See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/228194423

Legitimacy and the Social Field of Policing

Article  in  SSRN Electronic Journal · August 2011

DOI: 10.2139/ssrn.1914458

CITATIONS READS
2 159

2 authors:

Ben Bradford Jonathan Jackson

University College London The London School of Economics and Political Science
197 PUBLICATIONS   6,165 CITATIONS    258 PUBLICATIONS   10,809 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Procedural justice policing View project

Assessing the broad impacts of neighbourhood crime on the mental health of marginalized and vulnerable populations View project

All content following this page was uploaded by Jonathan Jackson on 14 December 2017.

The user has requested enhancement of the downloaded file.

 Legitimacy and the Social Field of Policing

Ben Bradford*, Methodology Institute and Mannheim Centre for Criminology, LSE
Jonathan Jackson, Methodology Institute and Mannheim Centre for Criminology, LSE

*Ben Bradford, Methodology Institute, LSE, B808, Columbia House, Houghton Street, London
WC2A 2AA, United Kingdom. b.bradford@lse.ac.uk

Abstract

Public actions that summon and assist police officers are vital for the effective and equitable functioning
of the criminal justice system. Such acts of cooperation link informal and formal mechanisms of social
control. They also reflect the legitimacy of the police. This paper reports the findings of a probability
sample survey investigating public trust, institutional legitimacy and cooperation with the police in
London, England. We find that the readiness of individuals to cooperate with the police is associated with
their trust in police procedural fairness and the legitimacy they invest in the police. This replicates prior
US-based research in the UK for the first time, but we also highlight some predictors of public
cooperation that have hitherto gone unexamined. Linking police legitimacy and cooperation to perceived
collective efficacy (confidence that local residents will intervene on behalf of the collective good),
concerns about disorder, and authoritarian positions on the perceived loss of discipline in society, we use
Bourdieu’s notions of field and habitus to highlight the role that the social position of the police plays in
generating public cooperation. Police legitimacy is shaped by its location in the social ‘field’ of policing
and the web of relationships that link organization and citizen.

Key words: cooperation, procedural justice, trust in the police, police legitimacy, formal and informal social
control

Acknowledgements: Many thanks to Betsy Stanko and the London Metropolitan Police for allowing us
continued access to their data. A previous version of this paper constituted part of the lead authors PhD,
which was funded by the Economic and Social Research Council.

Electronic copy available at: http://ssrn.com/abstract=1914458

The cooperation of the public is vital for the effective and equitable day-to-day functioning of the criminal
justice system. Many criminal offences become known to the police because they are identified by a
member of the public. Cooperation from citizens – as witnesses, jurors, or in other roles – is then required
throughout the criminal justice process (Goudriaan et al., 2006). To summon officers, to report crime or
suspicious activities, to provide information to help police identify a criminal – these are acts that bridge
formal and informal mechanisms of social control, engaging both citizen and police officer in the activity
of policing and regulation (Bursik and Gramsik, 2003). The justice system relies upon such cooperation:
these are acts that both legitimize the police and criminal courts and aid them in the fight against crime.
This study contributes to a growing literature on procedural justice and legitimacy. Examining
propensities to cooperate with the police in a British context, we test a series of hypotheses that revolve
around Tyler’s process-based model of policing. Exploring the legitimizing nature of cooperation, we
assess whether variation in intentions to cooperate with the London Metropolitan Police can be explained
by variation in the perceived trustworthiness of officers (particularly whether they can be trusted to treat
citizens fairly and take heed of the needs of local communities). We also extend the procedural justice
model of policing by locating the legitimacy of the police as an organization within a broader
understanding of policing as a social institution. We examine whether (a) subjective levels of collective
efficacy in the community, (b) public concerns about neighborhood breakdown, and (c) people’s anxieties
about moral decline in society are important in the relationship between public and police. Drawing on the
notion that the British police powerfully represent community and order (Reiner 2000; Loader and
Mulcahy 2003), we test whether individuals who believe that local people would intervene on behalf of
the collective good are also more likely to invest trust and legitimacy in the police and cooperate more
readily. We examine whether individuals who see problematic disorder around them tend to invest less
trust and less legitimacy in the police. And we explore whether authoritarian concerns about morality and
social discipline positively predict conferred legitimacy and intentions to cooperate.
The paper proceeds in five parts. We argue first that the activity of policing – engaging in acts of
social control – can usefully be seen as a ‘field’ in Bourdieu’s (1993) sense, and that the police are an
institutionalized organization within this field. Locating the police in the social field of policing opens up
the possibility that public opinions about and orientations toward the police do not rest only on issues of
procedural fairness: the socio-structural position of the police may be such that it is legitimated on other,
quite different, grounds. Second, we outline a concept of legitimacy that draws on – but also departs from
– that used in the procedural justice literature (see, for example, Sunshine and Tyler 2003a), and reframe
the predictions of procedural justice in relation to this concept. Third, we develop the idea that public
legitimations of the police may be founded in part on ideological positions and the need to exert informal
social control, criteria that only tangentially connect with the performance of the police as an organization.
Throughout the second and third sections of the paper we define a number of hypotheses: in the fourth, we
test these hypotheses against data from a representative population sample of Londoners (n= 5,120). We
close with some thoughts concerning implications for the procedural justice model and for our
understanding of the processes that reproduce, or undermine, the legitimacy of the police.
To anticipate the main findings, we find that Tyler’s procedural justice model of policing works
well in London. But it can also be built upon in ways that recognize the position of the police within the
field of social activity concerned with social control. Police legitimacy may be affected by a complex set
of processes that are distinct from issues of procedural fairness. Because it is tasked with enforcing social
control and dominant versions of normative order, police legitimacy may suffer when society is thought to
be fragmented and disorderly. Conversely, the police may garner legitimacy, and individuals may be more
likely to bridge informal and formal social control mechanisms by cooperating with the police in the fight
against crime, when resources of collective efficacy are seen to be high. But, equally, the social position
of the police may be reinforced by ideological viewpoints relating to the need for order and high levels of
social control: for some people, it may serve as a bulwark against threatening change. We find evidence of
strong associations between public opinions of the police and perceptions of the failure and success of
formal and informal social control processes (c.f. Skogan 2009).

POLICING AS A ‘SOCIAL FIELD’

Sociologists and criminologists have long struggled with a consensual definition of ‘police.’ The word is
both a noun and a verb, representing an organization, an institution, and a set of social practices. In this
paper we draw a strong distinction between policing, a human activity aimed at regulating the social

Electronic copy available at: http://ssrn.com/abstract=1914458

environment in line with dominant conceptions of normative order, and the police as an organization
specifically created for this task. The police have a certain set of exclusionary rights over the activity of
policing. Yet, despite its monopoly on the use of legitimate force, it is by no means the only agent
engaged in this endeavor. In the simplest of terms, all societies have had some kind of policing. But only
very few have had police, and the institution of policing has an existence prior to and beyond the police
organization.
Almost everyone, at some time, in some way, engages in activities aimed at reproducing social
order – that is, in acts of social control. Bursik and Gramsik define three types of such ‘informal’ social
control (1993; see also Warner, 2007). First, private social control is embedded in the relationships
between family and close friends. Second, parochial social control is exerted by more diffuse networks of
people operating within geographically and/or socially bounded areas. Third, indirect informal social
control (or what Warner calls public control) is bound up in the ability of individuals and social networks
to ‘secure public goods and services that are allocated by agencies located outside the area’ (Bursik and
Gramsik, 1993: 17; quoted in Warner, 2007: 101). This third form of social control implicates public
engagement with the police and the co-production of policing activity. But we can also draw on Carr’s
(2003) further blurring of the line between parochial and public control. Some forms of social control
operate across the boundaries of the informal and formal realms. His ‘new parochialism’ describes
situations in which individuals call upon and cooperate with agents of formal social control. Creating ‘a
partnership between parochial and public spheres’ (ibid: 1252) – wherein these different types of control
are not separate from each other but intimately linked – this type of social control is initiated at the
personal level and implemented at the public, or formal, level. Carr thus builds on earlier formulations that
also stressed the mutual interdependence of the parochial and public orders and the forms of social control
they maintain (Hunter, 1995: 221). Social control in this form directly implicates public cooperation with
the police.
Both police and public are therefore engaged in the activity of policing, and seen in this light
policing is a field (Bourdieu 1993; Emibayer and Johnson 2008), a structured space of positions or roles,
occupied by a variety of actors, which defines specific stakes and interests (Bourdieu 1993: 72). The field
of policing presupposes the desirability of ‘law’ and ‘order’. While those holding positions within it may
compete to maintain or change their place (different agencies struggle for ‘market share’, members of the
public attempt to influence the police to take certain actions or avoid others), they also share a common
understanding of the rules, and purpose, of the activity of policing (ibid.). The field structures the
understandings and activities of those operating within it. It makes ‘the meaning and value of the
properties objectified in things or embodied in persons’ (Bourdieu and Wacquant 1992: 23). The positions
occupied by actors in the field of policing influence how they conceive of crime, disorder and social
regulation and how they structure their possibilities of action in relation to these phenomena. Yet, their
decisions to act will be influenced by individual belief, circumstance and propensity, and the concept of
habitus as “the strategy generating principle enabling agents to cope with unforeseen and ever-changing
situations” (Bourdieu 1977: 72) is the indispensable relational other of the field. The field of policing
defines the social space in which actors engage in acts of social control aimed at maintaining law and
order, and where they cooperate (or not) with the police; the dispositions engendered by the habitus they
carry – as a social and psychological toolkit or set of ways to see the world – sensitize and mobilize them
in particular ways (Bourdieu and Wacquant 1992).
In Bourdieu’s usage, the field is envisioned as a site of conflict, as individual, organizational and
group actors use their economic and symbolic capital to either maintain or disrupt existing power
structures. Naturally, issues of conflict are intricately bound up with policing as a social institution, the
activity of the police as an organization, and the relationship between police and people. But the field of
policing – in Great Britain at least – may be more limited, and limiting, than Bourdieu’s general
conceptualization. The police has been characterized as an institutionalized organization (Crank 2003; c.f.
Meyer and Rowan 1977). As an organization it reflects (even embodies) the values of its institutional
environment – that is, the field of policing – to a degree uncommon in other areas of social life. In
addition, unlike many other fields and even though there is some competition for the police from what
might be termed alternative security providers (Bayley and Shearing 1996), in Britain at least the police
are by far the most important organization involved in policing. Under such conditions the organizational
boundary blurs and dissipates and, at the extreme, the organization becomes a dramatic enactment of the
myths pertaining to the social structures in which it is embedded (Meyer and Rowan 1977: 346).

