
is provided by the pattern of autocrime in the German Federal Republic since
1963 when all cars, both new and old, were required to be fitted with anti-theft
devices (see Figure 2:2). Increasing the security of the total population of cars in
this way produced a very marked decrease (62 %) in car theft during the first
complete year (1963) when all cars were protected over the last complete year
when no cars were protected (1960). In fact, security protection decreased the
risk of a car being stolen or taken without authority by a factor of nearly four,
and this decrease has endured: the risk of a car being stolen was virtually
identical in 1972 as in 1963, taking into account an 86 % increase in registrations.
In other words, the German case indicates that the incidence of car theft is
related not only to the number of cars on the road (as Wilkins suggests), or the
changing demand for them by different types of car thief (as on Gould's argu-
ment), but also to the degree to which they are secured.
There are considerable problems, therefore, in accurately predicting future levels

of car theft and unauthorised taking in this country. The apparent fact, how-
ever, that a substantial proportion of cars taken involve youths who 'joy-ride' or
miss the last bus home perhaps suggests that anti-theft devices will eventually
reduce the overall level of autocrime. According to MPD statistics for 1973, 85 %
of cars stolen or taken without authority were recovered, the great majority of
15
them within 30 days. For the most part, these cars can reasonably be assumed to
have been taken by casual unauthorised users who were probably responding
to the opportunity presented by the large number of relatively insecure vehicles
on the road. One might well expect that the fitting of steering column locks to an
increasing proportion of vehicles will eventually be reflected in a lower incidence
of 'joy-riding' and 'journey-making' on t n e P a r t °f the ^ ess painstaking, more
opportunist thief. For eventually the absolute numbers of unprotected cars on
the road will fall to figures low enough to alter materially the ease of finding a
car for illegitimate use. Thus in the MPD in 1973, 1:32 unprotected cars were
stolen or used without authority at a time when cars with steering column locks
accounted for about 37% of cars on the road. In the MPD in 1977, protected
cars will account for about 68 % of cars, and in 1980 for about 81 %.* It is worth
noting that at these two levels some 1:20 and 1:13 unprotected cars would have
to be stolen or taken without authority if the same number of such vehicles were
to be taken as in 1973. On the face of it, either proportion seems untenably high
given (apart from anything else) that risks of these magnitudes would hardly be
accepted complacently by owners pf old cars—or by their insurers.
DISPLACEMENT
A main finding of the present study is that although steering column locks have
substantially reduced the risk of cars fitted with them being illegally driven
away, they seem also to have had the effect of redirecting thieves to cars without
them. The results are therefore compatible with a 'specific' displacement effect as
discussed in Chapter 1. Or, at least, the findings support 'specific' displacement in
the current situation when, given the 1973 level of cars with steering column locks
(in the MPD about 37%) the absolute number of unprotected cars (some 1.2
million) seemed quite adequate to allow displacement to these: the potential thief
or joy-rider would have little difficulty in finding an unprotected car when he
wanted one. We have already pointed out, however, some of the difficulties of
knowing whether displacement to unprotected targets will as readily occur when
the number of these is heavily outweighed by the number of cars protected by
anti-theft devices.
As it becomes increasingly difficult to find unprotected cars, the 'specific' displace-
ment that will occur may be of the kind whereby car thieves—or, more precisely,
some car thieves—change their methods of operation. It seems likely that those
who at the moment steal cars for re-sale will, as well as developing more sophisti-
cated methods of moving secure cars and devoting more effort to the fraudulent
acquisition of cars, also displace some of their present activities to related
offences such as stealing parts and contents without moving the car and stealing
relatively vulnerable commercial vehicles. In contrast, displacement to other
1
These estimates assume that the yearly level of new registrations will remain constant at its
1973 level, and that the 'fall-off' in registrations of old cars will conform to present patterns
in the GLC (as they appear from GLC car registration data). No attempt has been made to
predict the situation beyond 1980.
17
autocrime is less likely to occur among those who casually and opportunistically
take cars for purposes (for instance, a ride home) which would not be obviously
served in other ways.
While this study provides evidence that curtailing opportunities for autocrime
might lead to a degree of'specific' displacement, it says nothing of course about
the extent to which reduced opportunities for car theft will 'generally' displace
behaviour to other forms of deviance. On the arguments presented in Chapter 1,
we ourselves would be hesitant to suggest that, with higher levels of vehicle
security, there will be any greater incidence of other crimes whose ends are not
congruent with those presently served by autocrime, and for which different
internal and external sanctions might apply. Thus, in relation to those casual car
users whose activities are unlikely to be 'specifically' displayed to other forms of
autocrime, we would argue also that their energies are unlikely to be 'generally'
displaced to mugging passers-by for money to get home, hijacking taxis, or
assaulting bus conductors.
GENERAL IMPLICATIONS FOR CRIME PREVENTION

