8.4.2 CCTV effectiveness and Context
When the Government funded the massive growth in CCTV across the UK there was no body of research to justify and guide the implementation of CCTV (Ditton and Short, 1999; Farrington and Walsh, 2002). Subsequently the effectiveness of CCTV across a number of contexts has been explored and research has started to establish an evidence base for where and how CCTV can be effective. Many of the studies into CCTV have produced contradictory results due to variations in the circumstances of the introduction of CCTV leading to varying effects (Tilley, 1998).
Research that utilises the scientific realism approach developed by Pawson and Tilley (1997) tried to identify how CCTV works and specifically in what contexts (see Tilley 1993, Gill and Spriggs, 2005). Academics (Armitage et al, 1999; Tilley 1993) have documented several ways or mechanisms that could result in CCTV bringing about change in an area and those devised by Tilley (1993, quoted in Gill and Spriggs, 2005) are as follows:
- Caught in the act - CCTV could reduce crime by increasing the likelihood that present offenders will be caught, stopped, removed, punished and therefore deterred.
- You've been framed - CCTV could reduce crime by deterring potential offenders who will not want to be observed by CCTV operators or have evidence against them captured on camera.
- Nosey Parker - a reduction could take place because more natural surveillance is encouraged as more people use the area covered by CCTV. This may deter offenders who fear an increased risk of apprehension.
- Effective deployment - CCTV may facilitate the effective deployment of security staff and police officers to locations where suspicious behaviour is occurring. Their presence may deter offenders, or may mean they are caught in the act.
- Publicity (general) - this may assist in deterring offenders.
- Publicity (specific) - CCTV cameras and signs show people are taking crime seriously, and thus offenders may be deterred.
- Time for Crime - CCTV may have less of an impact on crimes that can be done quickly as opposed to those that take a longer time, as offenders assume that they will have enough time to avoid the cameras, or to escape from police officers and security staff.
- Memory jogging - publicity about CCTV encourages potential victims to be more security conscious and to take precautionary measures.
- Appeal to the cautious - those who are more security minded use the areas with CCTV, driving out the more careless who are vulnerable to crime elsewhere.
The list above represents a starting point to consider how CCTV can impact on crime and numerous other mechanisms can be developed across a range of settings and offence types (Ratcliffe, 2006). Coupe and Kaur (2005) examined the impact of CCTV and alarms in detecting commercial burglary and they highlighted the complex interplay of mechanisms that can result in CCTV impacting on crime and how different crime prevention measures can have conflicting mechanism.
CCTV can provide evidence on film that leads to arrest, while visible CCTV cameras like alarms, may also deter burglars or displace them to other targets. In addition, visible or hidden CCTV cameras may alert a watchman or employee to the commission of a crime. On the other hand, activated alarms may frighten burglars so that they quickly flee the scene, reducing not only capture there, but also, where CCTV is additionally fitted inside the premises, of a subsequent arrest by catching the offender on film.
(Coupe and Kaur 2005: 53)
Sivarajasingam, Shepherd and Matthews (2003) examined the impact of CCTV in town/city centres and detailed how theoretically the cameras may impact on levels of violent crime:
Perpetrators may be detected and removed; CCTV may deter potential offenders who perceive an increased risk of detection; CCTV may direct security personnel to locations where precursors to offending have been detected, which may head off their translation into crime and reduce the severity of harm; CCTV could symbolise efforts to take crime seriously, and the perception of those efforts may both energise law abiding citizens and/or deter crime. The presence of CCTV may induce people to take elementary precautions, for fear that they will be shamed by being shown on CCTV.
(Sivarajasingam et al, 2003: 315)
The mechanisms outlined above highlight the potential problems of using recorded crimes rates to evaluate the impact of CCTV as the different mechanisms can have conflicting effects on crime rates (Ditton and Short, 1999: 212; Ratcliffe, 2006, Gill et al, 2007: 24). Although CCTV will not increase actual levels of crime the increased surveillance may result in more offences coming to the attention of the police, particularly violent offending (Brown, 1995). Using disaggregated crime data that identifies changes across individual offence types can help to understand the impact of CCTV across a target area. The range of additional crime reduction measures that often operate alongside CCTV system make if difficult to isolate the impact of the cameras and these can include changes to policing practices (Webb and Laycock, 1992), ad hoc police operations, improved lighting, community wardens and youth inclusion projects (Gill et al, 2007). Using crime statistics alone to evaluate CCTV means that many of the potential benefits of the cameras can be missed including supporting police activity leading to cost savings in relation to police time, increased detection rates, court time and the increased level of guilty pleas and guilty verdicts obtained when CCTV evidence in available (Home Office, 2007).