3
 Crucially for what follows, the literature on institutionalized organizations suggests they are
legitimated on bases other than formal effectiveness. External and ceremonial elements become especially
important. Institutionalized organizations also gain legitimacy from their social fitness, i.e. the extent to
which their form is regarded to fit the environment in which they operate (Scott 2001). Put another way,
institutionalized organizations are legitimated by the publics they serve less on the grounds off efficiency
and more on the basis of their mutual relationships, and by the apparent isomorphism of organization with
institutional structures. On this account police legitimacy will certainly rest in part on what it actually
does, particularly in relation to the symbolic, relational and representational aspects of its behavior; in
particular, the fairness with which it is perceived to treat people (Tyler and Huo 2002). But the legitimacy
of the police may also be influenced by its dominant place within the policing field and on its location
within a set of social positions that predispose certain actions and stances toward it. Because the police is
so closely associated with the activity of policing people who wish or see a need to engage in certain types
of social control activities may be pushed into engagement with it for reasons that do not directly
implicate their opinion of it as an organization, in terms of either its effectiveness or the fairness of its
behavior. Seen from some perspectives ‘police’ and ‘policing’ are one and the same, and the isomorphism
of organization and field (Meyer and Rowan 1977) may mean that to perceive a need for social control is
also to perceive a need for the police. Furthermore, when people think about ‘doing’ policing – when they
think about engaging in acts of social control – their possibilities of action will be strongly influenced by
the existence, role and power of the uniformed police. The police organization may provide a common
frame of reference around which definitions of crime, disorder and particularly the nature of desirable
action about these issues are formed, and it draws its legitimacy in part form its inevitable presence within
these definitions (Scott 2001: 61).

LEGITIMACY, COOPERATION AND EXPRESSED CONSENT

The notion of legitimacy is central to this study. Theorists from Weber (1978) onwards have viewed
legitimacy as a vital component of social institutions, both in the long run for their very survival and for
their successful functioning day-to-day basis. Individuals defer to – and cooperate with – legitimate
authorities because they feel it is right to do so (Tyler 1990; Sunshine and Tyler, 2003a; 2003b).
According to Sunshine and Tyler (2003a), legitimacy is:

‘a property of an authority or institution that leads people to feel that authority or

institution is entitled to be deferred to and obeyed. It represents an ‘acceptance by people
of the need to bring their behavior into line with the dictates of an external authority’.’
(Sunshine and Tyler 2003a: 556; quoting Tyler 1990: 25)

On this account, police legitimacy resides primarily in the perceived obligation to obey the commands of
officers. Much of the extant empirical evidence links public cooperation to the legitimacy of the police
and courts via the experience of procedural justice (see: Tyler, 1990; 2004; 2006; Tyler and Blader, 2000;
Tyler and Huo, 2002; Tyler et al., 2010). Procedurally just and fair treatment at the hands of authorities is
associated not only with satisfaction with decisions and decision-makers, but also (via the feeling that it is
right to obey the directives of the police) with increased propensities to offer assistance in the future.
A core premise of Tyler’s procedural justice model is that the experience of procedural fairness
fosters in people feelings of motive-based trust in (and shared group membership with) the authority
involved. On the one hand, fairness encourages the idea that citizens and the police are ‘on the same side.’
By treating people justly and equitably, police communicate to citizens that they are valued members of
the social group that the police represent (Tyler, 1990; Tyler and Blader, 2000) – a group that in the
British context can be conceptualized very broadly as the nation, state, or community (Jackson and
Bradford, 2009). The experience of fairness activates roles and responsibilities attendant to individual’s
relationship with the police that underline their responsibility to offer assistance and which suggest the
possibility of working with officers in co-producing the type of order presupposed by dominant notions of
policing (c.f. Kelman and Hamilton 1990). A police force legitimate on grounds of its procedural fairness
activates within those subject to it roles – ‘citizen’, for example – that encourage a sense that it is right and
proper to assist the police in its task. On the other hand, unfair treatment communicates division, social
denigration and exclusion, thus fostering an ‘us and them’ situation, meaning police legitimacy – and
therefore cooperation – may decline. Here, different roles and relationships with the police are activated,

4
and the reproduction of ‘law and order’ becomes a site of contest rather than cooperation. Finally, the
empirical work that supports Tyler’s thesis consistently finds that ideas about police effectiveness have
much weaker associations with its legitimacy. People appear to care much more about how they treated by
authorities than they do about the material outcomes those authorities may be able to provide them. Or, to
put it another way, the relationship between police and public is defined much more by active processes of
trust and engagement than by a passive regard of one by the other.
In Tyler’s research, a sense that the police are legitimate (as measured most importantly by the
perceived duty to obey officer’s instructions) leads to acts of cooperation. Perceptions of legitimacy are
empirically and conceptually distinct from intentions to cooperate. A subtly different idea of legitimacy
can be found in the work of David Beetham (1991). He insists that legitimacy is not simply a property of
an authority, or something which exists only in the subjective experiences of citizens, but is also
constituted by normative assessments of authorities and the actions of those they govern. When citizens
grant legitimacy to the police they do so on the basis that the relationship between individual and authority
is founded on common shared values. When individuals act in ways congruent with recognition of this
commonality, their behavior serves to confirm and reproduce the legitimate role of the authority. Beetham
therefore outlines three dimensions of legitimacy, each of which must be fulfilled for an authority to be
considered legitimate. First is conformity to a set of rules. Second is the justifiability of these rules in
terms of shared beliefs. Third is the expressed consent of those governed or otherwise affected by the
authority.1
There is, then, a certain level of conceptual ambiguity over whether cooperation with the police is
an outcome of its legitimacy or in part constitutive of it. To some – such as Tyler – the legitimacy of an
authority resides primarily in the views and orientations of those subject to it, and most specifically in the
idea or belief that the instructions of the authority should be followed regardless of their empirical content
(the ‘duty to obey’). To others – such as Beetham – legitimacy is constituted in and through the
relationship between authority and individual. While Beetham stresses the importance of citizen’s sense of
moral alignment between themselves and the authority (that is, individual’s perceptions of authorities) he
also states that the expressed consent of the governed is necessary if a particular social arrangement is to
be considered legitimate. One interpretation of this theory is that the legitimacy of the police rests most
importantly on citizen’s sense of moral alignment with it. Feeling that the police share their own norms
and values encourages individuals to cooperate with officers.
In this paper we apply Beetham’s work to the context of policing. We still consider cooperation as
a function of perceived legitimacy. Like Tyler, we specify cooperative acts as flowing from the belief that
the police have authority and act legally an in accord with shared moral beliefs. But these cooperative acts
are also part of a broader process model of legitimacy (c.f. Johnson et al. 2006). Expressed consent might
by found not only in acts of blunt obedience to officer’s instructions, but also in general public deference,
in offers of assistance, and in acceptance (indeed encouragement) of officers’ presence at specific
moments. While Tyler treats cooperation as an outcome of legitimacy, a broader view treats moments of
cooperation as both an outcome of the conferral of authority and as an act of legitimation. Cooperation
entails the recognition of the right of the police to act in, and gain control over, certain social situations.
Cooperative acts – such as calling upon and assisting the police – place obligations on officers as well as
citizens. Recognizant of the relationship between police and public, moments of cooperation serve to
define and delimit the role, rights and duties of both parties involved. In sum:

‘…what is important for legitimacy is evidence of consent expressed through actions

which are understood as demonstrating consent within the conventions of the particular
society … these actions … confer legitimacy; they contribute to making power
legitimate. They do this both through the public demonstration of people’s consent to
the power relationship, and through the resulting obligations that derive from them on
the part of both dominant and subordinate alike. They possess simultaneously a
symbolic and a normative force.’ (Beetham, 1991: 12, emphasis in original)

AN EXTENSION TO THE PROCEDURAL JUSTICE MODEL

Thus, far from rejecting Tyler’s work on police legitimacy we extend aspects of Sunshine & Tyler’s
(2003) framework. Positioning police and public within a set of social relations, we propose that the
legitimacy of the police (as experienced subjectively by members of the public) can be thought of as

5
residing in individual’s sense of broad moral alignment with officers and organization and also in their
expressions of consent to the role of the police, e.g. obligation to obey police directives and willingness to
cooperate with the police in the fight against crime. 2 The acts of cooperation we address in this paper
reflect citizens’ recognition of the police role as a formal agent of social control (authorized to investigate
crimes that have taken place and then to detain and charge suspects) as well as an indication of support in
these endeavors. In line with Sunshine & Tyler’s (2003) study, we hypothesize that a perceived duty to
obey will influence propensities to cooperate. But, following Beetham, we also hypothesize that moral
alignment will be the fulcrum of police legitimacy, strongly influencing both obedience and cooperation.3
Tyler links trust in police procedural justice to legitimacy defined as the duty to obey. We add to this the
expectation that trust in the police will be linked to consent expressed by willingness to cooperate with the
police. By using self-assessed propensity to cooperate with the police as our outcome variable, we
(necessarily inexactly) measure an additional aspect of ‘the public demonstration of people’s consent to
the power relationship’ (Beetham 1991: 12).
Of the first three hypotheses that we test in our London-wide probability sample H1 and H2 reflect
the replication of Sunshine & Tyler’s (2003) model. H3 represents our extension to the model:

H1 Those who perceive the police to be procedurally just and fair will be more likely to confer
authority, feel morally aligned with the police, and hold stronger intentions to cooperate.
H2 Those who perceive the police to be effective in cutting crimes and carrying out their routine duties
will not grant any more or any less legitimacy or intentions to cooperate, net of levels of trust in
procedural fairness.
H3 Moral alignment and duty to obey will both predict intentions to cooperate, but moral alignment
will be the most important predictor of cooperation.