The implications of this study for crime prevention, then, are of some weight. For
having shown that steering column locks have, for the time being at least,
probably displaced some autocrime to unprotected cars, we have shown how
optimistic was the hope that overall levels of car theft might be reduced through
a securing of a proportion of those vehicles at risk. A clear lesson of this research
is that the police will derive only limited benefit from preventive measures which
protect only a proportion of vulnerable property—as indeed has already been
argued by Riccio (1974) in relation to autocrime in the United States. If within
easy reach there is equally vulnerable and equally attractive property, anti-social
behaviour will probably be displaced to this. To derive real benefit the whole
class of property must simultaneously be secured—a principle, of course, which
has been borne out by the successful German experience of anti-theft devices on
cars.1
1
Another interesting example of a legislative measure which has been applied to a total class
of vulnerable property is the requirement brought into operation on 1 June 1973, that all
riders of motor-cycles (and similar two-wheelers) wear protective headgear. While the
measure was introduced for reasons of road safety, there is evidence that an unintended but
valuable consequence of the regulation has been a reduction in the number of two-wheelers
stolen and taken without authorisation. The number of two-wheelers so removed in the MPD
fell from 5280 in the 12 month period immediately prior to the introduction of the protective
headgear regulations to 3997 in the subsequent twelve months (a decrease of 24%). This was
particularly noticeable in relation to unauthorised taking, and contrasts with a rise of 35 % in
the theft and unauthorised taking of other motor vehicles in the same period. To the extent
that vehicle theft is opportunist, it is reasonable to think that some potential users (aware of
what was a well-published requirement) have been deterred from illegally taking two-wheelers
because of their increased visibility if not wearing a crash helmet. It is not unlikely, of course,
that some small proportion of the rise in 1973 in the theft and unauthorised taking of other
motor vehicles could be accounted for by displaced two-wheeler theft. Indeed, such an effect
would be a good illustration of how 'specific' displacement might operate between two
categories of similar offences involving property which serves generally similar ends.
18
Inevitably, however, a total securing of a class of property will cost more than a
partial securing of it; and it is worthwhile trying to assess whether, in Germany
for instance, the cumulative cost of fitting all cars on the road with anti-theft
devices has been justified in terms of some of the more definable savings made.
Indeed, since locks on cars serve no obvious purpose other than increasing car
security, their cost-effectiveness as a crime prevention measure is particularly
well worth considering.
Between 1961 and 1973, the cost of fitting all cars at risk in the German Federal
Republic with anti-theft devices can be estimated at £177m, on the assumption
that the cost of equipping each existing car was £15 and each newly-produced car
£10. On the further assumption (and it is a very optimistic one) that, of all cars
registered, the proportion stolen or taken without authority in each year since
1960 would have stayed at the 1960 level, the loss of some 2.6m cars has been
prevented over the 12 year period—apparently by the universal fitting of anti-
theft devices. Leaving aside that the protection of many newer vehicles will be of
continuing benefit after 1973, the total cost to car-owners of £177m, when
averaged over these 2.6m cars, gives a figure of about £70 per theft prevented, ie
to prevent the loss of one car, some seven individual car-owners have each had to
bear the relatively small expense of £10 to protect their car with an anti-theft
device. This cost can be offset against the cumulative savings made from the
total number of prevented losses in terms of police time, the costs to insurance
companies, the material costs to owners of stolen vehicles, and the costs asso-
ciated with the road accidents in which stolen vehicles are often involved. While
it is difficult to put a figure on these savings (and we acknowledge that for cars
retrieved quickly and undamaged, the inconvenience costs might be greater than
the material ones) it would seem, on the face of it, that the fitting of steering
column locks in Germany has been cost-effective.
Up to the present time, steering column locks have been cost-effective in this
country only for the owners of cars to which they have been fitted—a small
additional sum on the price of a new car has conferred the benefit on these
owners of a substantially reduced chance of their car being stolen. But the fitting
of steering column locks to new cars has not been of any great collective benefit
since, on our argument, the protection of only a proportion of cars on the road
has in all probability meant that car theft has been displaced to continuingly
vulnerable (though admittedly less valuable) older vehicles.
At first sight, then, the argument for requiring old cars as well as new ones to
be fitted with anti-theft devices might seem a strong one. In fact, even discounting
the difficulties of gaining public agreement, the time that would elapse before
action could be taken (realistically perhaps three years) might render the measure
superfluous. By 1978, an estimated 73 % of cars in the GLC will be protected by
anti-theft devices anyway, and as we have said, the owners and insurers of the
remaining vehicles may not be prepared to run the enhanced risks of these cars
being removed illegally.
19
In the meanwhile, the disproportionate increase in theft and unauthorised taking
over the past few years requires some explanation, and this will be helped by
more accurate information about the purposes for which cars are taken, the
immediate inducements which operate, and the ways in which different types of
illegal users acquire cars. Nevertheless, it would seem—given the still large
proportion of cars recovered intact—that casual offenders are heavily implicated
in the recent increase in autocrime. In direct practical terms, then, though we have
argued that steering column locks will ultimately prevent much casual taking of
vehicles, perhaps the benefits of such locks should be maximised by making it
more difficult for drivers to leave keys in the car (through the use of spring-
ejection locks and key warning systems), and more difficult for keys to be
acquired fraudulently. To the extent, however, that some of the increase in car-
taking is in the furtherance of theft of contents, it would be worth trying to
improve door locks. In any event, a technical approach to the problem of vehicle
