View this issue as a PDF: CrimHighlightsV7N2.pdf
This issue of Criminological Highlights addresses the following questions:
- Can school based activities for youths reduce crime?
- How should “racial profiling” be determined?
- Is offensive language an important determinant of citizens’ assessment of police?
- Why did England’s Crime Reduction Program fail?
- Does prison act as a deterrent?
- Do laws allowing governments to confiscate the proceeds of crime work?
- Can sexually violent predators be reliably identified?
- How should police performance be measured?
Grade 8 students from schools in which youths spend a lot of their time in unstructured, unsupervised activities have relatively high levels of delinquency.
Rates of delinquency vary dramatically across groups. One way in which groups of youths differ is in the amount of time spent socializing without supervision or away from authority figures. Previous studies have shown that the amount of time an individual spent in unstructured socializing was related to offending, drug and alcohol use, and dangerous driving.
This study goes one step further and asks whether youths who go to schools in which large numbers of youths have amounts of substantial unstructured, unsupervised time are more likely to offend above and beyond any individual characteristics of the youths themselves. This makes sense in part because if the social environment in which youths find themselves is largely unstructured and unsupervised, the youths will have more opportunities to commit offences.
Over 4,000 Grade 8 students from 36 schools in 10 U.S. cities answered questions about a number of things including how much time they spent “hanging around with your current friends, not doing anything in particular, where no adults are present.” They also reported their involvement in delinquent activities. Not surprisingly, the amount of time that the youth spent in unstructured, unsupervised activities was related to self-reported delinquency. The time spent in unstructured activities was more important than other individual characteristics of the youth (sex, race, parents’ education). More interesting, however, were the “school” effects: students attending schools in which youths generally spent lots of unstructured, unsupervised time were more likely to report high levels of delinquency above and beyond the youth’s own reports of time spent in unstructured activity. Said differently, attending a school in which many youths spend a lot of time “hanging out” with their friends away from adults is likely to increase the level of delinquency above and beyond the time that the individual youths spend in unstructured activities. In effect, the results show that the local culture (in this case the youth’s school) has an impact on a youth independent of his or her own circumstances.
Conclusion. Situations that are particularly conducive to delinquency are likely to occur when a youth engages in unstructured socializing away from authority figures. Schools with many youths who spend time “hanging out” in unsupervised settings are, therefore, likely to increase opportunities for offending for all adolescents in that school, regardless of how much time they themselves spend in unsupervised activities. The results would imply, therefore, that one way in which communities can minimize offending would be to provide attractive opportunities to youths that would encourage them to spend more time in supervised activities. If this were done within a community, the results of this study would suggest that there should be a reduction in the level of offending for all youths – those who spend a lot of time hanging out with their friends in unsupervised settings and those who do not.
Reference: Osgood, D. Wayne and Amy L. Anderson (2004). Unstructured Socializing and Rates of Delinquency. Criminology, 42(3), 519-549.
Do the police engage in disproportionate stops of people of certain races? The answer may depend on what is meant by ‘disproportionate.’
‘Racial profiling,’ or the disproportionate stopping (and/or searching) of people of non-dominant groups, is obviously a contentious issue in many countries including England, the United States, and Canada. Simply put, there are two problems in assessing whether racial profiling is taking place. First, one has to measure or estimate the numerator – the number of people of various groups who are stopped and/or searched. Second, one has to measure or estimate the denominator – the proportion of each group in the population.
Each of these measurement tasks is problematic. Getting valid data on the group membership of those stopped depends on having adequate police records, self-reports of citizens, or direct observations by third parties. Denominator data are equally, if not more, problematic. What population is one trying to estimate? The problem is that the population of residents may not reflect the population of people who are available to be stopped. Simply put, if the police are expected to stop people on the street in proportion to their membership in racial groups, one has to know the racial makeup of those on the streets not the racial makeup of the population as a whole.
This study argues that “the use of the residential population as the comparative basis [for police “stop” statistics] is problematic, not least because different sections of the population may use public space differently” (p. 890). In other words, the over-representation of any group (e.g., young males, blacks) may simply reflect the fact that the members of the group are more likely to be available in public places where the police stop citizens and therefore are at greater “risk” to be stopped. In two cities in England (Reading and Slough) observers systematically counted the numbers of white, black, Asian, and ‘other’ people in public areas where the police were known to be likely to stop and search people on the street. Thousands of people in each city were counted. These data were then compared to official records of stops and searches. Compared to census data, blacks were more likely to be stopped and searched in both cities. In Slough, for example, blacks made up 6% of the population but were the subject of 15% of the stops. However, blacks made up 17% of the available population (the population of those on the street). The data for Reading were similar. Blacks were not over-represented in stops when compared to the “available” population, but were when compared to the census figures. Asians in Slough were under-represented in stops. They made up 40% of the “available” population but were the subject of only 31% of the stops. In addition, young men had a dramatically higher likelihood of being stopped and searched than their numbers on the street would have predicted.
