The Impact of Homicide Upon Public Confidence and Reassurance

THE IMPACT OF HOMICIDE UPON PUBLIC CONFIDENCE AND REASSURANCE

A Report to ACPO (Homicide Working Group) and the National Policing Improvement Agency

Martin Innes, Trudy Lowe, Nicola Weston and Colin Roberts

Universities’ Police Science Institute, Cardiff University

February 2010

EXECUTIVE SUMMARY

This research study investigates the impacts of homicides upon public confidence and reassurance. In particular it focuses upon how and why some crimes involving illegal death have a profound and widely distributed impact on how people think, feel and act in relation to their security, whereas other, ostensibly similar incidents, are far less consequential and cause less social harm. For the purposes of this study two key definitions are employed:

Public confidence is conceived as a measure of perceptions, experiences and expectations of the police.
Reassurance measures broader levels of neighbourhood security.

The aim of this study is to provide an evidence base that can inform the development and improvement of current procedures for conducting community impact assessments.

Based upon in-depth studies of community reactions following seven criminal homicides that varied in terms of their circumstances and situational contexts, the analysis identifies five ‘impact models’. These seek to capture the patterns present in relation to the breadth and depth of harm that such crimes can have for community confidence and reassurance:

Private impact – is where the harms experienced are restricted to the family and friends of the victim(s).
Parochial localised impact – are crimes where, in addition to the immediate social contacts of the victim(s), there are negative repercussions for many people living in the local neighbourhoods. Importantly though, these negative consequences tend to be fairly geographically confined and do not ‘leak’ out into the wider public.
Parochial distributed impacts – are similar to the above. They also involve harm to people beyond the immediate family and friends. But instead the concern ‘travels’ across and through social networks. So for example, if the victim is a member of a minority community, then the crime may not have much of a discernible impact upon their immediate neighbourhood, but it could do for others self-identifying with that minority
Public impact – some homicides elicit a more generalised sense of revulsion and concern. These can be said to have a ‘public impact’. That is, they shape how the general public think, feel and behave in relation to their security in some way.
Public with parochial counter-reaction – in some circumstances though, the research finds that where a public reaction occurs this can in effect cause a neighbourhood level counter-reaction. The tenor of which is that the media and public accounts are actually artificially inflating the risks associated with the area where the crime occurred. In effect, the parochial counter-reaction serves to produce a more balanced local response to the crime and helps people to adjust back into their normal routines.

Albeit somewhat tentative, owing to some of the limitations associated with the empirical evidence available to inform this analysis it is suggested that in some situations it is important for police to recognise that community impact is a process, where the actual intensity, distribution and scale of public concern about an incident alters and evolves over time. One crucial influence upon any such shifts is the police investigation.

The empirical data informing the study suggest that the pattern of impact that an individual homicide incident has is shaped by the inter-play of three principal variables:

The crime – the nature of the violence used; the socio-demographics and lifestyle of the victim; where it happened; the relationship (or not) between the victim and offender, are all features that help to determine how many people will attend to the crime;
The context – aspects of the setting and situation in which the incident arises plays a key feature in shaping the depth and breadth of reactions. Areas where people have an established sense of vulnerability to crime may well react more strongly to homicides than those with strong levels of collective efficacy;
Police response – the data suggests that police actions can both amplify and mitigate levels of public concern following a major crime.

Understanding the relative and interactive significance of these factors is vital in being able to diagnose how and why some incidents become the focus for widespread public concern and approbation. For example, if in responding to a fairly routine kind of homicide the police make some error, then this can cause increased public concern. Likewise, police cannot control all aspects of the social context in which crime occurs. However, by thinking carefully about the constitution of local settings and circumstances, police should be able to predict the likelihood of different kinds of crime triggering pronounced social reactions.

In terms of developing this more predictive approach, the research suggests that in order to gauge the community impact of incidents, police need to measure along three key dimensions:

Physical distance – it is vital to look at whether the impact of the crime is fading the further away from the epicentre one is. In circumstances where the impact does not decline, this is suggestive of a crime possessed of the potential for generating a widely distributed negative impact;
Temporal distance – currently community impact assessments do not necessarily take sufficient account of the time dimension. Ordinarily one would expect that as time passes , most communities will progressively readjust and forget about a crime. However, the empirical data analysed herein suggests scenarios where this may not happen. Accordingly targeted reassurance interventions may be required to diffuse persistent fears and concerns;
Social distance – this concept finds that a crime will resonate more strongly with people if they identify in some way with it. This sense of identification could be on the basis of geography, lifestyle routine, faith, ethnicity and so forth. In practical terms it is therefore important for police to think about for whom a particular incident may have resonance and to implement their community impact assessment procedures accordingly.