security is likely to prove more acceptable than at least the alternative of intro-
ducing legal sanctions against drivers who leave their cars insecure.
20
3 Damage on buses: The effects
of supervision
On the argument (cf. Clinard and Wade, 1957; Wade, 1967) that vandalism tends
to be committed spontaneously rather than after deliberate planning, it would
seem a particularly appropriate offence to include in any examination of the role
of opportunity in crime. Vandalism is particularly likely to be directed at
abandoned houses, buildings under construction and closed school buildings in
secluded areas (Wade, 1967), all of which present opportunities for damage
perhaps largely to the extent that they are left unsupervised. Certainly, experi-
mental research in social psychology has indicated that under conditions of little
supervision, the occurrence of various forms of dishonest or irresponsible
behaviour increases (Hartshorne and May, 1928; Mischel and Gilligan, 1964;
Medinnus, 1966; Aronson and Mettee, 1968) while, in another context, attention
has been drawn to how urban street crime can be reduced through intensified
police activity (eg Press, 1971). With the exception of Newman's (1972) research
on public housing estates in the United States, however, little empirical work has
been done relating vandalism specifically to levels of supervision in the com-
munity.
The small-scale research project reported below provided a chance to study how
opportunities for vandalism on buses might be mediated by the ability of drivers
and conductors to supervise passengers—in this way illustrating one of the
principles proposed in Chapter 1 of this volume (see page 7), that opportunities
for crime are likely to be provided by lack of surveillance. Recognising that super-
vision is affected by design features of the bus—closer supervision of passengers
is possible, for example, on buses with conductors and on the lower deck of a
double deck bus (where the conductor would normally be)—the location and
extent of damage on four different types of double deck buses were related to the
different levels of supervision which the crew were able to provide. It was hoped
that the results, by elucidating factors in vandalism over which bus operators
might have some control, could provide a sound empirical base for attempts to
minimise damage.
THE STUDY
Sample
The sample of 99 buses was chosen from the two garages which service the
Southern Area of the Central Divisions of Greater Manchester Transport. A
25 % random sample was taken, stratifying for the four main types of double deck
bus—one-man operated, dual purpose, front-entrance conventional, and rear-
21
CRIME Al OPPORTUNITY
entrance conventional. The numbers of each type of bus making up the sample
and their principal characteristics are shown below:
Method
Since the bus company kept no individual job records of repair work related to
vandalism, damage was directly recorded by the research worker. Four different
types of damage were distinguished—holes, tears, scratches and writing—which
depending on their size were given a score of between 1 and 3. Damage was
recorded for individual seating units, defined as the seat itself and its immediately
surrounding area. Although these units varied slightly in size, analysis showed
that this was not an important factor in the-results. For the main analysis of the
data, damage was scored for the seating units in four locations of the bus: the
front, middle and rear third (excluding the back seat) and the back seat itself. In
comparing damage between these different locations, mean seating unit damage
scores were calculated to take account of the fact that the number of units in each
location varied between buses.
Damage was recorded over a period of five evenings when the buses were in the
garage. No reliability checks were made of the ratings as this was only intended
to be a small-scale exploratory study. As the damage was recorded by one
observer, however, it was not expected that the assessments would vary greatly
between parts of the bus or from one bus to another. In recording damage, no
allowance was made for the age of the bus or for when it had last been renovated1,
though account was taken of these factors in the analysis. Nor was any attempt
made at the recording stage to distinguish between accidental and deliberate
damage, though writing and many of the larger holes and tears (which would
attract larger damage scores) were quite obviously the result of vandalism.
1
An exception to this was that on the few occasions when parts of seats or complete seats had
been replaced they were separately recorded. As a result of such replacements, not all seats on
a particular bus would have been at risk for the same amount of time. In order to test whether
this affected the comparisons the two types of conventional bus, which had had most of the
seat changes, were compared both overall and by different locations without including any
seat which had been changed: differences in damage scores found between the bus types were,
however, unaffected.
22
DAMAGE ON BUSES: THE EFFECTS OF SUPERVISION
Although four types of damage were distinguished at the recording stage, they
were re-grouped for the purpose of analysis since when considered separately
there were too many zero scores to make detailed comparisons between buses
possible. In order to arrive at a basis for re-grouping, principal component
analyses were performed on the damage scores in the four different locations
and overall. As the results were very much the same whatever location was
studied, only the factor loadings from the overall analysis were used. Holes,
tears, and writing had roughly equal loadings on the first factor while scratches
alone loaded highly on the second (see Table 3:1)—the first two factors being the
only two with an eigenvalue of 1 or more. This suggested that holes, tears, and
writing had much in common, ie they may all have been the result of vandalism,
whereas the scratches were of a different order of damage—perhaps caused
accidentally by large objects such as baskets or umbrellas. It was therefore
decided to sum scores for holes, tears and writing and to omit scratches
altogether.
Table 3:1
Correlation Coefficients and Factor Loadings (all seat positions)
RESULTS
Damage on the lower and upper decks
For all four types of bus the seats on the upper deck suffered much more damage
than those on the lower deck (see Table 3:2). The difference was least pro-
nounced on rear-entrance conventional buses—mean seat damage on the upper
deck being five times as great as on the lower deck—and most pronounced on
one-man operated buses and on dual purpose buses—where damage on the
upper deck was over twenty times greater than on the lower deck.
Table 3:2
Mean seat damage score by deck and type of bus
f
One-man Dual Conventional with Conventional with