COOPERATION WITH THE POLICE AND THE STRUCTURE OF THE POLICING FIELD
So far we have considered the intention to cooperate with officers as an act of expressed consent that both
flows from and helps constitute the legitimacy of the police. But a sense of moral alignment and duty of
obedience, based primarily on assessments of the fairness of police activity, may not be the only factors
that shape public cooperation. Expressions of consent – acts that simultaneously recognize and confer
police legitimacy – may have other antecedents that reflect the social, cultural and indeed political position
of the police within the broader field of policing. In this study we measure cooperation by considering the
likelihood of calling the police to report a crime occurring in the neighborhood; the likelihood of helping
the police find someone suspected of committing a crime (by providing information); and the likelihood of
reporting dangerous or suspicious activities in the neighborhood (see below). Understood in these terms,
cooperation is an act that must – in part – be related to the perceived need for police intervention and the
desire to exert social control of the type envisaged by Carr’s notion of new parochialism. The generation of
a desire to exert social control via the police may be quite distinct from assessments of the behavior of the
organization itself. It follows that intentions to cooperate with police – and its legitimacy – may be
associated with social conditions that lead individuals to feel such need and their ideological orientation to
the world around them.

COOPERATION AND COLLECTIVE EFFICACY

One path to cooperation with the police may arise when individuals believe that their neighborhood lacks
collective efficacy, i.e. when they believe that it lacks the resources to regulate itself (cf. Sampson et al.,
1997). When people think that informal mechanisms of social control have been attenuated – and when
they judge local order to be under threat – they may be more likely to see a need for the police. More
broadly, the extent to which individuals are willing to contact and cooperate with the police will be related
to their normative assessments of the area in which they live and those they share it with – what ought or
ought not to occur, what should be done about deviancy, and whether it is worth getting involved. To
summon officers, to report crime or suspicious activities, to provide information to help police identify a
criminal – these are acts that imply active recognition of the propriety of the police remit over matters of
crime and disorder. By engaging both citizen and police officer in the activities of policing and regulation
(Bursik and Gramsik, 1993; Carr 2003) they also bridge formal and informal mechanisms of social control.
What then is the link between perceptions of the extent of informal social control in local areas and
intentions to cooperate with the police? An individual’s desire and propensity to engage in social control is

6
part of the habitus they carry – a set of durable dispositions regarding action (Bourdieu 1993: 86) – and this
propensity to act will be channeled in specific ways by entry into the field of policing. The social space
defined by the field and the relationships that comprise it offer distinct – and limiting – possibilities for
action for those who wish or see a need to engage in acts of social control. People may turn to formal
agents when informal social control mechanisms are perceived to be weak (or nonexistent). Net of other
relevant factors, when individuals feel that their local community does not have the resources to achieve
certain outcomes for the collective good, they may be more likely to turn to the police as a source of social
control, since this is the only viable option to achieve the outcomes they desire. This leads us to a fourth
hypothesis:

H4 Perceptions of collective efficacy (measured here as the combination of perceived social cohesion
and perceived informal social control, see: Sampson et al., 1997; Bursik, 1988; Bursik and
Grasmick, 1993; Sampson and Raudenbush, 1999) will, net of relevant controls (e.g. fear of crime
and perceived levels of crime and disorder), be negatively associated with intentions to cooperate
with the police. Citizens turn to the police especially when they believe that the local community
has few collective resources for self-regulation.

There is, however, an alternative formulation. Individuals who feel they live in communities where other
citizens will intercede when, for example, children are misbehaving, may be on average more likely to
trust the police (Girling et al., 2000; Jackson and Bradford, 2009). The British police – as important group
representatives – seem to garner support when local communities are seen to be cohesive, efficacious and
‘pulling in the same direction’. Because it is an institutionalized organization, so strongly associated with
the maintenance of social order (with the activity of policing), it is possible that individual-level trust in
police is fostered in part by a sense of community cohesion and collective efficacy based on a view that
social order is, in fact, being maintained. Trust garnered in this way will flow into legitimacy and
therefore decisions concerning whether to involve officers at times of perceived need. Furthermore the
legitimacy of the police, as a symbol of the community, may rest in part on such foundations. As Reiner
(2000) notes, the police may be seen as successful when their presence is not needed, and it may garner
legitimacy from the perceived success of wider processes of social control. Hypothesis five therefore
reverses hypothesis four:

H5 Perceptions of collective efficacy will be positively associated with intentions to cooperate with the
police: citizens are more willing to cooperate with the police when they believe their community is
cohesive and shares common values. This effect should be mediated in large part by individual’s
trust judgments concerning the police.

Note here the stress on individual-level action. Criminological research often measures social
control mechanisms at the community level. Some studies predict outcomes such as crime rates by using
neighborhood-level measures of social control aggregated up from individual survey responses (see
Sampson et al., 2002). Other studies mix individual and area-level measures (ibid.). The relative strengths
or weaknesses of social control mechanisms are in turn related to factors such as the social composition of
an area, population stability, relationships with the police, and social cohesion and trust, again often
aggregated up to the neighborhood-level (Sampson and Bartusch, 1998; Sampson et al., 1997; Silver and
Miller, 2004; Warner 2007; Wells et al., 2006).
By contrast, we are concerned here with the perceptual and cognitive factors that influence
decisions by citizens to invoke the police as agents of formal social control. The approach of this paper
therefore differs in important ways from studies that have measured neighborhood-level processes (for
example, Sampson and Bartusch, 1998; Silver and Miller, 2004). At the neighborhood level, aggregate
satisfaction with the police has been found to be linked to aggregate levels of informal social control:
support for the police is held to influence informal social control processes (Silver and Miller, 2004), the
reverse of the relationship specified here. In other regards, however, the current approach echoes results
from other studies which have looked at the effect of community level characteristics on individual
behavior. In a study which examined the associations between area-level characteristics and crime
reporting, Goudriaan et al. (2006) found that high social cohesion, measured at the aggregate level, was
associated with higher chances of reporting crime victimization at the individual level. Similarly, Warner

7
(2007) reports a strong association between aggregate social cohesion and trust and individual
propensities to engage in ‘direct’ social control (self-assessed readiness to involve landlords, police or
others in neighbor disputes).

COOPERATION AND SOCIAL THREAT

The way people think about the police – what it represents to them, what they think it is for, how it relates
to their experiences – may then have a powerful effect on the legitimacy they grant it. On this, British
criminologists have developed a range of distinctive accounts concerning the cultural significance of the
police. In much of this work the English police (in particular) are held by citizens to be representatives of
law and order, the nation-state, respectability, and even a certain form of Englishness (Girling et al., 2000;
Loader, 2006; Reiner, 2000; Smith, 2007; Waddington, 1999). We might ask, therefore, what opinions,
outlooks or ‘structures of feeling’ (Williams, 1964; see Loader and Mulcahy, 2003) are implicated by
adherence to the police as representative of, for example, a stable, cohesive national past (Girling et al.,
2000; Reiner, 2000)? This is relevant to the current study because such representations will influence the
social position of the police and the ways in which people (imagine) they might interact with it. What do
prevailing representations of the police imply in terms of the policing field, the dominant habitus
concerning police and policing, and the legitimacy of the police organization? We assess in this study the
roles variously played by (a) perceptions of neighborhood disorder and incivility; and (b) authoritarian
stances towards morality and discipline
It is something of a paradox that when British people think about the police, issues of crime per se
may not be uppermost in their minds. Indeed, it seems that worry about crime has only a tangential
connection with public assessments of the police (Jackson and Bradford, 2009; but see Gray et al. 2011).
Rather, it is the deeper social concerns that may underlie people’s ‘crime talk’ (Sasson, 1995) that seem
most important. A growing body of work (for example Girling et al., 2000; Loader and Mulcahy, 2003;
Jackson and Sunshine, 2007) stresses that when people think about the police and their ‘crime-fighting’
activities, they also think about what ‘crime’ stands for (erosion of norms and social ties that underpin
group life) and what ‘policing’ stands for (organized defense of those norms and ties). Individuals who are
concerned about local neighborhood disorder and broader long-term social change, who perceive a
modern world in moral and social decline, or who buy into a ‘community lost’ narrative, may blame the
police as representatives of the social order that allows these things to happen (Jackson and Bradford,
2009).
In the British context concerns about low-level disorder and anti-social behavior are often held to
reflect fears about the breakdown and fragmentation of society (Girling et al. 2000). Perceptions of
disorder and may thus damage confidence in police effectiveness, fairness and group engagement
(Jackson and Sunshine, 2007), and therefore, perhaps, police legitimacy. Such a perspective positions the
police as exactly the kind of ‘proto-typical group representatives’ (Sunshine and Tyler, 2003b) envisaged
by the procedural justice model. It also underlines that, since the field of policing is a site of struggle as
much as it is an arena of cooperation, disagreement and dissent to the role of police can also arise from its
position and role within the field.
The key idea here is that citizens hold accountable group authorities that are perceived to let the
norms, values and standards of public behavior erode ‘on their watch’. Police legitimacy will suffer when
the social world is perceived as disordered and heading in the wrong direction. Concerns about local
disorder are also thought to be strongly linked to informal social control (Carr, 2003; Sampson and
Bartusch, 1998; Sampson et al., 1997; Warner, 2003, 2007; Wells et al., 2006), with high levels of
disorder associated with lower propensities to engage in informal social control. Hypothesis six is that
concerns about neighborhood disorder are associated with low levels of trust and a lack of moral
alignment between police and citizen.

H6 High perceived disorder will be linked with lower propensities to cooperate. Much of this
association will be mediated by trust and legitimacy as moral alignment. Holding the police to
account for poor social conditions, those who perceive a situation of neighborhood disorder and
decline will report lower levels of trust in the police and lower levels of moral alignment with the
police.