In summary, the research finds that some homicides do acquire the properties and attributes of a ‘signal crime’, altering how people think, feel or act. However, it is equally the case that ostensibly similar crimes involving fatal violence elicit far less in terms of community reactions. This is because of how the relative power of an incident to impact and cause harm to people beyond the victim’s closest social contacts, depends in part upon the social context and police response. The aim of this study has been to establish a framework for helping police to recognise and categorise the different types of impact patterns that individual crimes can have and the ‘shape’ of these. In setting out the five impact models, the research has identified that particular crimes act in very different ways upon levels of public confidence and reassurance. It has been found that some incidents have a deep and profound negative impact, but on a relatively few number of people. In contrast, other incidents perhaps have a more diffuse impact, but with the effects ‘travelling’ across a far larger number of people.

In seeking to understand the harms induced by different homicides, it has been found that aspects of the police response can effectively amplify rather than mitigate the overall level of social harm that results. Equipped with this understanding the police in thinking about their response to particular crimes can start to categorise incidents according to their potential impacts and develop appropriate community impact management strategies. For doing something about a crime with a ‘deep but narrow’ social impact will require a different set of interventions from one that has ‘broad and shallow’ consequences.

INTRODUCTION

This report outlines key findings from a series of field-work based research studies designed to develop an evidence-based perspective on the extent to which incidents of criminal homicide impact upon public confidence and reassurance. Specifically, the report has three principal aims:

To establish whether homicides do impact upon levels of public confidence and reassurance, and their relative salience compared to other crime and disorder problems?
To explain how and why some incidents exhibit profound and long-lasting impacts, but others do not;
To consider the policy and practice implications for the police response to homicide.

In engaging with these aims, the empirical data and analysis detailed herein builds upon a wider research programme examining processes of social reaction to crime, disorder and policing more broadly. This work has sought to develop and harness a structured and systematic methodology to investigate how particular signal events come to influence how people, think, feel and behave in relation to their security. Applied to the problem of homicide, it is an approach that affords an opportunity to trace out, in some detail, how different types of major crime impact upon community perceptions and experiences of security. This is of value in that, as part of their standard operating procedures for responding to homicides, police are increasingly seeking to utilise ‘community impact assessment’ instruments to measure the consequences of individual crimes within and across particular communities, in order that they can better manage the social harms associated with such crimes. These harmful impacts can relate to particular facets of the crime itself, but also policing interventions that sometimes amplify rather than reduce levels of community insecurity.

The analysis reported is based upon data derived from two research programmes, including:

A series of four intensive field-work based case studies conducted between 2007 and 2009 exploring the impacts of particular homicides upon the local communities in which they occurred;
Fieldwork conducted as part of the National Reassurance Policing Programme between 2003 and 2005, during which qualitative research was conducted on community reactions to a number of nationally high-profile homicides.

For all the incidents studied, a similar methodology was employed, utilising face-to-face in-depth interviews conducted with a sample of people living near to where the crime occurred (see Methodological Appendix).

Informed by these data, that have been integrated at the analytic stage, the research identifies five ‘impact models’ in terms of how homicides shape the perceptions and experiences of individuals, groups and communities. The notion of an impact model is employed to capture and articulate broad empirical patterns that can be detected in terms of how particular incidents influence and shape the levels of confidence and reassurance within and across different community contexts. Stated briefly, the five impact models are:

Personal: where the impact of the crime registers acutely with those with direct ‘strong social ties’ to the victim, such as their family and friends. There are though few repercussions beyond this social network;
Parochial-localised: is where in addition to the family and friends of the victim, the crime alters the security needs across a neighbourhood. The effects are then structured on a geographic basis;
Parochial-distributed: is where the consequences are felt across an extended social network, but one that is organised on a basis other than geographic proximity. So for example, if a homicide victim is from a minority ethnic community, then the ‘fear’ generated by such an incident might be detectable amongst other members of that group, even though they do not live near where the crime occurred.
Public: is where the anxiety and trepidation created by a particular crime ‘travels’ across multiple communities;
Public with parochial counter-reaction: is a pattern whereby, because they are in possession of local knowledge, a neighbourhood community develops a different, more balanced, understanding of the presenting risks and threats, from the public ‘at large’.