operated purpose front entrances rear entrances
(n = 48) (n = 22) (n = 12) in = 17)
Lower deck 0.22 0.12 0.23 0.37

Upper deck 5.12 2.47 2.70 1.97
23
Table 3:2 highlights, secondly, the particularly great amount of damage on the
upper deck of one-man operated buses: they suffered almost twice as much
damage there as any of the other three types of bus (p < .001) where the extent of
upper deck damage did not vary significantly. Another indication of the severity
of damage on one-man operated buses was that some of the seats on twelve of
these buses (a quarter of the total) had been replaced by hard wearing plastic
seats which are fitted only where vandalism is prevalent. None of the other buses
had been fitted with these seats.
The greater vulnerability of seats on the upper deck of all bus types was not
unexpected given that the upper deck would be less supervised than the lower
deck. On conductor-operated buses the conductor would spend most of his time
on the lower deck, leaving the upper deck unattended for long periods. Lack of
supervision, of course, could also explain the particular vulnerability of upper
deck seats on buses operated without a conductor: drivers would observe the
activities of passengers on the upper deck only through the observation mirrors
sited at the front of the bus. The fact that the dual purpose buses are more
similar to conventional buses than to one-man operated buses in the extent of
damage suffered on the upper deck is probably explained by the bus company's
estimate that dual purpose buses operate with conductors on about 80% of their
journeys.
From Table 3:2 it can be seen, thirdly, that on the lower deck, despite relatively
small amounts of damage, there were again differences between bus types.
Analysis of variance showed these to be significant (p < .05). While one-man
operated buses suffered most damage on the upper deck, it was rear-entrance
conventional buses which suffered most from damage committed on the lower
deck. Although conventional buses are slowly being phased out—the older rear-
entrance ones first—the bus company has not adopted a policy of letting the
damage to the oldest buses go unrepaired. Indeed, if there had been such a
policy, one would have expected such buses to have suffered more damage on the
upper deck than was the case.
The probable explanation for this difference in damage on the two decks is that,
in addition to the effect of supervision, the effect of the buses' age has also to be
taken into account. Separate analyses of damage to buses within each of the four
types had not shown any direct relationship between age and extent of damage,
but this was probably because the age range was too narrow for each bus type.
To have treated all buses together, however, would have confounded the analysis
by inclusion of the supervision factor. On the upper deck, the amount of super-
vision was clearly a more important factor than the age of the bus because the
one-man operated buses, despite being at risk for the least time, had suffered
most damage and the oldest buses with a conventional rear-entrance had suffered
the least. On the lower deck, however, the oldest buses, in particular the rear-
entrance conventional ones, had suffered the most damage, some of it possibly
caused not by vandalism but by extensive ordinary wear. Nevertheless, super-
24
vision may still have had a limited effect upon the extent of damage on the lower
deck, as the least supervised one-man operated buses had suffered more damage
than the older dual purpose ones.
The location of damage on the upper deck within different bus types
Analysis of variance of damage scores for the upper deck again revealed that
there were significant differences between the bus types in the location of damage
(p < .001), the greatest differences being between the one-man operated and all
the others, for each of the locations.
It can be seen from Table 3:3 that for all bus types, the back seat had suffered far
more damage than any other location and, with the exception of the rear-
entrance conventional buses, the damage generally got worse towards the back of
the bus. These results were not unexpected given the particular perspectives of
this study: the back seat would be the only seat where the activities of the
occupants would be unobserved by other passengers, and the rear of the bus
would also be less visible to most passengers than the front.
On the rear-entrance conventional buses, although the back seat was still most
severely damaged, there was a greater amount of damage at the front of the bus
than in the centre or rear. It was thought that these findings might partially be
accounted for by a 'displacement' of damage (see Chapter 1, pages 5-6) resulting
from the position of the staircase—rear-entrance conventional buses being the
only buses with their staircases at the back. People sitting near to the staircase
might be more reluctant to commit acts of vandalism when there would be a
danger of being surprised by another passenger or the conductor.1 At the same
time, the relatively small amount of damage on the back seat of rear-entrance
conventional buses and, indeed, the relatively small amount of damage on the
1
In fact, when the two types of conventional bus were separately examined, no significant
difference was recorded in the location of damage, although the tendency for rear-entrance
buses to suffer more damage at the front and less at the back was consistently found for each
of the individual damage types which make up the aggregate score.
25
upper deck in general (see Table 3:2) suggests that the position of the staircase on
these buses, as well as displacing some vandalism, might also have prevented
some.
On the one-man operated and dual purpose buses with, respectively, staircases
situated at the centre and at the front, it is not possible to estimate whether the
location of damage has similarly been affected by staircase position because any
'displacement' would be in the direction of the back of the bus, which was likely
to be most severely damaged wherever the staircase was positioned
DISCUSSION
Two main findings, then, emerge from this study: first damage was greatest
on buses without a conductor, even though these were the newest of the buses
studied;1 second, on all buses, including those with conductors, damage was
greatest in areas of low supervision (especially the upper deck and the back seat).
Neither of these findings may be surprising, but the magnitude of the differences
in the damage recorded should not be overlooked. For example, on one-man
operated and dual purpose buses there was about 20 times as much damage on
the upper as on the lower deck. The finding of a relationship between the super-
vision of passengers and the amounts of damage on buses is broadly in agreement
with Newman's (1972) finding reported in 'Defensible Space' that the least
observed parts of housing estates, such as lifts and staircases, suffered the highest
rates of crime.
Although a relationship was found in this study between lack of supervision and
damage, it is conceivable that the location of damage within buses, but not the
extent of damage between different types, may be affected by where different
people choose to sit. With this in mind a small observational study involving 47
bus journeys was conducted to find out where people of different age and
sex tend to sit. On any one journey only passengers on one deck of the bus were
studied. For practical reasons the study took place during Monday to Friday of
one week, and only on one-man operated buses, although it was accepted that on
other types of bus and at weekends the age and sex of passengers might be
different. The results showed that slightly more people travel on the lower deck,
and that about 60 % of them were women and girls. On the upper deck about
70% of the passengers were male and 21 % were estimated, again without check-
ing reliability, to be under 16; on the lower deck 14% were estimated to be of
that age. A separate analysis of where people choose to sit on the upper deck
showed that children (especially boys) are more likely to be found at the back of
the bus, especially on the back seat (48 % of the back seat passengers were under
16), while the older passengers who travel on the upper deck more often sit at
the front (60% of passengers over 16 sat at the front).
1
Although conductor operated buses tended to serve certain routes, there was no reason to
think that these would attract more or less vandals than the routes served by one-man operated
buses.
26
These results, then, show that those people who are considered least likely to
engage in acts of vandalism (older passengers, women and girls) have a tendency
to sit in the areas least damaged (the lower deck and the front of the upper).
Nevertheless, these differences between groups of people in where they choose to
sit are not nearly as large as the differences in the amount of damage to the
various seat locations. Moreover, it should not be overlooked that very sub-
stantial differences were found between bus types in amount of damage and that
these were most likely to have resulted from differential supervision. It would
seem likely, therefore, that differences in amount of damage between the various
parts of the bus are also largely due to supervision factors, with perhaps a smaller
part of the differences attributable to different kinds of passengers choosing
particular seats. Even more likely is that these two factors are to some extent
interdependent—potential vandals probably choose unsupervised areas even if
they do not intend to cause damage. Vandalism itself might still be a spontaneous
reaction to the opportunities that the situation presents.
Further knowledge about this process could be achieved in several ways. A study
of where people sat on the upper deck of the two types of conventional bus might
show whether those types of people who sat at the back of the front entrance
buses were more likely to be found at the front of the rear entrance buses, given
the effects that the staircase would have on the levels of supervision. Failing this,
interviews with bus passengers might reveal reasons for seat choices.