8
 However, a contradictory process may again be at work. Consider what perceptions of disorder –
as opposed to the objective condition of disorder – represent. Perceived disorder may express a sense that
society is threatened and that negative change is occurring. Underlying individual’s propensity to see
disorder may be a deeper sense of social malaise, and those who perceive social and moral order to be in
decline (and that the established order is under threat) may cleave more strongly to the police. They may,
for example, see it as bulwark against this decline and therefore deserving of support. In addition to
perceptions of disorder, therefore, we also require a measure of people’s general sense that society is
‘heading in the wrong direction’ (or that it is not). In fact, the measures we use to capture perceived loss of
morality and discipline in society are similar to the sort of items that political psychologists routinely use
to measure authoritarianism (see below); it may therefore be that our measures capture both the perceived
loss of morality and also a kind of submission to authority and support for authority figures. Seen in these
terms, authoritarianism is an aspect of the individual’s habitus, part of the psychological toolkit they use to
understand and act in specific social settings. Those who think that many things are happening which
ought not to be may identify more with the police, and they may be more ready to invoke the police to
correct what they see to be a deteriorating state of affairs. Indeed, given the peculiarly close relationship
between the British public and its police force, vestiges of which remain even after a long relative decline
in trust and confidence (Loader and Mulcahy 2003; Reiner, 2000), it might be expected that this ‘positive’
relationship might be at least as common as the ‘negative’ one outlined above that relates to perceptions of
local disorder. While there is much in the literature to support this idea, in some cases at least (Girling et
al., 2000; Loader and Mulcahy, 2003), there has so far been little empirical work to examine possible
mechanisms for such a link. Our seventh hypothesis is therefore:

H7 Those who perceive a situation of moral decline and who value strong authority in society will
profess a greater readiness to assist the police. Controlling for other relevant factors, they will on
average hold the police to more legitimate because they see it as a bulwark against social decline.

SUMMARY
We thus consider a complex web of factors that may be related to people’s willingness to contact and
cooperate with the police that reflect above all the nature of the relationship between police and public. Our
hypotheses are framed by Tyler’s work on procedural justice and legitimacy but they also bring in the
structural position of the police within the policing field, the nature of its relationships with citizens, and
the habitus – a set of understand and act in within the policing field – that these citizens carry and which
predisposes them toward or certain relationships with the police. Figure 1 provides a conceptual map – a
purely stylistic way of presenting the suggested relationships and of ordering the observational data (see
Figure 3 below). In reality, things are likely to be considerably more complicated. In particular, both model
and reality will be reciprocal with, for example, levels of trust in the police reflecting back over time onto
ideas about crime and disorder (Skogan, 2009). But for now, our objective is more modest: to test, using
cross-sectional data, one plausible way of specifying the particular patterns of partial associations.

Figure 1 near here

DATA AND METHODOLOGY

DATA
The London Metropolitan Police Service Public Attitudes Survey (PAS) is a rolling 12 month survey with
an annual sample size in excess of 20,000 respondents. We call upon data from the first quarter of the
2009/10 survey (n=5,120). Respondents were from a random sample representative of residents aged 16
and over. The sample is drawn across the capital, resulting in a dataset in which respondents are clustered
in 593 electoral wards (electoral ward being one of the basic administrative units in the capital). Topics
covered in the PAS include public confidence in the police, perceptions of crime and disorder, attitudes
towards and contact with the police, victimization, and the fear of crime. The structure of the sample is
shown in Table 1.

Table 1 near here

9
MEASURES
Figure 2 above demonstrated that a number of variables are needed to investigate the issues at hand. The
ultimate response variable (we use structural equation modeling, and there are a number of response
variables in our model) in the analysis is propensity to cooperate with police. Prior to this sit the other
aspects of police legitimacy, trust judgments, views of local social cohesion and disorder and, ultimately,
respondent’s assessments that the social order is under threat. Opinions of and worry about crime (see
Farrall et al., 2009, for a recent review of the fear of crime literature and discussion of the meaning and
measurement of ‘fear’) constitute control variables in the analysis. To represent the key sets of public
opinions required we estimated a simultaneous confirmatory factor analysis model with ten latent
constructs and a total of 44 separate indicators, using MPlus (version 5.2) and with the indicators
identified as categorical variables. Factor loadings and fit statistics are shown in Table 2. The model fit
was adequate.

Table 2 near here

Some of the latent constructs shown in Table 2 had relatively high covariances, which might
imply that the separate latent constructs involved were misspecified and should be combined. The largest
such covariance was between ‘moral alignment with police’ and ‘perceived duty to obey’ (.60); that is,
between two components of police legitimacy. However, a second, nine factor SCFA model which
combined these two constructs fitted the data significantly less well, according to the exact fit statistics
(difference in X2 values = 2,036). We therefore retained the ten separate measures in the subsequent
analysis.
Some discussion of the nature and content of the key measures is also in order. Firstly, we began
by measuring confidence in police fairness (procedural justice) and confidence in police engagement with
the community separately (Bradford et al., 2009; Jackson and Bradford, 2009). Yet, while these can be
seen as conceptually distinct ideas – and are treated as such in the work of Tyler and others – the measures
were so highly correlated there was little option other than to treat them as one (or else suffer significant
multicollinearity problems). This in itself is an interesting finding: it suggests that people living in London
draw very little distinction between, on the one hand, the fairness with which officers wield their
authority, and, on the other, the extent to which the police understand and represent citizens at the group
level.
Similarly, while ideas about community cohesion (‘people in this neighborhood can be trusted’)
and perceptions of informal social control (‘local people and authorities have control over the public space
in this area’) are also conceptually distinct, answers to these individual questions (and indeed the distinct
latent constructs underlying them) were also so highly correlated that it again made little sense to treat
them as separate ‘things’. One interpretation of this is that when people in London think about how
cohesive are their communities they do so in a way which heavily implicates assessments of the extent of
informal (and formal) social control and efficacy. We can therefore treat this combined indicator as a
measure of collective efficacy, since it reflects both respondents’ trust in those around them and – in a
closely related way – their sense that other people are willing to intervene on behalf of the public good
(Sampson et al., 1997).
By contrast, as the model comparison reported above demonstrates, respondent’s sense of moral
alignment with police and their perceived duty to obey (and indeed propensity to cooperate) were
empirically separate ‘components’ of legitimacy. This finding echoes Beetham’s concept of legitimacy in
that the different aspects of legitimacy he identified do not simply collapse into one construct but exist in
some sense alongside each other.
The focus of the measure of moral decline is on behaviors of young people and punitiveness.
While we label this measure ‘moral decline,’ this is a quite particular type of morality (indeed, one could
treat these items as indicative of a form of authoritarian ideology). It is concerned mainly with order,
authority, and what should happen to those who defy it. Unlike many of the other variables included in the
analysis, these measures do not access ‘local’ concerns; they are, implicitly at least, directed to the societal
level.
Finally, the slightly ambiguous nature of the ‘cooperation with police’ questions should be
recognized. The wording of the preamble – ‘how likely would you be to do the following things’ – is such
that some respondents might interpret it to mean ‘how likely is it that the following things might happen

10
about which you might have to do something.’ Answers could then be predicated, in part, on ideas about
the level of crime and disorder in respondent’s local areas. This does not seem to be an excessive risk,
however, as we have measured respondent perceptions of the extent of the crime problem and ‘worry’
about crime. We use these as control variables to partial out (to some extent) the need – perceived or
actual – for the police, thus homing in on the propensity to cooperate rather than the need to cooperate.

ANALYSIS STRATEGY
Multivariate analysis was conducted in two stages. First was multi-level regression modeling of the
cooperation variable. This allowed us to control for a range of possible confounders. If associations
between the key variables were not robust to a range of socio-economic and experiential controls then any
conclusions drawn about paths to cooperation with the police might be confounded by the specific
contexts from which the data were drawn. To represent the key constructs in the regression modeling we
utilized ordinal latent trait modeling in the software package Latent Gold (version 4). Separate models
were estimated for each construct and factor scores extracted for use in the regression modeling. Factor
loadings are shown in Table 1. These were high in every case, supplying further evidence that the
constructs and indicators used in this study are robust.3 Secondly, to more fully explore the theorized
associations between social concerns, trust in the police and cooperation, we utilized structural equation
modeling of the key latent constructs. Tests of the hypotheses set out above are reviewed formally in the
discussion section.

Table 3 near here

DESCRIPTIVE STATISTICS AND BIVARIATE CORRELATIONS

To aid interpretation, the factors extracted from the latent trait models were rescaled 0-10:4 Table 3
provides key properties of these latent variables. There were weak to moderate correlations between all
the potential explanatory variables and the ‘cooperation’ response variable, with one notable exception:
there was no correlation between trust in the police effectiveness and intention to cooperate.
Unsurprisingly, the highest pair-wise correlations were between the three measures or components of
police legitimacy. It also appears that a number of the variables were somewhat skewed, particularly the
measure of cooperation. This skew resulted from the fact that over a third of respondents answered ‘very
likely’ to all three manifest indicators of cooperation, thus scoring the highest possible value on the factor
representing it. Such skewness in the dependent variable represents a potential problem for the analysis
which follows, although we present below some evidence that such problems did not in fact emerge.

RESULTS
STEP ONE: REGRESSION MODELLING
For the first step of the analysis we estimated a random effects linear regression model that predicted
public propensities to cooperate with the police (using STATA 10.0, with respondents clustered in wards).
Using this design allows us to take some account of the actual conditions of respondent’s local areas and
thus home in on perceptions of disorder as subjective assessments that are tied to concerns about the
nature and trajectory of social change. Two models were estimated. Both contained relational concerns,
ideas about crime and trust in the police as the key predictors. The second differed from the first only in
that it added assessments of moral alignment with police and perceived duty to obey. Results are shown in
Table 4.

