THE DENIAL OF VOICE AND THE REMOVAL OF RESPONSIBILITY: REDUCTIONS TO THE CRIME SURVEY ENGLAND AND WALES
New cost-saving measures applied to the Crime Survey for England and Wales from 2017/18 will reduce the number of people interviewed each year and so challenge the ability of academics, police forces and policy makers to draw on a robust local evidence-base that is much needed to inform and steer decision-making and service delivery tailored to local needs. That the Crime Survey is being squeezed again at a time when the in-house analytic capability of police forces is at an all-time low, when forces are no longer mandated to routinely survey victims of crime and many have stopped local public surveys to save money, is of major concern when all the signs point to rising crime and vulnerability in society.
Emma Williams – a little history
Last night (November 15th) I attended Professor Ben Bradford’s inaugural lecture at City Hall in London. The paper was focused on policing diversity, immigration and the impact of this on police legitimacy. It was a brilliant paper which yet again highlighted to me the clear ignorance around the governments’ decision to reduce the questions in the Crime Survey England and Wales.
I will avoid making this blog a history of certain criminological theories and the influence they had on the development of victimisation surveys in the UK. But I cannot write this without making reference to Jock Young and the Left Realist school. These writings exploring and questioning ‘the truth’ of official crime statistics and the lack of discussion about the dark figure of crime led to the development of victimisation surveys and the British Crime Survey, which later became the Crime Survey England and Wales (managed by ONS). Such data finally shed light on the reality of victimisation for many communities and groups in the UK. Without such insight we would not effectively understand multiple and repeat victimisation, incidences of hate crime, the reality of domestic abuse or the reality of how certain areas and groups experience crime and violence differently and disproportionately. Plus, critically, just how much of this goes unreported to the police.
The crime survey plays a vital role in helping us understand how the public view the police which is essential to understanding the level of legitimacy the public feel the police have in the UK. The surveys reveal issues around differing definitions of what constitutes a crime, personal narrative about harm or an act of violence. Such insight helped to confirm feminist criminologists’ exposure of the level of threat women feel as a result of their gender. This allowed for some real challenge to the notions of women’s fear of crime being irrational as it revealed the type of everyday perceived violence that women can experience. The dark figure of crime and victimisation is vital to understand – it facilitates (or should) more focused resource deployment, policy and strategic initiatives aimed at informing the community and dealing with their concerns, local policing plans and targeted work where certain groups are overly represented or have lower levels of confidence in the police. Force data and local surveys have largely gone due to severe cuts but now it seems it is deemed as ‘not a priority’ by our government also.
Helen will talk more about some of the practical implications of these cuts later in the paper but I want to briefly discuss them in relation to something Ben raised last night. I cannot do his words justice in this short piece but what his analysis showed us is how much we need to understand the experiences of immigrants as just one vulnerable group, and how much a sense of identity and belonging in a new country / community is linked to their perceptions of legitimacy and trust in the police. The cutting back of questions to the crime survey inhibits and limits the exploration of this groups’ experiences of crime and policing – particularly given how much crime experienced by these groups goes unreported. The crime survey can help us track issues over time – indeed by exploring perceptions of the police we might even be able to further understand our social order and our ability to integrate and include immigrants within the UK. This understanding gives us a broader sense of understanding our social order – a gauge on which to consider our wider social order not just simply victimisation.
Therefore, when I read articles about how the police are no longer dealing with low level disorder and minor crime – predominantly I would argue because they have had to make decisions based on such severe budget cuts – it makes me think that there may be more reasons for the cuts to the CSEW than simply money. If we don’t ask about some of our most vulnerable citizens experiences of crime, harm and victimisation, if we don’t fully understand the disparity in crime and victimisation then we a: can justify and excuse ourselves for not dealing with it and b: leave some very serious questions about our country’s social order and willingness to deal with the type of exclusions that more and more of our communities face, unexplained. If the evidence is not there then we don’t need to deal with it………..