To find out more about the circumstances of acts of vandalism it would be
necessary to undertake an observational study (with the difficulty that the
observation might inhibit the behaviour being studied). It would be valuable to
know when vandalism is most likely to occur and which sort of bus vandalism is
an individual pursuit and which a group one, since it has been argued that groups
are likely to take greater risks than individuals (Rettig, 1966; Wallach, Kogan
and Bern, 1965). It might be expected, for example, that vandalism occurring in
the more supervised areas was more likely to be committed when there was a
group of two or three children together.
Thus, while the findings of the present study may not greatly enhance our under-
standing of, say, the motivational factors involved in vandalism, they certainly
point to powerful situational determinants of the behaviour and, moreover, they
have important implications for its prevention. The findings suggest that, in
considering the design and manning of buses, bus companies should take into
account the possible effects of their policies on the prevention of vandalism.
Quite clearly, though, bus companies have other factors than vandalism to
consider, in particular staff recruitment and wages. Given the policy of operating
buses with a minimum of staff, it would seem that the only way to reduce
vandalism in the short term is to reduce the opportunity for offences to be com-
mitted in ways other than providing conductor supervision. Greater Manchester
Transport is already trying to do this by colouring flat surfaces so that they do
not show felt-tip writing, by using non-flammable materials wherever possible,
27
and by ensuring that fixtures and fittings cannot easily be removed. The company
is also experimenting with more elaborate devices to minimise damage, such as
closed circuit television on the upper deck and warning devices fitted to seats. All
of this adds to the cost of vandalism (already estimated by GMT to be about
£150,000 a year) and it may be that one-man operated buses will only prove less
expensive to operate if most of the damage is left unrepaired or if more hard
wearing seats are fitted, especially on the upper deck.
SUMMARY
This chapter has been concerned with the effect of supervision by drivers and
conductors on the location and extent of damage on four different types of bus
operating in Manchester. From the particularly large amount of damage to seats
on the upper deck of one-man operated buses, it appeared that, for this deck at
least, the absence of conductor supervision was the most important factor
affecting the amount of damage sustained. At the same time, on all types of bus,
including those with conductor supervision, the degree to which passengers were
supervised by the bus crew seemed to relate clearly to the extent and location of
damage, in that areas of low supervision (such as the upper deck and the back
seats) were damaged most.
There was some evidence that the position of the staircase can lead to a displace-
ment of damage, probably because vandals seated near to the stairs can be
surprised by anyone climbing them. This displacement can be advantageous
overall if it is in a direction away from the back seat where most vandalism
occurs; this was illustrated by the lower rate of damage on the upper deck of
conventional rear-entrance buses compared with the front entrance ones.
Although the least supervised areas of the bus were the most damaged, it was
also found that younger male passengers—perhaps the ones who would more
often commit damage—were somewhat more likely than others to sit in these
places. While this finding suggests that any explanation of the different levels of
damage within different parts of the bus needs to account for the interrelation-
ship between seat choice and supervision factors, it does not, of course, under-
mine the importance of the differences found in damage committed between bus
types. Thus the results indicate that situational factors, such as supervision, can
play an important part in determining the extent of damage on buses, and
indeed, that to reduce bus vandalism, these factors must be taken into account.
28
4 Concluding remarks
This report has suggested that it would be valuable if criminologists were to take
a greater interest in techniques which seek to reduce crime through manipulation
of the physical rather than social environment. Such techniques encompass the
preventive work of the police and some of the architectural and design principles
which in some isolated contexts have recently been discussed in criminology.
They can be seen to rest on the premise that the situational contexts in which
different types of crime are committed need to be separately analysed, and that
prevention should be based on specifically-directed measures which, by making
crime more difficult to commit, discourage the opportunist offender and deter the
professional by increasing his chances of being apprehended. Attention was
drawn in particular to the relevance for practice of considering the role of
opportunity as one of the situational factors in offending.
The studies reported in Chapters 2 and 3 have shown that of the various oppor-
tunities for crime afforded by the environment, at least those presented by the
lack of physical security or surveillance can be profitably examined in relation to
crime prevention practice: Chapter 2 showed that the routine fitting in this
country since 1971 of steering column locks to new cars has been effective-in
substantially reducing the risk to cars so fitted of being stolen or driven away
illegally; Chapter 3 showed that parts of buses not easily supervised by the crew
were very much more likely to be vandalised than areas which were visible to
them.
As well as their more immediate value, the results of both studies also bear on
some general points—in the first place, the question of displacement. Both studies
provided some evidence compatible with the sort of 'specific' displacement of
crime which it was suggested in Chapter 1 was likely to occur. Thus vandalism on
the upper deck of buses was displaced to less visible areas depending on the posi-
tion of the staircase, while the effect of protecting some cars with steering column
locks has been to increase the risk to other cars. Indeed, the relevance for crime
prevention of anticipating 'specific' displacement was underlined by reference to
the position in the German Federal Republic where a dramatic reduction in levels
of theft and unauthorised taking was achieved by requiring all cars to be fitted
with steering column locks at more or less the same time. In other words, the
German experience has underlined the need to consider carefully whether, and in
what circumstances, partial securing of a class of property can be considered
worthwhile.
Secondly, the steering lock study in particular has demonstrated the scope for
29
CONCLUDING REMARKS
evaluating current prevention practices which, despite an impressive growth in
commercial security services and the recent expansion of crime prevention as a
specialised police function,1 have largely escaped examination. As Wheeler et al.
(1967) have argued, one attraction of physical crime prevention is that it may
prove considerably cheaper than attempts to alter the attitudes and abilities of
potential offenders. Physical prevention methods (which are designed only to
reduce crime) should certainly prove easier to analyse in relation to costs than the
more diffuse and multi-purpose practices of social prevention. An admittedly
crude attempt at an analysis of this kind was made in the present report where it
was argued that the use of steering column locks in Germany has been apparently
cost-effective. In the vandalism study, it was suggested that bus companies, in
attempting to reduce vandalism, would have to weigh the costs of reverting to
conductor-operated buses against the costs of using more hard-wearing materials.
Finally, it is hoped that this report has begun to illustrate that physical prevention
is not simply a matter of intensive policing and crude security, but that it can, in
imaginative and unobtrusive ways, utilise technological and architectural exper-
tise to protect vulnerable property from theft and vandalism, curtail the means of
committing crime (for instance, by restricting the availability of dangerous
weapons), and take advantage of the natural supervision of the environment by
ordinary individuals. Hopefully, it has illustrated too that if physical prevention
implies a different form of 'social engineering' from that of social prevention, it
does not necessarily involve a greater degree of behavioural control. These are
small beginnings, however, and for the future there is undoubtedly scope for
undertaking further studies related to the simple classification of opportunities
presented earlier in the report, as well as scope for refining it. There is also a need
to test the notion of 'general' displacement (ie the displacement of one type of
criminal activity to disparate forms of crime), though this may prove to be as
elusive a phenomenon as that of general deterrence. Thereafter perhaps, the most
pressing need will be for research which will allow the importance of opportunity
relative to. other factors in criminal behaviour, to be more precisely determined.
Only then will it be clear whether opportunity merits as central a place in
criminological explanation as it is given in the title of this report.
1
Through the establishment in the 1950s of posts of crime prevention officers and the setting up
in 1963 of a training centre at Stafford. Crime prevention officers number about 500 at present
and are attached to police divisions throughout the country. An important part of their work
is preparing security surveys for firms and retailers, while they are also responsible for giving
general security advice to the public and for bringing attention to local crime risks. Although
security consultancy is their chief manifest role, they are seen as having a latent function in
deterring crime through advertising the presence of the police.
30
References
Aronson, E. & Mettee, O. R. 1968. 'Dishonest behaviour as a function of differ-