Table 4 near here

Looking first at model 1 in Table 4, there was a strongly significant positive association between
perception of moral decline and stated propensity to cooperate with the police – on average, the more a
respondent perceived the general moral order to be under threat the more ready they were to say they
would cooperate with and support the police. Perceptions of collective efficacy (which combines concerns
about levels of social cohesion informal social control) also had a small but significant positive
association with the response variable. Opinions of police community engagement and fairness had a
strongly significant positive association with the response variable. By contrast, and perplexingly, trust in

11
police effectiveness had a small but significant negative association with cooperation. It may be that,
holding trust in police fairness constant, believing the police to be more effective is associated with a
lower propensity to cooperate because the police are trusted to be efficient in dealing with crime on their
own. Further analysis (results not shown here) supported this idea to an extent – if trust in police fairness
was excluded from the model there was no significant association between trust in effectiveness and
cooperation. Finally, and somewhat contrary to expectations, assessment of local disorder had no
significant association with cooperation.
Turning to the key control variables, perhaps the most interesting feature was that satisfactory
contact had a positive statistical effect, while unsatisfactory contact had no significant impact. This runs
counter to the expectation of negative asymmetry in the effect of personal encounters with the police
(Bradford et al., 2009; Skogan, 2006). The two variables representing respondents’ ideas about crime had
different relationships with the response variable. Perceiving more crime in the local area had no
significant association with propensity to cooperate with the police. A higher perceived level of crime did
not mean that respondents were more likely to say they would contact and assist the police (which would
imply that they answered the cooperation questions on the basis of the perceived likelihood of having to
do so). Yet, net of ideas about the level of crime and the other variables in model one, higher worry about
crime was associated with greater readiness to offer cooperation. These results suggest that respondents
were indeed answering the cooperation questions in ways amenable to the research design.
Model 2 added perceptions of moral alignment and the duty to obey, the two ‘core’ measures of
legitimacy available for this study. These variables were strongly significant predictors of cooperation.
Further, once these measures of legitimacy were added, then the coefficient for ‘moral decline’ fell to less
than half its original value. This finding suggests that perceptions of moral decline in society are indeed
linked to cooperation because cooperation is an aspect of police legitimacy – those who feel the social
order to be under threat hold, on average, the police to be more legitimate, and once other measures of
legitimacy are taken into account the link between moral decline and cooperation is weakened, although,
importantly, not broken (see below). Coefficient sizes and significance levels elsewhere in the model were
little affected by the addition of the two legitimacy measures.
In order to address the potential problems created by the skewed nature of the ‘cooperation’
response variable (caused by the large number of respondents who answered ‘very likely’ to all three
questions which measure it), a binary indicator of intention to cooperate with the police was created.
Those replying ‘very likely’ to all three questions (n=2,010) were coded 1, with all other responses coded
0. A binary logistic regression predicting responses to this new variable was then estimated, using the
same explanatory variables as model 2 shown in Table 4. The key explanatory variables in the study
retained the same relationships with this new response variable, both in terms of direction of effect and
statistical significance (full results are available from the lead author). There was one exception – the
coefficient for collective efficacy was no longer significant at the 5 per cent level, although the direction
was as before and the association verged on significance. Overall, this exercise provided good evidence
that the skewness of the response variable did not create problems in the linear regression analyses.5

STEP TWO: STRUCTURAL EQUATION MODELLING

In the models described above, two key strands of variables can be identified that may influence
cooperation as a component of police legitimacy and the exercise of new parochial social control. The first
is trust in the police; and the second is perceptions of the social order which the police may represent,
reflected here by concerns about moral decline, local disorder, and collective efficacy.
Yet these elements are likely to be interrelated, with, for example, concerns about collective
efficacy associated with trust in the police and the exercise of new parochial social control. Furthermore,
if moral identification is considered to be the core component of legitimacy it is likely that the effects of
trust and social concerns on cooperation will be mediated, at least partially, by this factor. Stage two of the
analysis utilizes structural equation modeling (using MPlus 5.2, treating the indicators as categorical and
using full information maximum likelihood estimation) to specify latent constructs mirroring those used
previously. Such analysis allows a more nuanced investigation of the associations identified in stage one.

Figure 2 near here

12
 Results of the structural equation modeling are shown in Figure 2. The model supports the idea
that all the key variables in this study are associated with stated propensities to cooperate with the police –
opinions of the police itself, concerns about declining morality, and perceptions of disorder and collective
efficacy. The relationship between these variables is somewhat more complex than could be suggested by
the linear regression model.
As found by Jackson and Bradford (2009) ideas about social cohesion and collective efficacy
were correlated with subjective disorder of respondent’s local areas – people who perceived more disorder
around them also felt that collective efficacy was lower. Lower perceived collective efficacy was
associated with lower levels of trust in the police, again as found by the earlier study. The structural model
further confirmed that lower collective efficacy was associated with a reduced readiness to cooperate, an
effect largely mediated by moral identification with the police. Those who perceived lower levels of
collective efficacy not only trusted the police less, on average, than others, they were less likely to
cooperate with officers. Their social concerns damaged the legitimacy of the police in their eyes.
However, net of these associations, there was a small but significant direct link between collective
efficacy and cooperation. Those who perceived higher collective efficacy were on average slightly more
likely to say they would cooperate with the police, suggesting a somewhat greater propensity to engage
with agents of formal social control when informal controls were seen to be stronger.
The measure of moral decline retained a direct, strong, positive link with propensity to support;
and it was also associated with the other aspects of police legitimacy. Those who perceived a general
sense of moral decline were more likely to express a strong sense of moral identification with the police –
and to a lesser degree were more likely to say there is a duty to obey police directives – and these factors
were both themselves associated with propensity to cooperate. This split effect suggests that those who
have a sense that the social order is under threat are more likely, on average, to cooperate with police,
precisely because such cooperation is both an act of informal social control – which they are more likely
to think necessary – and because cooperation is an aspect of police legitimacy and they identify more
strongly with the police than others: as a bulwark against social decline, perhaps). Social concerns
expressed by a narrative of moral decline (and with a strong authoritarian overtone) appeared to bolster
the legitimacy of the police.
Perceived disorder in contrast had (again as in the linear regression model) no significant direct
link with propensity to cooperate. But the structural model showed it did have indirect negative
associations. Those who saw more disorder around them evinced less trust in the police, and this was
associated with less readiness to involve and cooperate with officers (and, in general, lower police
legitimacy). This finding chimes with that of Jackson and Bradford (2009), who found that high levels of
perceived disorder damaged ‘overall’ confidence in the police. Again, social concerns of this type
damaged trust and police legitimacy.
Trust in police procedural justice retained a significant unique association with intention to
cooperate in the structural model, and also a major indirect statistical effect via moral identification (the
significance of this indirect path was confirmed via the relevant Mplus function). This corresponds with
the predictions of Tyler et al. (Tyler and Huo 2002; Tyler and Fagan 2008) – it seems that cooperation
with the police can be enhanced by treating people fairly and with respect, and engaging with them
positively, whatever the other material and social factors in play. As in the linear models described above,
controlling for the other variables in the model higher levels of trust in police effectiveness were
associated with lower propensities to cooperate.
Finally, as would be expected the three components of police legitimacy were all associated with
each other. The model specified that respondents sense of moral alignment with the police was the most
fundamental aspect of legitimacy, with paths leading from moral alignment to both duty to obey and
cooperation (which are considered to be different aspects of expressed consent). The data supported this
assumption, with moral alignment having statistically significant and substantively large associations with
both duty to obey and cooperation. Notably, the path from duty to obey to cooperation was much weaker,
which supports the idea developed here that cooperation with the police, in the British context at least,
does not develop from legitimacy as duty to obey but is an empirically separate aspect or instantiation of
police legitimacy, rooted most importantly in assessments of the procedural fairness but also linked to the
relationship between individual and police within the policing field.

DISCUSSION

13
Our assessment of public intentions to cooperate with the police has drawn together social-psychological
and sociological perspectives to provide an empirical assessment of the associations between trust and
cooperation that locates individual ideas and opinions of the police within the wider social context of
policing in London. Psychologically, we drew on Tyler’s procedural justice model (Tyler, 1990, 2006;
Tyler and Blader, 2000; Tyler and Huo, 2002). Sociologically, we located individuals and organization
within the structured space of the policing field. Since our data were individual level, we concentrated on
public concerns about community and narratives of moral decline that may shape cooperation and which
provide part of the broader background against which people experience, and judge, the police. The
structural position of the police means that it is both held accountable for many social problems (as
representatives of nation or state it is held responsible for disorder and low social cohesion) and invoked
by the public, through calls and assistance, in order to defend and reinstate order and moral consensus
(Jackson and Sunshine, 2007).
Evidence was found in favor of hypothesis one: high confidence in police procedural justice was
positively associated with a greater propensity to cooperate with the police. Some, but not all, of this
effect was mediated by the moral alignment variable. Hypothesis two was also supported, but with an
unexpected twist: high levels of trust in police effectiveness were associated with less readiness to
cooperate. Our London data therefore correspond with the predictions of Tyler’s procedural justice model.
Hypothesis three was supported by the data. Of the three components of police legitimacy
considered here, moral alignment was the most important. There were significant paths from moral
alignment to both duty to obey and propensity to cooperate. Controlling for moral alignment, these other
components of legitimacy were themselves only rather weakly related. Moral alignment also appeared to
mediate much of the statistical effect of key explanatory variables (assessments of police procedural
justice, perceptions of moral decline and of local disorder) on cooperation. However, it did not mediate
these associations entirely. Police procedural justice and sense of moral decline retained significant and at
least moderately-sized direct associations with cooperation. This suggests that expressed consent to the
role of the police – as an aspect of its legitimacy – is bound up not only with a sense of moral alignment.
By agreeing, tacitly at least, to involve police when confronted by socially unacceptable behaviors,
Londoners may legitimize the police for reasons beyond assessments of the extent to which institutional
and personal moralities are felt to be aligned.
Of the two hypotheses (four and five) concerning the relationship between perceptions of
collective efficacy and cooperation, it was hypothesis five that found empirical support. High perceived
collective efficacy was associated with greater trust in the police and, through this, greater legitimacy and
thus higher propensities to cooperate. When people think that others around them are more likely to
engage in informal social control, their sense of trust in the police – as a proto-typical group representative
– may be enhanced and they may be more likely to offer it their support. Furthermore, net of these
associations, those who saw higher levels of collective efficacy were still somewhat more likely to say
they would cooperate. It seems that the police in London garner legitimacy when communities are felt to
be cohesive. Conversely, of course, those who felt their local area lacked collective efficacy trusted the
police less and invested less legitimacy in it.
Hypothesis six was supported by the data. While there was no direct path between perception of
disorder and cooperation in the structural model, the strong association between higher disorder and lower
trust in the police implied that cooperation with the police suffers when disorder is seen to be high: trust
and legitimacy may be undermined under such conditions. Finally, hypothesis seven also found support.
Belief in the decline of morality and authority in society was associated with a stronger propensity to call
upon the police as a resource of social order and control.
This study therefore supports the idea that there are at least two routes toward cooperation with
the police. On the one hand, those who felt that the police are fair and engaged with the community (thus
demonstrating that they are group representatives who both communicate high group status to citizens and
represent and defend community norms and values) were more likely to offer their cooperation. We find
support for Tyler’s hypothesis that trust in police fairness is more important than trust in police
effectiveness in shaping legitimacy and, therefore, cooperation. On the other hand, trust in the police and
police legitimacy were also influenced by ideas and feelings running through a second path relating to the
position of the police as an organization within the field of policing. As specified in our model,
individual’s sense of ‘moral decline’ (as measured, recall, by indicators closely aligned assessments of
authoritarian attitudes) is an important aspect of their habitus that affects how they think about, and