Conclusion. The study demonstrates the importance of considering in a more sophisticated manner what the “expected” rate of stops would be for various groups if “profiling” did not exist. In this study, resident population figures for two cities would suggest racial profiling was taking place, while available “street population” estimates would suggest that blacks in particular were not the subject of racial profiling. The analysis suggests that one not only needs analyses of profiling that take into account the “available” population, but one also needs to consider why certain locations (as opposed to certain groups) are identified as appropriate for increased police proactive stops.
Reference: Waddington, P.A. J., Kevin Stenson and David Don (2004) In Proportion: Race, and Police Stop and Search. British Journal of Criminology, 44, 889-914.
Offensive language by police officers is at least as important as their behaviour in determining the way they are seen by ordinary citizens.
What are the important dimensions of misconduct by the police from the perspective of ordinary citizens? Traditionally, police misconduct in relation to interactions with citizens has been categorized as involving three dimensions: the use of unnecessary force (e.g., hitting or beating a citizen), abuse of authority (e.g., threats or the refusal of the officer to give his/her badge number/name), and discourtesy or the use of inappropriate language (e.g., racial slurs, insulting language). This paper examines the relative importance of these dimensions in determining how police are seen by ordinary members of the public.
Eleven hundred New York City residents were each read a set of short vignettes describing an interaction between a police officer and a citizen. The officer’s language was described in neutral terms or in a range of different discourteous or obscene terms such as by calling the citizen a “fuckin’ piece of trash” (p. 686) or using a racial slur. Abuse of authority was manipulated by simply stating that the officer threatened to arrest the citizen or engaged in a range of different forms of abuse such as “threatening to grab or kick the civilian”, or “refusing to provide a name or badge number” (p. 686). The use of unnecessary force was injected into some scenarios by saying such things as the officer “punched the civilian” or “drew his or her gun and aimed it at the civilian” (p. 687). The event precipitating the citizen-police interaction was also described in various ways. Some were ambiguous (e.g., the police officer was described as simply stopping the car and asking the citizen for his or her driver’s licence, etc.) while in other cases the citizen was described as having been observed committing an offence.
One might have expected that the rated seriousness of the misconduct would increase incrementally as one moved along a continuum from offensive language through abuse of authority to the use of unnecessary force. This was not the case. Independent of the reason for the encounter, the description of the civilian’s response to the officer and various other factors, “a police officer’s discourtesy or offensive language remained highly salient as an explanation of the respondent’s evaluation of the seriousness of misconduct” (p. 691). Language, it seems, matters and it matters a lot. In particular, “unnecessary force in the presence of offensive language has a greater impact on… ratings” (p.692) than did abuse of authority (though abuse of authority did add significantly to the rated seriousness of the misbehaviour).
Conclusion: Offensive language “may be part of everyday speech [but] it carries a very different meaning when voiced by police officers” (p. 702) in an encounter with a citizen. Along with abuse of authority and use of unnecessary force, language turns out to be very important in shaping citizens’ views of the police. At the same time, however, non-cooperative behaviour on the part of the citizen does lessen, somewhat, the rated seriousness of police misbehaviour. The mitigating impact, however, is small compared to effects of police misbehaviour. Though the public may, under some circumstances, tolerate police misconduct, “the public’s tolerance for [police] misconduct in an encounter with a civilian does not extend to unnecessary use of force” (p. 703).
Reference: Seron, Carrroll, Joseph Pereira, and Jean Kovath. (2004). Judging Police Misconduct: “Street-Level” versus Professional Policing. Law and Society Review, 38, 665-710
The Crime Reduction Program (1999-2002) in England & Wales was a failure because of the manner and speed with which it was implemented and because it did not reflect what was known about crime reduction.