ential levels of induced self-esteem'. Journal of Personality and Social Psychology,
9, 121-127.
Baldwin, J. 1974. 'The role of victim in certain property offences'. Criminal Law
Review, June, 353-358.
Baldwin, J. & Bottoms, A. E. 1975. The Urban Criminal. London: Tavistock.
Boggs, S. L. 1965. 'Urban Crime Patterns'. American Sociological Review, 30,
899-908.
Bundeskriminalamt. 1973. Tagungsbericht der Arbeitstagung 'Kraftfahrzeugdieb-
stahV beim Weisbaden.
Briar, S. & Piliavin, I. M. 1965. 'Delinquency, situational inducements and
commitment to conformity', Social Problems, 13, 35-45.
Burt, C. 1944. The Young Delinquents. London: University of London Press.
Clarke, R. V. G. & Martin, D. N. 1971. Absconding from Approved Schools.
London: HMSO.
Clifford, W. 1974. 'Crime as a problem for research in a developmental perspec-
tive'. Justice of the Peace, 138, 703-704.
Clinard, M. B. & Wade, A. L. 1957. 'Towards the delineation of vandalism as a
sub-type in juvenile delinquency'. Journal of Criminal Law and Criminology, 48,
493-499.
Cloward, R. A. & Ohlin, L. E. 1961. Delinquency and Opportunity. London:
Routledge & Kegan Paul.
Cornish, D. B. & Clarke, R. V. G. 1975. Residential Treatment and its Effects
on Delinquency. London: HMSO.
de Alarcon, R. 1973. 'Lessons from the recent British drug outbreak'. Anglo-
American Conference on Drug Dependence, Royal Society of Medicine, April.
Gibbons, D. C. 1971. 'Observations on the study of crime causation'. American
Journal of Sociology, 77, 262-278.
Gould, L. C. 1969. 'The changing structure of property crime in an affluent
society'. Social Forces, 48, 50-59.
31
Grenough, J. L. 1974. 'Crime prevention: a new approach'. Journal of Police
Science and Administration, 2, 339-343.
Hartshorne, H, & May, M. A. 1928. Studies in Deceit. Vol. 1 of Studies in the
Nature of Character. New York: Macmillan.
Hassell, C. & Trethowan, W. H. 1972. 'Suicide in Birmingham'. British Medical
Journal, 1,717-718.
Home Office. 1974. Criminal Statistics: England and Wales, 1973. London:
HMSO, Cmrid. 5677.
Jacob, J, 1961. The Death and Life of Great American Cities. New York: Random
House.
Jeffery, C. R. 1971. Crime Prevention Through Environmental Design. Beverly
Hills: Sage Publications.
Mansfield, R., Gould, L. C. & Namenwirth, J. Z. 1974. 'A socioeconomic model
for the prediction of societal rates of property theft'. Social Forces, 52, 462-472.
Matza, D. 1964. Delinquency and Drift. New York: Wiley.
Medinnus, G. R. 1966. 'Age and sex differences in conscience development'.
Journal of Genetic Psychology, 109, 117-118.
Merton, K. 1957. Social Theory and Social Structure. Glencoe, Illinois: The Free
Press.
Metropolitan Police. 1974. Crime Statistics: Summary for 1973. London: New
Scotland Yard.
Mischel, W. 1968. Personality and Assessment. New York: Wiley.
Mischel, W. & Gilligan, C. 1964. 'Delay of gratification, motivation for the
prohibited gratification, and responses to temptation'. Journal of Abnormal and
Social Psychology, 69, 411-417.
Morris, N. & Hawkins, G. 1970. The Honest Politician's Guide to Crime Control.
Chicago: University of Chicago Press.
Newman, O. 1972. Defensible Space: People and Design in the Violent City.
London: Architectural Press.
Nieburg, H. L. 1974. 'Crime prevention by urban design'. Society, 12, 41-47.
Platt, T. 1974. 'Prospects for a Radical Criminology in the United States'. Crime
and SocialJustice, 1,2-20.
President's Commission on Law Enforcement and Administration of Justice. 1967.
The Challenge of Crime in a Free Society. Washington: U.S. Government Print-
ing Office.
Press, S. J. 1971. Some Effects of an Increase in Police Manpower in the 20th
Precinct of New York City. New York: Rand.
32
Radzinowicz, L. 1966. Ideology and Crime, New York: Columbia University
Press.
Rettig, S. 1966. 'Group discussion and predicted ethical risk-taking'. Journal of
Personality and Social Psychology, 3, 629-633.
Riccio, L. J. 1974. 'Direct deterrence—an analysis of the effectiveness of police
patrol and other crime prevention technologies'. Journal of Criminal Justice, 2,
207-217.
Scottish Council on Crime. 1975. Crime and the Prevention of Crime. Scottish
Home and Health Department, Edinburgh: HMSO.
Short, J. F. (Jr.) & Strodtbeck, F. L. 1965. Group Processes and Gang Delin-
quency. Chicago: University of Chicago Press.
Sinclair, I. A. C. 1971. Hostels for Probationers. London: HMSO.
Smart, R. G. 1974. 'The effect of licencing restrictions during 1914-1918 on
drunkenness and liver cirrhosis deaths in Britain'. British Journal of Addiction, 69,
109-121.
Sparks, R. F., Genn, H. G. & Dodd, D. J. In press. Surveying Victims. London:
John Wiley & Sons.
Wade, A. L. 1967. 'Social processes in the act of vandalism'. In Clinard, M. B. &
Quinney, E. T. (Eds.), Criminal Behaviour Systems: A Typology. New York:
Holt, Rinehart & Winston.
Walker, N. 1965. Crime and Punishment in Britain. Edinburgh: University of
Edinburgh Press.
Wallach, M. A., Kogan, N. & Bern, D. J. 1965. 'Group decision-making under
risk of adversive consequences'. Journal of Personality and Social Psychology, 1,
453-460.
Wheeler, S., Cottrell, S. L. (Jr.) & Ramasco, A. 1967. 'Juvenile delinquency—its
prevention and control'. In Task Force Report: Juvenile Delinquency and Youth
Crime. President's Commission on Law Enforcement and Administration of
Justice. Report on Juvenile Justice and Consultants' Paper. Washington: US
Government Printing Office.
Wilkins, L. T. 1964. Social Deviance. London: Tavistock.
Yablonsky, L. 1962. The Violent Gang. New York: Macmillan.
33
Publications
Titles already published for the Home Office
Postage extra
Studies in the causes of Delinquency gnd the Treatment of Offenders
1 Prediction Methods in Relation to Borstal Training
by Dr Herman Mannheim and Leslie T. Wilkins. £1.15
2 Time Spent Awaiting Trial
by Evelyn Gibson. 27p
3 Delinquent Generations
by Leslie T. Wilkins. 16p
4 Murder
by Evelyn Gibson and S. Klein. 30p
5 Persistent Criminals
by W. H. Hammond and Edna Chayen. £1.25
6 Some Statistical and other Numerical Techniques for Classifying Individuals
by P. McNaughton-Smith. 17£p
7 Probation Research. A Preliminary Report
by S. Folkard, K. Lyon, M. M. Carver, and E. O'Leary. 42p
8 Trends and Regional Comparisons in Probation (England and Wales)
by Hugh Barr and E. O'Leary. 25p
9 Probation Research. A Survey of Group Work in the Probation Service
by H. Barr. 40p
10 Types of Delinquency and Home Background. A Validation Study of Hewitt and Jenkins'
Hypothesis
by Elizabeth Field. 14p
11 Studies of Female Offenders
by Nancy Goodman and Jean Price 30p
12 The Use of the Jesness Inventory on a Sample of British Probationers
by MartinDavies. lip
13 The Jesness Inventory: Application to Approved School Boys
byJoyMott. 17Jp
Home Office Research Studies