14
potentially act in relation to, the police. Holding trust in police fairness and effectiveness constant, those
who perceived moral decline were more likely to hold the police legitimate (perhaps because of a need to
support figures of authority) and more likely to be willing to engage in informal social control. By
contrast, higher levels of perceived disorder and lower collective efficacy were associated with less
favorable opinions of the police and, through these, lower propensities to cooperate. Contrariwise,
respondents who felt more secure about the condition of their local area were more willing to engage in
social control activities and cooperate with the police – in large part because they had more trust in the
police and therefore granted the police more legitimacy.
These divergent patterns underline the complex and contradictory nature of the relationship
between police and public. If the police represent to many individuals nation, community and the
established order, this appears to imply a bi-directional relationship between assessments of social order,
threat and cohesion and judgments about the police in terms of trust and legitimacy. A view that
established order and morality was under threat at a societal level was linked to higher legitimacy, both in
terms of ‘moral alignment’ and in terms of duty to obey and readiness to cooperate: that is, in terms of a
higher level of expressed consent. But a view that the local area was in social and physical decline was
associated with a more critical stance, lower legitimacy and a lesser propensity to express consent to the
role of the police.
The strongest version Bourdieu’s notion of the field places great emphasis on its structuring
properties. Because it determines the social positions and options for actors embedded in it, the field
predisposes and may even determine their behavior. It may not only be, then, that some individuals
perceive a situation of ‘moral decline’ (for example) that causes them to legitimize the police in an active
sense (believing that one has a duty to obey officer’s directives, for example). Rather, perceptions of
social and moral threat may be filtered through the dispositions engendered by the habitus carried by such
actors and by their location in the field of policing, relative to the police organization, such that stances
and roles are activated that promote legitimizing actions. The direct path from ‘moral decline’ to
cooperation in our model may be an illustration of just such a process. The logic of the relations between
individual and organizational positions within the policing field may provide a mechanism through which
social concerns are translated directly into legitimating actions. In particular, the dominating place of the
police within the field more or less forces those who with rather authoritarian viewpoints about moral
decline into some kind of engagement with it. Because the police is an institutionalized organization,
strongly isomorphic with the field of policing, such legitimating processes may operate in part outside
assessments of the performance of the organization itself (whether in terms of fairness, effectiveness or
some other criteria).
In sum, the evidence presented above suggests that the structural relationship between police and
public means individuals may turn to the police as a symbolic or material resource to use against crime
and disorder. Symbolically, those who perceive a situation of moral decline may see the police as a
bulwark against such threat. Materially, cooperation with the police is an act of ‘new parochial’ social
control that people who feel they live in cohesive communities, and who have relatively high levels of
trust in the police, feel themselves able to draw on. By indicating their readiness to cooperate with
officers, their (potential) actions speak to directly to Beetham’s (1990: 12) expanded notion of legitimacy.
Acts of cooperation demonstrate a form of consent to the role of the police and to the relationship between
police and public that has both a symbolic and normative strength. But such cooperation is in no sense
inevitable, and people may equally engage in the struggles that Bourdieu would argue are rendered
inevitable by the unequal distribution of power within the policing field. Actors may withdraw
cooperation when they feel officers are not doing the right thing or when they feel that the police, or the
social structures it represents, has let them down (when, for example, the state is seen to be abandoning
communities to disorder). By withdrawing cooperation, they are making quite specific claims about the
propriety of the police role and the extent of the power vested in it: about, in other words, its legitimacy.

LIMITS OF THE PRESENT STUDY

The limits of the present study must be recognized. Most importantly, we are not able with the available
data to predict actual acts of cooperation, but rather stated propensities to cooperate. Saying one will
cooperate may be easy, and more socially acceptable, than the alternative; actually cooperating may be
more difficult. But while it is clearly optimal to capture both in a given study, it is still valuable to
examine stated propensities. Arguably, survey responses indicating a readiness to cooperate with the

15
police not only capture people’s intentions, but they may in themselves express some justification for
police legitimacy, since they explicitly recognize a concrete police role in certain situations (rather than
simply recording respondent’s views about how well the police are ‘doing’ in general terms). Saying one
would not contact or assist the police in the face of criminal or ‘suspicious’ activity communicates a quite
definite negative stance. A second limitation of this study is that it is cross-sectional in design. We cannot
know whether the direction of the paths traced here are as formulated or whether, for example, there are
feedback loops between the legitimacy granted to the police and a sense of trust in them. Future studies
using panel data would be a welcome addition to work on this topic. For now, capturing and explaining
variation across a major world city is, we hope, a significant contribution, especially as variation in issues
of perceived social order, fear of crime, trust in the police and the legitimacy of legal authorities is
unlikely to be amenable to intervention and manipulation.

CONCLUSION
This study has found that the legitimacy of the police – as measured by respondent’s sense of moral
alignment with police and their propensity to express consent to the police role – was higher among
individuals with high confidence that police were fair, understand and represent community norms and
values; among individuals who saw threats to moral values in society (people saw a need for the police in
the face of a loss of moral authority); and among individuals who perceived a strong collective will in
their neighborhood to regulate behavior in public space and defend civil norms and values.
The idea that the police are ‘proto-typical group representatives’, associated with the perceived
failures and successes of the existing social order (Sunshine and Tyler, 2003b; Jackson and Sunshine,
2007), is supported by our data. If the police can act in ways which indicate shared group membership to
people through displays of procedural fairness and associated behaviors, this may enhance legitimacy,
cooperation and support. But the observed relationship between police and public is also located within a
broader social and cultural framework. Diverse social-psychological and sociological structures and
phenomena – the desire or need to maintain social order, cohesiveness and moral order and the existence
of the police as formal monopolists of legitimate force – may act on and through the behavior of
individual actors within the field of social activity oriented to maintaining social order in such a way that
the legitimacy of the police tends to be reproduced. By engaging in acts of social control, individuals re-
affirm the position of the police within the social structure in which both people and institution are
embedded, and they appear to do this without necessarily referring to assessments of the police as an
organization (the extent to which they trust it to be fair, for example).
Acts that express consent to the role of the police may be most importantly influenced by trust in
its fairness and legitimacy understood as a sense of moral alignment. But they can also flow from a need
to exert social control that is prior to specific assessments of the way the police behave. We might also
point to the symbolic power wielded by the police to construct and reproduce the public’s sense of what
order and security are, should be, and how they can be maintained (Loader 1997). The police is not a
passive recipient of support and cooperation generated by external social and political pressures – it is an
active agent in the creation and reproduction of such pressures. Future research might profitably examine
the relationship between police representations of crime and disorder and the support it garners from the
public.
A central finding of our study has been that some key factors influencing not only public
assessments of police activities, but also the right of the police to be recognized and involved as the
institution entitled and enabled to deal with crime, are far removed from arenas traditionally under the
purview of the police. Police legitimacy may be bound up in much wider structures of feeling and thought
(Loader and Mulcahy, 2003) and, ultimately, with extremely deep-seated social and psychological needs.
Yet, procedural justice concerns also open up a space for individual agency and exigency. For all that
calling the police may be an act constitutive of its legitimacy, what subsequently transpires reflects back
onto that legitimacy. If officers are found to act in an unfair or unjust manner, if they fail to communicate
shared group membership to those with whom they have contact, then the legitimacy of the police – as
represented precisely by propensities to support or cooperate in the future – may well suffer. Extant social
structures and processes may predispose people toward seeing the police as legitimate, but neither process
determines the social position of the police. How officers behave – the manner in which they wield their
authority – is once again found to be central to the legitimacy of the police.

16
 NOTES
1. Legitimacy can be seen as primarily rooted in moral identification ‘in the eye of the beholder’,
residing in the beliefs of those subject to an authority. On the other hand it can also be viewed in
terms of a normative justifiability, provided by organisational adherence to established legal and
ethical frameworks and assessed by external or objective criteria. We do not seek to address this
distinction here – suffice to say that in the context of British policing, elements of both
conceptualisations may be important among the public at large.
2. Like Sunshine & Tyler (2003) we treat cooperation as an outcome of obligation to obey. In this
respect, there are no empirical implications to treating cooperation as legitimizing acts. Nevertheless
the conceptual point remains, we believe, important.
3. Public assessments of the extent to which police follow the rules governing their behaviour, and the
law more generally, are also likely to be important aspects of police legitimacy (Beetham 1991;
Bottoms & Tankebe 2008). However, we do not have measures of this aspect of police legitimacy
available for this study.
4. While it is generally accepted that formal statistical tests of ordinal latent trait models are too
sensitive to sample size to be of practical use, there is as yet little agreement as to the relevance of
approximate fit statistics of the type often used in confirmatory factor analysis.
5. We used the ‘rescale’ programme written for Stata by Johan Martinsson at the Department of
Political Science of University of Gothenburg.
6. Similarly, plots of residuals against fitted values generated from the models shown in Table 3
suggested no problem with heteroscedasticity.