In 1999, an ambitious crime reduction program was implemented in England and Wales. Substantial resources were allocated to evaluations so that the best programs could be chosen for implementation across the country. Building on an optimistic review of what was known about crime reduction, the Crime Reduction Program (CRP) seemed to offer something to everyone, encouraging “the notion of building up a store of ‘scientific’ knowledge about ‘what works’ which could be translated into reliable advice on ‘best practice’, [and] disseminated to practitioners in user-friendly form” (p. 216).
Government funding structures, however, demanded rapid identification and implementation of crime reduction programs and of the evaluations before adequate planning and staffing could take place. The initial plan for a ten-year lifespan of the program was in conflict with the electoral cycle of government which needed early successes. What started as a series of experiments on crime reduction quickly morphed into programs with explicit crime reduction targets (a 25-30% reduction in five years) that were based on political wishes rather than empirical assessments of what was possible. Local communities were asked to design their own programs rather than to build or implement programs based on known best practices. Staffing these programs was difficult within the governmental requirement to spend money before the end of each fiscal year. Staff turnover was high since stability of employment could not be guaranteed. Programs were overly ambitious in terms of the nature of the promised interventions (since, among other things, programs were funded on a competitive basis). The result was that few programs were fully implemented in a constructive manner. Given the nature of the funding, community organizations tended to treat them “as a ‘windfall’ to contribute to pre-existing plans….” (p. 223). As time went on, there was more pressure to deliver “tangible crime reductions, at the expense of longer term pay-offs in terms of research knowledge” (p. 224). This favoured the targeting of high volume, less serious crime and simplistic hardware-oriented approaches (e.g., CCTV), since implementation and larger reductions appeared more plausible in these domains.
Evaluators, who were sometimes aware of the problems of design and implementation, were forbidden to pass their knowledge on to the program implementers. The result was that many of the evaluations carefully documented the implementation failure of the program in rich detail but could not determine if there had been an effect on crime.
Conclusion. By creating impossible crime reduction targets, the crime reduction program was not seen as delivering the politically necessary results when crime (across the country) began to rise. The result was that “politicians were quick to lose patience with the program as initially conceived” (p. 231) and brought it to a close after three rather than ten years. The problems faced by the program appear to have been “tied up with broader tensions and contradictions inherent in the differences in cultures, perceptions and time-frames of policy makers and politicians, practitioners and academics, including differences in their understandings of the nature, purposes and reliability of ‘research’ itself” (p. 232).
Reference: Maguire, Mike (2004). The Crime Reduction Programme in England and Wales. Reflections on the Vision and the Reality. Criminal Justice, 4, 213-237.
In 1993, Michael Howard, then Home Secretary and subsequently head of Britain’s Conservative Party, suggested that “prison works,” contradicting the same government’s 1990 statement that “prison was an expensive way of making bad people worse” (p. 390). His statement was, of course, consistent with those of many others at the time (e.g., James Q. Wilson), but he also cited British research saying that, for prisoners, “avoidance of imprisonment was the most frequently mentioned reason for not wanting to reoffend” (p. 392). That research – whose purpose was to investigate “the contradiction between such allegedly rational calculus and subsequent… offending” – also showed that 62% of prisoners reported that they had reoffended in the two years following release. Nevertheless, the authors of the cited research suggest that “the stated desires, intentions and beliefs of offenders cannot be meaningless and are probably related to how people act” (p. 393).
Shortly before this group of 130 men was about to be released in the early 1990s, “over 80% reported that they wanted to go straight but only 25% thought they would definitely be able to go straight” (p. 395). The predictions that about-to-be-released prisoners made of their likelihood of reoffending were fairly accurate. Those who thought they would go straight and felt confident that they had the ability to do so were less likely to reoffend within two years than were those who were pessimistic about their ability to go straight. This same group of offenders was examined (through official records) ten years after release. Over 80% had committed at least one subsequent crime. However, those who were hopeful of their ability to go straight were much less likely to have been re-imprisoned than were those who were less optimistic about their future.
However, “hope for the future” appears to be relevant only for those people not exposed to too many social problems: “When faced with a large number of exogenous social difficulties… a person’s attitudes and internal motivation may be overwhelmed by reality” (p. 399).