1 Workloads in Children's Departments
by Eleanor Grey. 37£p
2 Probationers in their Social Environment
by Martin Davies. 87|p
3 Murder 1957 to 1968
by Evelyn Gibson and S. Klein. 60p
4 Firearms in Crime
by A. D. Weatherhead and B. M- Robinson. 30p
34
5 Financial Penalties and Probation
by Martin Davies. 30p
6 Hostels for Probationers
by Ian Sinclair. £1.15
7 Prediction Methods in Criminology
by Francis H. Simon. £1.25
8 Study of the Juvenile Liaison Scheme in West Ham 1961 to 1965
by Marilyn Taylor. 35p
9 Explorations in After-Care
by Ian Sinclair, Martin Silberman, Brenda Chapman and Aryeh Leissner. 85p
10 A Survey of Adoption in Great Britain
by Eleanor Grey. 95p
11 Thirteen-year-old Approved School Boys in 1962
by Elizabeth Field, W. H. Hammond and J. Tizard. 35p
12 Absconding from Approved Schools
by R. V. G. Clarke and D. N. Martin. 85p
13 An Experiment in Personality Assessment of Young Men Remanded in Custody
by H. Sylvia Anthony. 52£p
14 Girl Offenders Aged 17 to 20 years
by Jean Davies and Nancy Goodman. 52|p
15 The Controlled Trial in Institutional Research
by R. V. G. Clarke and D. B. Cornish. 29p
16 A Survey of Fine Enforcement
by P. Softley. 47p
17 An Index of Social Environment
by Martin Davies. 47p
18 Social Enquiry Reports and the Probation Service
by Martin Davies and Andrea Knopf. 36£p
19 Depression, Psychopathic Personality and Attempted Suicide in a Borstal Sample
by H. Sylvia Anthony. 36|p
20 The Use of Bail and Custody by London Magistrates' Courts Before and After the
Criminal Justice Act 1967
by Frances Simon and Mollie Weatheritt. 57p
21 Social Work in the Environment
by Martin Davies. £1.10
22 Social Work in Prison
by Margaret Shaw. £1.45
23 Delinquency amongst Opiate Users
by Joy Mott and Marilyn Taylor. 41 p
24 IMPACT. Intensive matched Probation and After-Care Treatment: Vol. 1. The design'of
the probation experiment and an interim evaluation
by M. S. Folkard, A. J. Fowles, B. C. Me Williams, W. McWilliams, D. E. Smith and
G.R. Walmsley. 59p
25 The Approved School Experience
by Anne B. Dunlop. £1.22
26 Absconding from Open Prisons
by Charlotte Banks, Patricia Mayhew and R. J. Sapsford. 95p
27 Driving while Disqualified
by Sue Kriefman. £1.22
28 Some Male Offenders' Problems
Part I. Homeless offenders in Liverpool
by W. McWilliams
Part II. Casework with short-term prisoners
by Julie Holborn. £1.80
35
29 Community Service Orders
by K. Pease, P. Durkin, I. Earnshaw, D. Payne, J. Thorpe. 75p
30 Field Wing Bail Hostel: the first nine months
by Frances Simon and Sheena Wilson. 85p
31 Homicide in England and Wales 1967-71
by Evelyn Gibson. 90p
32 Residential Treatment and its Effects on Delinquency
by D. B. Cornish and R. V. G. Clarke. £1.00
33 Further Studies of Female Offenders
Part A. Borstal girls eight years after release
by Nancy Goodman, Elizabeth Maloney and Jean Davies
Part B. The sentencing of women at the London higher courts
by Nancy Goodman, Paul Durkin and Janet Halton
Part C. Girls appearing before a juvenile court
by Jean Davies. £1.55
HMSO
Government publications can be purchased from the Government Bookshops at the addresses
listed on cover page iv (post orders to PO Box 569, London, SE1 9NH), or through booksellers.
Printed in England for Her Majesty's Stationery Office

