The Structural Determinants of Fear of Crime: An Analysis Using Census and Crime Survey Data from England and Wales*
| Author | Pat Pack,John Salked,Chris Hale |
| Published date | 01 September 1994 |
| Date | 01 September 1994 |
| DOI | http://doi.org/10.1177/026975809400300303 |
International Review ofVictimology, 1994,
Vol.
3, pp. 211-233
0269-7580/94$10
1994 A B Academic
Publishers-Printed
in
Great Britain
THE
STRUCTURAL
DETERMINANTS
OF
FEAR
OF
CRIME:
AN
ANALYSIS
USING
CENSUS
AND
CRIME
SURVEY
DATA
FROM
ENGLAND
AND
WALES*
CHRIS HALE
1,
PAT
PACK and JOHN SALKED2
~
Faculty
of
Social Sciences, University of Kent, Canterbury, Kent CT2 7NX, U.K.
Applied Statistics Research Unit, University
of
Kent, Canterbury, Kent CT2 7NF, U.K.
ABSTRACT
The research presented
in
this article develops the explanation
of
fear
of
crime beyond that which
concentrates upon the characteristics and attributes
of
individuals to consider structural or neigh-
bourhood level determinants. After reviewing the theoretical arguments for such an approach, an
empirical model which combines neighbourhood level data constructed from the
1981
British
Census with individual level data from the 1984 British Crime Survey
is
presented. The results
support the contention that neighbourhood structure
is
an important factor when discussing fear
of
crime.
INTIWDUCTION
This work extends the recent logic
of
developing contextual models
of
victimi-
zation to the analysis
of
fear of crime. The notion that researchers must look
beyond individual characteristics and, as well, seek the determinants
of
fear at
the neighbourhood or community level is not, at the theoretical level, particularly
novel. However it remains the case that when attempting to explain variation in
levels
of
fear, empirical work using British data has relied upon individual level
measures from victimization surveys. The research presented below is a prelimi-
nary attempt to address this shortcoming. Information from the 1981 British
Census is used to construct theoretically relevant community level variables,
which are then combined with individual level data taken from the 1984 British
Crime Survey (BCS). It therefore complements work which has used the same
data sources and similar methodology to model crime victimization (Osborn et
al., 1992; Trickett et al., 1992). The results show that neighbourhood effects are
indeed discernable, a finding already noted with a smaller data set from a U.S.
*This
work
is
based on research funded by the United Kingdom Economic and Social Research
Council (ESRC) reference number R00023121
0.
An earlier version
of
this work was presented to
the annual conference
of
the American Criminological Society, San Francisco, November 1991.
The authors would like
to
thank Andrew Dart, of the Kent Comparative Criminal Justice Project,
for helpful comments on earlier drafts
of
this paper and Pat Mayhew,
of
the Home Office Research
and Planning Unit, for the data from, and advice on, the 1984 British Crime Survey. Finally the
paper has benefited greatly from the comments
of
anonymous referees. Any errors remain,
of
course,
the authors' responsibility.
212
city by Covington and Taylor ( 1991
).
The present study is however the first, with
a national level data set, which includes both suburban and rural as well as urban
settings, to show that structural contextual variables contribute substantively to
explaining fear.
As Smith ( 1989) points out, variability in the extent and distribution
of
fear
of
crime has been observed at all spatial scales and will be dependent upon the
cultural climate
of
a country, the social climate
of
a community and the emotional
state
of
individuals. Within England and Wales, for example, there are wide
regional differences in the estimated proportions
of
those who feel a bit
or
very
unsafe 'walking alone in this area after dark'. Figures from the 1982 BCS range
from a low
of
26% in East Anglia up to 42% in London (Smith, 1989; Table 10.1
).
Examining these regional variations suggests that, in addition to seeking expla-
nations for the prevalance
of
crime at a neighbourhood or community level, fear
of
crime might also exhibit a neighbourhood effect. Indeed, the social disorgan-
ization theory
of
Shaw and McKay (1942), whilst presented as a means
of
explaining crime and delinquency, appears equally apposite to examining fear.
In a recent article which provides an elegant test
of
this theory using data from
the 1982 and 1984 BCSs, Sampson and Groves (1989) comment that
'In
general
terms, social disorganization refers to the inability
of
a community structure to
realize the common values
of
its residents and maintain effective social controls.'
Particularly,
if
fear
of
crime is considered as a measure
of
individuals' unease
with their local environment and community, we would suggest that social
disorganization may be expected to have an impact upon it independent
of
its
effect upon actual crime levels.
There
is
evidence in the literature that anxieties related to a deterioration in the
local environment, a worsening
of
community life, a sense of social isolation and
a lack
of
local economic and political power appear to be displaced on to the
problem
of
crime. (Merry, 1981; Hunter and Baumer, 1982; Skogan and Max-
field, 1981; Smith, 1983, 1986). Lewis and Salem (1986) also stress the import-
ance
of
community informal social control in the development
of
fear
of
crime.
They suggest this is more difficult to maintain in residential areas which are
disorganized due to social change or social marginalization. Fear
of
crime is more
than a response to threatened or actual victimization; it is
'a
judgement that the
government and the social structure will not be able to provide the collective good
of
safety.' (Taylor
eta/.,
1986;
p.
176). This can be expected to be particularly
true in a society where the Government has made much
of
the issue
of
law and
order as a potent symbol
of
moral and economic decline (see Brake and Hale,
1991). In the words
of
Smith (1989;
p.
198) '
...
fear
of
crime may be concep-
tualized as an expression
of
the sense of powerlessness and uncertainty that
accompanies much
of
urban life.' Indeed,
if
we examine the intervening con-
structs which Shaw and McKay suggest are important in their model, we find that
at least two are related directly to ones commonly used in empirical work
on
fear
of
crime. First, the ability
of
a community to supervise and control teenage peer
groups
is
obviously connected to the incidence
of
groups
of
teenagers hanging
To continue reading
Request your trial