The contours of these models will be discussed in more depth and detail in due course.

In developing these impact models in order to try and understand how and why particular major crimes have profound consequences, we have sought to measure the impact of these incidents along two dimensions:

Public confidence – herein is used to gauge attitudes to the police;
Reassurance – is concerned with levels of fear and insecurity more generally.

This reflects an attempt to recognise the fact that homicides are potentially important because of their consequences for how members of the public view the police, but also because of their implications for the overall sense of community well-being and resilience. Before discussing these issues in more detail, the report commences by quickly seeking to position the substantive focus in relation to our broader understandings of homicide, and why it is so important. This is followed by a discussion of what we know about the factors shaping confidence in policing. Informed by empirical data collected through a series of field studies the report then examines the impact models in turn to determine their key characteristics. Additional sections investigating how levels of confidence and reassurance after homicides shift over time, and the influence on this of the police response are then included. The final section of the report considers the implications for the policing of homicide, the conduct of community impact assessments and the measurement of public confidence and reassurance.

POSITIONING THE STUDY

The focus of the current work upon seeking to develop an evidence-based understanding of the community impacts of homicide on public confidence and reassurance reflects the fact that whilst there is a growing body of social scientific literature that maps and investigates different aspects of social reactions to homicide, including studies of: how it is constructed in fictional and factual media representations (Jenkins, 1994); the police investigative response (Innes, 2003a; Foster, 2008); the repercussions for victims families (Rock, 1998); and the stigmatisation of offenders friends and families (Condry, 2007). And whilst there are increasing numbers of studies that map the aetiology (Polk, 1994), victimology (Brookman, 2005) and geographic distribution of fatal violence (Shaw, Tunstall and Dorling, 2005), there is very little systematic empirically grounded research that has considered how crimes and homicides may shape and alter the social contexts in which they occur. It seems to have been assumed that such crimes have a significant impact upon those areas where they arise. However, emerging evidence suggests that in reality the picture is more complex. For example, Carr’s (2005) ethnographic study of a Chicago neighbourhood, recounts how, following two separate shootings in the area, neighbourhood residents sought to collectively mobilise to respond to the perceived threats signalled by these crimes. However, on both occasions the nascent efforts failed. Following a third incident though, a more concerted attempt to generate enhanced ‘collective efficacy’ emerged that, according to Carr resulted in a number of improvements to the quality of life in the neighbourhood.

The broad tone of Carr’s analysis is in keeping with the conceptual framing of social reactions to crime found in the ‘signal crimes perspective’ (SCP). The SCP posits that some crime and disorder incidents act as warning signals to people about the distribution of risk to their everyday security, and exposure to these signals structure peoples’ subsequent beliefs and behaviour (Innes, 2004). Importantly, empirical research informed by this conceptual scaffolding found that strong public reactions can be induced by a range of issues from antisocial behaviour (Bottoms and Wilson, 2006) to high profile murders (Innes, 2003b).

That major high profile crimes can elicit significant reactions is well documented. For example, Chancer (2005) analyses a number of high profile American crimes as case studies to show how they were generative of important legislative changes. Similarly, Jonathan Simon’s (2007) recent book has unpacked how a number of pivotal incidents have profoundly shifted the framing of crime more generally in political discourse and action. Closer to home, signal crimes such as the murder of Stephen Lawrence have been harnessed as the catalysts for large, politically driven, institutional reform efforts (Rowe, 2007; Rock, 2004). But whilst there are studies of the effects of particular cases upon the institutional architecture, and the politics and culture of criminal justice, the slightly more ‘mundane’ question about what impact do such crimes have upon the neighbourhoods and communities in which they occur has not been systematically or thoroughly addressed.

This lacuna in the evidence-base for policing exists despite growing police awareness that major incidents can have pronounced and long-lasting negative consequences for public security, reassurance and confidence in the police. To address such matters at the level of practice, community impact assessment (CIA) techniques are increasingly deployed as part of the police response to major crimes and critical incidents. However, to date little work has evaluated the efficacy or effectiveness of these techniques, either in terms of their ability to accurately measure the impacts of particular crimes, nor in steering responses that can mitigate the social harms associated with them. Neither has a connection been constructed in terms of the policing of homicide and levels of public confidence in the police, despite the fact this is now a key performance indicator.