Conclusion. It would appear that simple rational choice theory, as applied to offending, once again is found to be inadequate. These findings suggest that “one important individual factor to consider… is the role of hope among prisoners. A rational choice to avoid imprisonment, verbalised in moments of reflection or in interviews with researchers, is not on its own going to make much difference if the person does not believe they have the ability or the resources to make this change. One needs both the ‘will and the ways’ for this rational choice to be meaningful… The notion that… ‘decent but austere’ prisons can both scare inhabitants straight through sheer deterrence, and also somehow become hotbeds for hope and developing self-efficacy seems a far fetched fantasy…. The subsequent criminal careers of the majority of the sample contradict the easy assumption that a distaste for imprisonment, itself, leads to a lifestyle that avoids repeating the experience” (p. 401).
Reference: Burnett, Ros and Shadd Maruna. (2004) So ‘Prison Works’, Does it? The Criminal Careers of 130 Men Released from Prison under Home Secretary, Michael Howard. The Howard Journal, 43, 390-404.
One of the more popular legal approaches to combating organized crime and terrorism is to require the reporting to government of suspicious transactions and to allow governments to freeze and confiscate assets believed to be associated with crime. Depriving criminals of their illicit earnings is seen as hitting them “at their most vulnerable spot: their assets” (p. 517). The theory seems to be that this will hurt the organization, whereas imprisonment of members of the criminal organization may hurt the individuals, but not the organization.
In 1993, the Netherlands adopted legislation making it possible for prosecutors to seize assets if they could convince a court that involvement in criminal activity had increased an offender’s revenue. Financial investigations were enabled by allowing prosecutors to identify funds obtained by illegal advantage up to two years after conviction. The range of assets that could be seized was increased, and, in addition, procedures for out-of-court deals were created.
The Netherlands Ministry of Justice estimated that the government would gain €200 million (about C$320 million) in three years. This was seen by some as having the potential to encourage abuse of power by the police and prosecutor by putting enormous pressure on them to collect as much money as possible. In fact, during a six-year period (1995-2001), a total of €129 million in confiscation orders were made, but only €27 million were actually collected. “These figures underline the fact that it is extremely difficult to ensure the execution of court orders [just as it has been in the U.K]… The public prosecution department has been very reluctant to take people into custody when they are unwilling to pay. Thus the alleged big stick has become a minor twig” (p. 521).
Based on the assumption that people should not profit from crime, these laws developed without much knowledge of the details of the problem. The laws seemed to be based on the theory that serious “organized” offenders “regard the threat and fact of imprisonment as the cost of doing crime” and are willing to pay this price, but are not willing to lose their profits. There is no evidence that this is correct: “The assumption that by depriving one criminal of his proceeds, the financial situation of an organisation as a whole may be influenced is… not supported by any empirical research” (p. 525). For “ordinary” white collar criminals, it seems that “if necessary, this type of offender is willing to pay huge amounts of money in order to prevent themselves from being sent to prison” (p.525) suggesting that imprisonment is more of a worry than is the loss of money.
Conclusion. The reasons for these disappointing results relate to both “weak policy theory” and “defective law enforcement” (p. 522). The theory that organized crime is likely to be affected by legal threats to assets is not supported by the evidence. Law enforcement agencies are unable or unwilling to spend resources attempting to “follow the money.” Indeed, when one looks at the cases that have contributed to the money that has been collected, “in 84% of the cases, the estimated amount of illegally obtained income does not exceed €45,000 [approximately C$72,000]” (p. 528). Almost half of the cases involve amounts under €4500. The law does not seem to be applied, primarily, to “organized crime” since the amounts often involved are quite small. Confiscation is a low priority for the police and prosecutors.
Reference: Nelen, Hans (2004). Hit Them Where It Hurts Most? The Proceeds-of-Crime Approach in the Netherlands. Crime, Law, and Social Change, 41, 517-534.
One problem with civil commitment laws as an approach to incarcerating those thought to be sexually violent predators is that experts cannot agree, when assessing individual cases, who a dangerous sexual predator is.
Sixteen U.S. states have laws that allow for the civil commitment after incarceration of those thought to be dangerous sexual predators. The U.S. Supreme Court has required that the targets of such laws have (1) a history of criminal sexual behaviour, (2) “a mental abnormality or personality disorder predisposing the individual to sexual violence, and (3) a likelihood of future sexually violent behaviour” (p. 357).
This study examined the most basic statistical requirement of such commitment laws: the likelihood that two “experts” would give the same assessments to a potential “Sexually Violent Predator.” The records of all 295 male convicted sex offenders in Florida who were assessed by at least two professionals under this provision during one year (1 July 2001-30 June 2002) were examined. A total of 25 different experts (licensed psychologists or psychiatrists) assessed at least one prisoner. The focus of the study is simple: did the two professional assessors come to similar conclusions? If they did not, the question of validity becomes irrelevant: Reliability of assessment is a necessary, but not sufficient, condition for predictive validity.