CCTV can work on a number of different levels across a range of different contexts and this has resulted in mixed research findings in terms of CCTV effectiveness. Welsh and Farrington (2002) conducted a meta-analysis on studies of CCTV effectiveness and collected 46 studies but only considered 22 of the research papers to be rigorous enough for inclusion in their review. Half (eleven) of the studies found a desirable effect on crime, five found an undesirable effect on crime, five found a null effect, and one was classified as an uncertain effect. The largest impact on CCTV was found across car parks where there was evidence that crime reduced by 41% in the experimental compared to control area, which was significant. The research identified that CCTV had little or no effect on violent crime but the authors advocated the need for more high quality research that 'established the causal mechanism by which CCTV has any effect on crime' which should involve methodologically rigorous evaluations and interviewing offenders. A further meta-analysis of CCTV studies conducted in 2008 by Walsh and Farrington the confirmed earlier findings that CCTV was effective in car parks and they advocated narrowing the use of CCTV to reflect research findings related to its effectiveness.
The Home Office's National Evaluation of CCTV (Gill and Spriggs, 2005) attempted to address some of the deficiencies identified in previous CCTV evaluations by combining a process and impact evaluation that incorporated control areas and identified other crime control initiatives that were operating in the target area to evaluate their impact on recorded crime levels. Thirteen CCTV systems were evaluated across a range of system including town centres, city centres, car parks, hospital and residential areas. The inclusion of residential areas reflected the governments push to include these types of areas into the Phase 2 of the Crime Reduction Programme (Home Office, 2007). The main findings were:
Out of the 13 systems evaluated six showed a relatively substantial reduction in crime in the target area compared with the control area, but only two showed a statistically significant reduction relative to the control area, and in one of these cases the change could be explained by the presence of confounding variables. Crime increased in seven areas but this could not be attributed to CCTV. The findings in these seven areas were inconclusive as a range of variables accounted for the changes in crime levels, including fluctuations in crime caused by seasonal, divisional and national trends and additional initiatives.
(Gill and Spriggs, 2005i)
The quotation above highlights the difficulties in obtaining a true picture of the impact of CCTV schemes given the complex environments where they often operate. The research concluded that for CCTV to be effective it needs to be implemented with a clear strategy that takes into account local crime problems and identifies the mechanism by which the system will address the problems. CCTV should not be implemented as a stand alone crime prevention tool but needs to be integrated into prevention measures already in place and operate alongside local police structures to create rapid responses to incidents and effective use of images for evidential purposes.
Research has found mixed results regarding the effectiveness of CCTV but what has emerged is a body of literature that has started to identify the specific context where CCTV works and types of mechanisms that need to be in place. Brown (1995) found that in certain circumstances CCTV can make a positive contribution to addressing crime and this was reliant on CCTV being used by the police as an integral part of a command and control strategy. The research highlighted the deterrent effects of CCTV but suggested that CCTV must also be used to effectively manage police resources through rapid responses to incidents. Brown found that the use of CCTV in Newcastle and King's Lynn resulted in a reduction in recorded crime particularly across burglary, criminal damage and vehicle crime offences. The research found CCTV had no overall impact in Birmingham on crime levels (Brown, 1995: 46). Research indicated that the layout of the streets has an impact on the ability of CCTV to detect crime and areas with less side streets and more long straight roads more conducive to CCTV (see also Gill et al 2005b).
Ditton and Short (1999) found that recorded crime fell and detections rose after CCTV was implemented in Airdrie but in Glasgow recorded crime increased and detection increased. The research found a differential effect of the cameras across crime type with drug offences, low-level public order and minor traffic violations increasing whilst various forms of acquisitive crime fell. Airdrie is a little town where awareness and a sense of ownership of the cameras were high compared to Glasgow where the cameras merged into the structure of the city, and these situational differences may have impacted on the effect of the cameras.
Most evaluations of CCTV in town/city centres have used recorded crime and found that cameras had very little impact on violent crime (Gill and Spriggs, 2005) but through the use of accident and emergency data Sivarajasingam et al (2003) found that the 'effectiveness of CCTV lies less in preventing assaults and their precursor, but more in preventing injury through increased police detection and intervention' (ibid: 315). CCTV was found to increase police detection but was associated with reductions in the seriousness of violent incidents. There was no evidence of the deterrent effect of CCTV in relation to violent crime but the effectiveness of CCTV within this context is related to surveillance facilitating a faster police response that limits the length of violent incidents and therefore the severity of injuries.
The body of CCTV research literature emphasises that it is more effective in relatively simple target areas with clear lines of sight. CCTV has been shown to reduce crime in car parks (Tilley, 1993; Gill and Spriggs, 2005) and this may be partly explained by the cameras monitoring an environment where access and egress can be carefully monitored. CCTV in car parks has reduced crime by acting primarily as a deterrent and this mechanism has been facilitated by clear CCTV signage and the visibility of the cameras (Gill and Spriggs, 2005). The impact on 'theft of' and 'theft from' motor vehicles was different with a larger positive impact across 'theft of' offences which may be due to these types of offences taking longer and offenders having to drive out of exits monitored by cameras (Tilley, 1993).