17
 REFERENCES
Barker, Rodney. 1990. Political Legitimacy and the State. Oxford: Clarendon Press.
Bayley, David H. and Clifford D. Shearing. 1996. The Future of Policing. Law and Society Review 30(3):
585-606.
Beetham, David. 1991. The Legitimation of Power. London: Macmillan.
Bittner, Egon. 1990. Aspects of Police Work. Boston: Northeastern University Press.
Black, Donald. 1993. The Social Structure of Right and Wrong. San Diego: Academic Press.
Bottoms, Anthony E. and Justice Tankebe. 2008. Re-Conceptualising Police Legitimacy. Paper Presented
at the annual Stockholm Criminology Symposium, Stockholm, Sweden.
Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press.
Bourdieu, Pierre. 1990. The Logic of Practice. Cambridge: Polity Press.
Bourdieu, Pierre and Lo c Wacquant. 1992. An Invitation to Reflexive Sociology. Chicago: Polity Press.
Bradford, Ben. 2009. The Quality of Police Contact: Procedural Justice Concerns Among Victims of
Crime in London. Available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1596754
Bradford, Ben, Jonathan Jackson and Elizabeth A. Stanko. 2009. Contact and Confidence: Revisiting the
Impact of Public Encounters with the Police. Policing and Society 19(1): 20-46.
Brunson, Rod K. 2007. ‘Police Don’t Like Black People’: African-American Young Men’s Accumulated
Police Experiences. Criminology and Public Policy 6(1): 71-102.
Bursik, Robert J. 1988
Bursik, Robert J. and Harold G. Gramsick. 1993. Neighborhoods and Crime: The Dimensions of Effective
Community Control. New York: Lexington Books.
Carr, Patrick J. 2003. 'The New Parochialism: The Implications of the Beltway Case for Arguments
Concerning Informal Social Control. The American Journal of Sociology 108(6): 1249-1291.
Carr, Patrick J., Laura Napolitano and Jessica Keating. 2007. ‘We Never Call the Cops and Here is Why’:
A Qualitative Examination of Legal Cynicism in Three Philadelphia Neighborhoods. Criminology
45(2): 445-480.
Crank, John P. 2003. Institutional Theory of the Police: A Review of the State of the Art. Policing: An
International Journal of Police Strategies and Management 26(2): 186-207.
Emirbayer, Mustafa and Victoria Johnson. 2008. Bourdieu and Institutional Analysis. Theoretical
Sociology 37: 1-44.
Farrall, Stephen, Jonathan Jackson and Emily Gray, 2009. Social Order and the Fear of Crime in
Contemporary Times. Oxford: Oxford University Press.
Ferraro, Kenneth 1995. Fear of Crime: Interpreting Victimization Risk. New York: SUNY Press
Girling, Evi, Ian Loader and Richard Sparks. 1998. Narratives of Decline: Youth, (Dis)Order and
Community. British Journal of Criminology 38(3): 388-403.
Girling, Evi, Ian Loader and Richard Sparks. 2000. Crime and Social Control in Middle England:
Questions of Order in an English Town. London: Routledge.
Goudriaan, Heike, Karin Wittebrood and Paul Nieuwbeerta. 2006. Neighbourhood Characteristics and
Reporting Crime. British Journal of Criminology 46: 719-742.
Gray, Emily, Jonathan Jackson and Stephen Farrall. 2011. Feelings and Functions in the Fear of Crime:
Applying a New Approach to Victimisation Insecurity. British Journal of Criminology 51(1): 79-94.
Hinds, Lyn and Kristina Murphy. 2007. Public Satisfaction with the Police: Using Procedural Justice to
Improve Police Legitimacy. The Australian and New Zealand Journal of Criminology 40(1): 27-42.
Hough, Mike. 2003. Modernisation and Public Opinion: Some Criminal Justice Paradoxes. Contemporary
Politics 9(2): 143-155
Hough, Mike. 2004. Modernisation, Scientific Rationalisation and the Crime Reduction Programme.
Criminal Justice, 4(3): 239-254.
Hough, Mike. 2007. Policing, New Public Management and Legitimacy in Britain. In Legitimacy,
Criminal Justice, and the Law, eds. Jeffrey Fagan and Tom R. Tyler. New York: Russell Sage
Foundation Press.
Hunter, A. 1995. Private, Parochial and Public Social Orders: The Problem of Crime and Incivility in
Urban Communities. In Metropolis: Center and Symbol of Our Times, ed. Philip Kasinitz. New York:
New York University Press.
Jackson, Jonathan and Ben Bradford. 2009. Crime, Policing and Social Order: On the Expressive Nature of
Public Confidence in Policing. British Journal of Sociology 60(3): 493-521.

18
Jackson, Jonathan and Jason Sunshine. 2007. Public Confidence in Policing: A Neo-Durkheimian
Perspective. British Journal of Criminology 47(2): 214-233.
Johnson, Cathryn, Timothy J. Dowd and Cecilia L. Ridgeway. 2006. Legitimacy as a Social Process.
Annual Review of Sociology 32: 53-78.
Kelman, Herbert C. and Hamilton, V. Lee. 1990. Crimes of Obedience: Towards a Social Psychology of
Authority and Obedience. Yale: Yale University Press.
Loader, Ian. 1996. Youth, Policing and Democracy. Basingstoke: Palgrave McMillan.
Loader, Ian. 1997. Policing and the Social: Questions of Symbolic Power. British Journal of Sociology,
48(1): 1-18.
Loader, Ian. 2006. Policing, Recognition and Belonging. The Annals of the American Academy 605: 201-
221.
Loader, Ian and Aogán Mulcahy. 2003. Policing and the Condition of England: Memory, Politics and
Culture. Oxford: Oxford University Press.
MacDonald, John and Robert J. Stokes. 2006. Race, Social Capital, and Trust in the Police. Urban Affairs
41(3): 358-375.
McAra, Lesley and Susan McVie. 2005. The Usual Suspects: Street-life, Young People and the Police.
Criminal Justice 5(1): 3-36.
Meyer, John W. and Brian Rowan. 1977. Institutionalized Organizations: Formal Structure as Myth and
Ceremony. The American Journal of Sociology 83(2): 340-363.
Office for National Statistics. 2009. Resident Population Estimates by Ethnic Group, (Percentages), 2007.
London: Office for National Statistics.
Reiner, Robert. 2000. The Politics of the Police (3rd edition). Oxford: Oxford University Press.
Sampson, Robert J. and Dawn Jeglum Bartusch. 1998. Legal Cynicism and (Subcultural?) Tolerance of
Deviance: The Neighborhood Context of Racial Differences. Law and Society Review 32(4): 777-
804.
Sampson, Robert J., Stephen W. Raudenbush and Felton Earls. 1997. Neighborhoods and Violent Crime:
A Multilevel Study of Collective Efficacy. Science 227: 918-924.
Sampson, Robert J., and Stephen W. Raudenbush. 1999.
Sasson, Theodore. 1995. Crime Talk: How Citizens Construct a Social Problem. New York: Aldine de
Gruyter.
Scott, W. Richard 2001. Institutions and Organizations (2nd Edition). Thousand Oaks: Sage Publications,
Inc.
Skogan, Wesley G. 2006. Asymmetry in the impact of encounters with the police. Policing and Society
16(2): 99-126.
Skogan, Wesley G. 2009. Concern About Crime and Confidence in the Police: Reassurance or
Accountability. Police Quarterly 12(3): 301-318.
Silver, Eric and Lisa L. Miller. 2004. Sources of Informal Social Control in Chicago Neighbourhoods.
Criminology 42(3): 551-583.
Smith, David. 2007. The Foundations of Legitimacy. In Legitimacy and Criminal justice: International
perspectives, ed. Tom R. Tyler. New York, Russell Sage Foundation Press.
Sunshine, Jason and Tom R. Tyler. 2003a. Moral Solidarity, Identification with the Community, and the
Importance of Procedural Justice: The Police as Prototypical Representatives of a Groups Moral
Values. Social Psychology Quarterly 66(2): 153-165.
Sunshine, Jason and Tom R. Tyler. 2003b. The Role of Procedural Justice and Legitimacy in Public
Support for Policing. Law and Society Review 37(3): 513-548.
Tankebe, Justice. 2008. Police Effectiveness and Police Trustworthiness in Ghana: An Empirical
Appraisal. Criminology and Criminal Justice 8(2): 185 – 202.
Tyler, Tom R. 1990. Why People Obey the Law. New Haven: Yale University Press.
Tyler, Tom R. 2004. Enhancing Police Legitimacy. The Annals of the American Academy 593: 84-99.
Tyler, Tom R. 2006. Psychological Perspectives on Legitimacy and Legitimation. Annual Review of
Psychology 57: 375-400.
Tyler, Tom R. 2009. Legitimacy and Criminal Justice: The Benefits of Self-Regulation, Ohio St. J. Crim.
L. 307.
Tyler, Tom R. and Steven L. Blader. 2000. Cooperation in Groups. Procedural Justice, Social Identity and
Behavioral Engagement. Philadelphia: Psychology Press.

19
Tyler, Tom R. and Robert J. Boeckmann. 1997. Three Strikes and You Are Out, But Why? The
Psychology of Public Support for Punishing Rule Breakers. Law and Society Review 31: 237-265.
Tyler, Tom R. and Jeffrey Fagan. 2008. Legitimacy and cooperation: why do people help the police
fight crime in their communities? Ohio State Journal of Criminal Law 6: 231-275.
Tyler, Tom R. and Yuen J. Huo. 2002. Trust in the Law: Encouraging Public Cooperation with the Police
and Courts. New York: Russell Sage Foundation.
Tyler, Tom R., Stephen Schulhofer and Aziz Z. Huq. 2010. Legitimacy and Deterrence as Strategies for
Motivating Muslim American Cooperation with Police Counter-terrorism Efforts. Law and Society
Review 44(2): 365-402.
Waddington, P.A.J. 1999. Policing Citizens: Authority and Rights. London: University College Press.
Warner, Barbara D. 2003. The Role of Attenuated Culture in Social Disorganization Theory. Criminology
41(1): 73-97.
Warner, Barbara D. 2007. Directly Intervene or Call the Authorities? A Study of Forms of Neighborhood
Social Control Within a Social Disorganization Framework. Criminology 45(1): 99-129.
Weber, Max. 1978. Economy and Society: Volume Two. Berkley, University of California Press.
Wells, William, Joseph A. Schafer, Sean P. Varano and Timothy S. Bynum. 2006. Neighborhood
Residents Production of Order: The Effects of Collective Efficacy on Responses to Neighborhood
Problems. Crime and Delinquency 52(4): 523-550.
Williams, Raymond. 1964. The Long Revolution. Harmondsworth: Penguin.