by Knight & Forster Ltd. Leeds LSI 1 9UB
Dd 290018 K20 1/76
36

124-Mayhew Et Al-Crime As Opportunity

Uploaded by

Copyright:

Available Formats

124-Mayhew Et Al-Crime As Opportunity

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

124-Mayhew Et Al-Crime As Opportunity

Uploaded by

Copyright:

Available Formats

HOME OFFICE RESEARCH STUDY NO 34

LONDON: HER MAJESTY'S STATIONERY OFFICE

Foreword Page iii

Chapter 1 'Physical' crime prevention: A criminological perspective 1

Chapter 2 Steering column locks and car theft 9

Chapter 3 Damage on buses: The effects of supervision 21

Chapter 4 Concluding remarks 29

Criminological interest in crime prevention has been largely confined to 'social'

THE OPPORTUNITY FACTOR

DISPLACEMENT AND CRIME PREVENTION

CLASSIFYING AND MEASURING OPPORTUNITIES FOR CRIME

vulnerable to shoplifting than those in the traditional corner-shop.) Thirdly,

THE EFFECTIVENESS OF STEERING COLUMN LOCKS

taking: a vehicle is considered stolen if it is not recovered within 30 days, other-

FUTURE LEVELS OF THEFT AND UNAUTHORISED TAKING

There are considerable problems, therefore, in accurately predicting future levels

GENERAL IMPLICATIONS FOR CRIME PREVENTION

One-man Dual Conventional with Conventional with

Lower deck 0.22 0.12 0.23 0.37

Aronson, E. & Mettee, O. R. 1968. 'Dishonest behaviour as a function of differ-

Titles already published for the Home Office

Home Office Research Studies

Printed in England for Her Majesty's Stationery Office

You might also like