Dr Helen Innes – some practicalities
So having looked at why the CSEW matters, I would like to turn to explore how reducing the overall sample size by 600 households (from 35,000 down to 34,400 annually) and the survey response rate to 71 percent will have a detrimental impact upon the conduct of evidence-based policing and other forms of research. The kinds of research that the Crime and Security Research Institute at Cardiff University, and other similar units, have been able to conduct at the request of some police forces and Police and Crime Commissioners has used local data in the Crime Survey to directly inform difficult decisions about service re-configuration and delivery in today’s austere climate.
The loss of 600 households from a continuous national survey of around 35,000 doesn’t sound a lot, but it is important to put this in a longer-term perspective. For it represents a further ‘chipping away’ at a sample that at its peak included 46,000 interviews across England and Wales. Whilst the total number remains large, the effect of these cumulative reductions is felt disproportionately when it comes to sub-national or local analyses, such as those using the geographical unit of police force area (PFA) to align with the agendas of locally elected Police and Crime Commissioners (PCCs).
As part of a broader move to empower local communities, PCCs were introduced in 2012 to play a key role in driving innovation and reform in local policing priorities to accord with local need and to evidence these in their annual crime plan. On average, however, the number of interviews per PFA has fallen markedly since the Crime Survey became continuous in 2001; from 1,000 in each police force area down to 650 in 2012-13 and cut again today by an average of 13 interviews per area, or a total of 637. The numeric effect of this reduction will not be uniform across England and Wales because some force areas (e.g. the Metropolitan Police) are larger than others, but it does mean that the opportunity to leverage local insights from the survey will be further compromised across the country.
The impact on evidence-based research is twofold. First, it follows that a smaller sample for each police force area places more stringent access conditions on the data for researchers given that there is an increased risk of disclosure of respondents’ personal details. The danger is that this limits the reach and potential of local data (which may well be the only consistent and representative survey sample of victims and non-victims of crime available) to a select few in academia. Second, of course, is that a smaller sample reduces the precision of estimates from the data, meaning that confidence in the reporting of findings from any single year becomes limited at a local level, particularly when looking at important but small sub-groups within a sub-national population such as victims of crime.
This matters because the Crime Survey is so much more than a counterpoint to police-recorded crime statistics that report each quarter if, and what, crime is going up or down. Being able to see the volume, trend and clustering of crime is important nationally and regionally as the nature of crime itself changes and evolves, but so too are peoples’ reported experiences of crime and policing. How these intersect with where and how people live and the drivers of vulnerability to social harm is something that will vary between and within police forces. It is to be hoped that, in reducing the response rate to this survey, there is no disproportionate impact on capturing the voices of those most vulnerable to victimisation and repeat victimisation.
Although we remain fortunate in this country to have a number of high-quality surveys capturing data on public attitudes, family, education, health and well-being, rarely do they permit these areas to be connected with public opinions on crime and policing. This is particularly the case in Wales, for example, where policing is not a devolved responsibility.
The Crime Survey really is the best large-scale tool we have for understanding and anticipating demand on police and victim services in England and Wales, what crime means to people, how it harms them and where we should look to prioritise action and intervention given limited resources across the whole public sector. An example of this from our own work is a local Crime Survey analysis we did on behalf of a Police and Crime Commissioner focusing on public perceptions of victim support services within their force area. By highlighting areas of unmet need based on the qualitative harm associated with reported victimisation rather than the category of crime itself, this evidence informed commissioning decisions about the future direction of quality and delivery of these services.
It is now up to individual police forces and commissioners to decide their ongoing commitment to surveying victims of crime and utilising survey-based evidence to best tailor the services they can provide to their public. Facing this choice, a pertinent question might be whether it is enough to rely on robust ‘national’ indicators (e.g. of victim satisfaction) that the Crime Survey of England and Wales provides rather than drift towards a more fragmented, piecemeal picture of regional variability from locally commissioned, small scale surveys utilising diverse sampling techniques and questions. It would be foolish to deny that cost-efficiencies have to be made, but perhaps an alternative in this case would be for PCCs to come together collectively to financially support the Crime Survey in achieving a sample that makes it robust, useful and comparable for locally oriented analyses in England and Wales.