CCTV systems rarely work in isolation and often form part of a crime prevention strategy. Webb and Laycock (1992) found evidence that CCTV can reduce robberies on the London Underground but the cameras were part of a package of measures to reduce crime in the area that made it difficult to identify the impact of the cameras alone. The research concluded that 'CCTV does not seem to be very useful in large complex and crowded environments to deal with surreptitious behaviour such as pick pocketing or shoplifting' (Webb and Laycock, 1992: 23) as the quick nature of the offences made it unlikely that they would be picked up by operators. Given that it was unlikely offenders would be detected by the cameras their effectiveness was mainly linked to whether offenders associated the cameras with an increased risk of getting caught on the London Underground.
CCTV is a type of situational crime prevention and is often used to facilitate a change in the behaviour of offenders. Mayhew (1984) suggested that formal surveillance would deter potential offenders and this follows the rational choice theory perspective (Clarke and Felson, 1993) that proposes offenders act in a rational manner and by calculating whether the perceived benefits outweigh the cost in a given situation. The application of the deterrent effect of CCTV to routine activity theory means that the presence of CCTV can be perceived to act as the capable guardian and therefore demotivate offenders. The majority of CCTV systems rely on the deterrent effect of the cameras but the deterrent is often symbolic and 'more or less incompetent deterrence because cameras are highly visible but those under surveillance are hardly visible for an observer due to irregular monitoring, informational overkill or even deployment of dummy cameras' (Hempel and Topfer, 2004: 33).
Research has examined the effect of CCTV on offenders' behaviour across a range of contexts and identified that CCTV tends to be an effective deterrent against planned offences. Allard, Wortley and Steward (2008) examined whether the presence of CCTV in prisons reduced the number of incidents that were defined as 'breaches of law or rules that may result in criminal prosecution or breach hearings and emergencies'. The research found that CCTV had a greater impact on non-violent than violent prisoner misbehaviour and affected planned behaviour to a greater extent than unplanned behaviour. The spontaneous nature of violence means that the deterrent effect of CCTV can be removed and it tends to be more 'effective when behaviour is motivated' (Allard et al, 2008: 416).
Research into public space CCTV has identified similar patterns and indicated that CCTV impacts more on premeditated crimes (Brown, 1995; Welsh and Farrington, 2002; Gill et al, 2005). Analysing the impact of CCTV on public behaviour, Mazerolle (2002) found that the cameras created an initial deterrence in the two-month period after installation but to prolong the effect recommended increasing the deterrence of using signs and short sporadic cameras deployment. Tilley (1993: 24) suggested that 'when the real potential of CCTV to lead to apprehension loses credibility amongst criminals, the effect will begin to fade, though by (over)-statement of successes periodic effectiveness can be re-established'. The positive impact of CCTV on levels of robbery in the London Underground was found to fade over time and this may have been due to offenders discovering that the CCTV did not increase the risk of being caught (Webb and Laycock, 1992: 15). High camera density and quality lighting may increase the perceived risk for offenders (Gill et al, 2007). Research indicates that only by combining the different mechanism by which CCTV works (Armitage, et al, 1999, Tilley 1993, Gill and Spriggs, 2005) and integrating other crime prevention measures can the optimal use of CCTV occur and research is currently building the evidence base to fully understand where and how the mechanisms work.
CCTV does not create a physical barrier to crime and therefore can rely to a large extent on changing offenders' behaviour. Therefore key to the success of CCTV is offenders' views regarding its effectiveness. Evaluations that use crime levels to investigate the impact of CCTV on offenders need to be supplemented with offender interview based research to develop a full picture of how CCTV can be utilised fully to address criminal behaviour (Farrington and Walsh, 2002; Gill and Loveday, 2003). Gill and Loveday interviewed 77 convicted offenders in prison and the general consensus amongst those interviewed was that they did not worry about CCTV but there was evidence that some offenders chose to take precautions against the cameras by wearing clothes that hid their identity or offended in camera blind spots. Many of the offenders committed 'swift offences' and therefore believed that police notified by the cameras would not arrive in time to apprehend them (see also Short and Ditton, 1998). Roughly half the sample of offenders believed that CCTV increased the risk of getting caught but those that had been caught by CCTV perceived it as more of a threat. There was a lack of understanding amongst the offenders regarding image quality and how the images could be used to increase detection. The types of mechanisms that need to be utilised to increase the perceived risk of CCTV for offenders include using publicity detailing successes of the cameras and the capabilities of systems.