Figure 1. Conceptual map

Concerns about Trust in police

collective fairness and
efficacy community
representativeness

Concerns about Legitimacy: Legitimacy:

moral decline Moral alignment expressions of
with police consent

Trust in police
Concerns about effectiveness
neighbourhood
disorder

20
Figure 2. Social concerns, confidence in policing, and cooperation with the police

21
Table 1. Sample structure
Percentages

Gender Ethnic group

Male 45 White British/Irish 53
Female 55 Black African 7
Age Mixed 7
15-21 6 Black Caribbean 6
22-24 5 Indian 5
25-34 21 Bangladeshi 3
35-44 20 Pakistani 3
45-54 16 Other 15
55-64 12 Recent contact with the police
65-74 12 No 81
75 plus 8 Yes and satisfactory 15
Housing tenure Yes and unsatisfactory 8
Home owner 47 Employment Status
Social renter 28 Full time 41
Private renter 21 Part-time 9
Other 4 Unemployed 4
Car access Retired 23
Yes 60 Student 7
No 40 Other economically inactive 13
Recent victim of crime Other 2
Yes 8
No 92 Total n (numbers) 5,120
Source: London Metropolitan Police Public Attitudes Survey 2009/10

22
Table 2. Key measures - constructs and indicators
Factor loadings from
simultaneous Factor loadings from
confirmatory factor ordinal latent trait
analysis models
Cooperation with the police: How likely would you be to do the following things?
Call the police to report a crime occurring in your neighbourhood? 0.96 0.84
Help the police to find someone suspected of committing a crime by providing them with information? 0.90 0.82
Report dangerous or suspicious activities in your neighbourhood to the police? 0.95 0.83
Moral alignment with the police: To what extent do you agree or disagree that when you deal with the police in London:
The police usually act in ways that are consistent with my own ideas about what is right and wrong 0.94 0.84
The police can be trusted to make decisions that are right for people in my neighbourhood 0.95 0.82
My own feelings about what is right and wrong usually agree with the law 0.86 0.78
Obligation to obey: Please tell me to what extent you agree or disagree with each of these statements
You should do what the police tell you even if you disagree 0.92 0.81
You should accept the decisions made by police, even if you think they are wrong 0.94 0.87
You should do what the police tell you to even when you don’t like the way they treat you 0.86 0.79
Perceived levels of collective efficacy
People in this neighbourhood can be trusted 0.81 0.67
People act with courtesy to each other in public space in this area 0.77 0.66
You can see from the public space here that people take pride in their environment 0.80 0.70
Local people and authorities have control over the public space in this area 0.75 0.66
If I sensed trouble whilst in this area, I could get help from people who live here 0.76 0.66
The people who live here can be relied upon to call the police if someone is acting suspiciously 0.78 0.65
Perception of crime problem: Are these things a problem in this area?
Burglary 0.86 0.74
Non violent theft, for instance, pick pocketing/bag snatch 0.89 0.74
Car crime – stealing cars or from cars 0.88 0.77
Knife crime – people carrying or using knives to threaten or commit violence 0.89 0.69
Trust in police effectiveness: How well do the police actually carry out these services?
Tackle gun crime 0.70 0.70
Support victims and witnesses 0.76 0.71
Tackle dangerous driving 0.78 0.74
Tackle drug dealing and drug use 0.79 0.76
Enforcing road legislation to improve traffic flows 0.73 0.69
Responds to emergencies promptly 0.78 0.66
Provide a visible patrolling presence 0.76 0.67
Trust in police fairness and community engagement
They would treat you with respect if you had contact with them for any reason 0.76 0.63
The police in this area treat everyone fairly regardless of who they are 0.77 0.65
The police in this area are friendly and approachable 0.85 0.71
The police in this area are helpful 0.89 0.77
They are dealing with the things that matter to people in this community 0.85 0.72
They understand the issues that affect this community 0.84 0.70
They can be relied on to be there when you need them 0.85 0.74
The police in this area listen to the concerns of local people 0.80 0.66
Concerns about neigherborhood disorder: Are these things a problem in your area?
Noisy and/or nuisance neighbours 0.60 0.64
Teenagers hanging around in the street 0.81 0.64
Drinking in the street 0.80 0.76
Worry about crime: How worried are you about:
Having your home broken into and something stolen 0.81 0.76
Being mugged 0.87 0.85
Being physically attacked by strangers 0.93 0.89
Being insulted or pestered by anybody while in the street or any other public place 0.90 0.84
Concerns about moral decline in society
Young people today don’t have enough respect for traditional values 0.71 0.75
People who break the law should be given stiffer sentences 0.91 0.72
Fit statistics
Chi square 8771.8
p <0.005
CLI 0.91
TFI 0.97
RMSEA 0.08
Source: London Metropolitan Police Public Attitudes Survey 2009/10

23
Table 3. Correlation matrix and summaries for key explanatory variables

Rescaled continuous factors extracted from ordinal latent trait models

1 2 3 4 5 6 7 8 9 10
Cooperation with the police (1) 1
Concerns about moral decline in society (2) 0.21* 1
Legitimacy - moral alignment with police (3) 0.33* 0.11* 1
Legitimacy - perceived duty to obey (4) 0.25* 0.09* 0.55* 1
Trust in police fairness and community engagement (5) 0.20* 0.01 0.41* 0.28* 1
Trust in police effectiveness (6) -0.02 -0.14* 0.20* 0.10* 0.38* 1
Perception of crime problem (7) -0.04* -0.09* 0.09* 0.03* 0.10* 0.25* 1
Worry about crime (8) 0.14* 0.22* -0.08* -0.07* -0.16* -0.27* -0.36* 1
Concerns about neighborhood disorder (9) -0.06* 0.05* -0.21* -0.09* -0.22* -0.28* -0.35* 0.33* 1
Perceived levels of collective efficacy (10) 0.13* -0.01 0.26* 0.15* 0.40* 0.21* 0.27* -0.17* -0.29* 1

Summary of variables
Mean SD Skewness Kurtosis Minimum Maximum
Cooperation with the police 7.56 2.37 -0.59 2.83 0 10
Concerns about moral decline in society 7.15 1.98 -0.10 2.27 0 10
Legitimacy - moral alignment with police 6.39 2.14 0.14 2.92 0 10
Legitimacy – perceived duty to obey 6.04 2.34 -0.10 2.90 0 10
Trust in police fairness and community engagement 5.63 1.63 0.06 3.46 0 10
Trust in police effectiveness 5.77 1.62 0.18 3.26 0 10
Perception of crime problem 7.03 2.91 -0.61 2.39 0 10
Worry about crime 4.28 2.30 0.24 3.26 0 10
Concerns about neighborhood disorder 3.74 2.23 -0.16 2.46 0 10
Perceived levels of collective efficacy 5.66 1.52 0.15 3.60 0 10
* = p<0.05
All factors coded such that high = more (confidence, worry, disorder, greater moral decline etc.).
Source: London Metropolitan Public Attitudes Survey 2009/10

24
 1
Table 4. Linear regression model predicting stated propensities to assist the police
(High scores = greater propensity).

95% CI for B 95% CI for B

B Lower Upper B Lower Upper
Constant 4.40 *** 3.85 4.96 3.55 *** 2.97 4.14
Gender (ref: male)
Female 0.02 -0.09 0.14 0.03 -0.08 0.14
Age (ref: 21 and under)
22-24 0.16 -0.17 0.50 0.15 -0.18 0.49
25-34 0.40 ** 0.15 0.66 0.34 ** 0.09 0.60
34-44 0.50 *** 0.25 0.76 0.41 ** 0.16 0.67
45-54 0.57 *** 0.31 0.83 0.47 *** 0.21 0.73
55-64 0.53 *** 0.25 0.81 0.40 ** 0.12 0.68
65-74 0.77 *** 0.49 1.05 0.60 *** 0.32 0.88
75 and over 0.74 *** 0.43 1.05 0.53 ** 0.22 0.84
Ethnic group (ref: White British/Irish)
Mixed -0.24 -0.48 0.01 -0.19 -0.43 0.05
Indian -0.23 -0.50 0.03 -0.23 -0.50 0.03
Pakistani -0.11 -0.46 0.23 -0.12 -0.46 0.22
Bangladeshi 0.06 -0.27 0.39 -0.02 -0.35 0.31
Black Carribean -0.25 * -0.50 0.00 -0.25 * -0.50 0.00
Black African -0.14 -0.37 0.10 -0.19 -0.42 0.05
Other -0.14 -0.31 0.03 -0.15 -0.32 0.02
Social class (ref: AB)
C -0.17 -0.35 0.01 -0.17 -0.35 0.00
DE -0.23 * -0.44 -0.02 -0.25 * -0.46 -0.04
Car access (ref: no)
Yes 0.19 ** 0.05 0.32 0.19 ** 0.06 0.33
Tenure (ref: Home owner/other)
Social renter -0.31 *** -0.47 -0.15 -0.32 *** -0.48 -0.16
Private renter -0.30 ** -0.48 -0.12 -0.29 ** -0.46 -0.11
Other -0.37 * -0.70 -0.04 -0.34 * -0.67 -0.02
Victim of crime (ref: no)
Yes -0.19 -0.43 0.05 -0.19 -0.42 0.05
Contact with the police (ref: none)
Satisfactory contact 0.31 ** 0.12 0.49 0.31 ** 0.13 0.50
Unsatisfactory contact2 -0.02 -0.25 0.20 -0.04 -0.27 0.18
Trust in police fairness and community engagement 0.16 *** 0.11 0.20 0.15 *** 0.10 0.19
Trust in police effectiveness3 -0.06 ** -0.10 -0.02 -0.05 * -0.10 -0.01
Perception of crime problems3 -0.02 -0.05 0.00 -0.02 -0.05 0.00
Worry about crime3 0.12 *** 0.09 0.15 0.11 *** 0.07 0.14
Concerns about neighborhood disorder3 -0.02 -0.05 0.01 -0.03 -0.06 0.00
Perceived levels of collective efficacy3 0.05 * 0.01 0.10 0.05 * 0.01 0.10
Concerns about moral decline in society3 0.28 *** 0.25 0.31 0.11 *** 0.08 0.14
Moral identification with police 0.22 *** 0.19 0.26
Perceived duty to obey police 0.10 *** 0.07 0.13
ICC 0.13 0.13
Unweighted n 5120 5120
* = p<0.05; ** = p<0.01; *** = p<0.001
1 Random effects models, respondents grouped within wards.
2 Includes 'neither' and 'don't know'.
3 Coded such that high = more (intention to cooperate, confidence, worry, disorder, greater moral decline etc.).
Source: London Metropolitan Police Public Attitudes Survey 2009/10

25

View publication stats