The results were not encouraging. Using an accepted standard for assessing the strength of the agreement between the two assessors, only the risk assessment instruments achieved good reliability. The reliability of the risk measure is not surprising: high risk scores can be obtained on the basis of criminal history or demographic factors alone (factors that are reliably assessed and may predict reoffending in any prisoner but not necessarily sexual reoffending).
Most importantly, reliability was poor for most diagnostic categories. Only “pedophilia” and “other mental illness” had fair reliability. Collapsing the “mental illness” measures into one overall diagnostic measure “to see if evaluators can agree that an individual meets criteria for some paraphilia” (p. 363) resulted, again, in poor reliability. Probably the most important single finding was that “the inter-rater reliability of the civil commitment recommendation proved to be poor.” Given that these laws typically require that the individual “suffer a mental disorder that predisposes the individual to future sexual violence” the unreliability of the diagnosis is a serious problem.
Conclusion. The fact that mental health professionals cannot agree on whether a prisoner has or does not have a mental disorder that predisposes him to commit a sexual offence suggests that civil commitment laws such as this one – whose purpose it is to incarcerate people for crimes they might commit – cannot fulfill their purpose. Reliability of assessment or diagnosis must be achieved before it is worth even considering issues of the validity of the assessments. Given the poor showing on the reliability measures, it is clear that the laws which civilly commit those thought to be sexually violent predators cannot achieve their stated function.
Reference: Levenson, Jill S. (2004). Reliability of Sexually Violent Predator Civil Commitment Criteria in Florida. Law and Human Behaviour, 28, 357-367.
Traditional measures of police performance – based on managerial or technical attempts to find an equivalent to the private sector’s “bottom line” – should be replaced.
The development of performance measures for police is inherently a normative or political process. “One… has to have a theory about whose views of the ‘good’ and the ‘right’ should count in setting a standard for policing” (p. 3). Who does the valuing of the police? Should the focus be largely on instrumental goals (e.g., reducing crime, increasing security) or on fair and impartial enforcement regardless of instrumental results?
This paper suggests that the most appropriate perspective for thinking about police performance measures is that of citizens who “have to decide what kind of police department they would like not knowing whether they would end up being individuals who called the police for assistance, individuals who supported the police by paying taxes, or individuals who were arrested by the police” (p. 4). The goal of “satisfying customers” is not sufficient since identifying customers is complex. The satisfaction of those who call the police is clearly important, but those same people are also taxpayers who may not feel it is appropriate to spend large amounts of public resources responding to minor calls for service. Fair treatment may not be valued by a citizen until such time as that same citizen is the recipient of the treatment. Hence the suggestion is made that the citizen who should be “considering a question about the just and fair use of state resources [is someone who is doing so] without knowing the particular position he or she occupies in society” (p. 7). In this way, citizens are more like owners than customers and “are interested in the overall, aggregate performance of the department” (p. 8) not just the quality of the experience of individuals who call for service. Hence the quality of the experience of all of those who come in contact with the police is important, in part because in the exercise of state authority, any citizen can be obligated by the police to do certain things.
The factors that might be nominated for discussion in this context would then be wide ranging. It would be for the “community” (as defined above) to decide on what weight these, and other, factors might be given:
- Reducing crime. It is generally accepted that there are severe limits on the ability of police to reduce crime. Consequently, giving too much weight to this value is almost certainly unfair. But it is also clear that the police can contribute to crime prevention through means other than arrests (a common performance measure)
- Holding offenders to account.
- Reducing fear and enhancing security.
- Structuring the burden of defending against crime. The wealthy are able, through various means (e.g., the location of their homes), to protect themselves from crime. In a similar fashion, giving the police power to enforce the law can increase the likelihood that those who attack the weak will suffer the same consequences as those who attack the powerful.
- The regulation of public spaces and traffic safety.
- Providing a range of emergency medical and social services.
The process of “developing an ordered sense of the values that would be worth pursing” (p. 14) by a publicly funded police department might include the following:
- Distinguishing utilitarian from principled values. Both are important, though often the utilitarian values are, without thought, given precedence.
- Deciding on the importance of various stakeholders and groups who need to be satisfied. “It is sometimes the duty of a police department to say ‘no’ to [groups of people such as businesses and law abiding citizens]” (p. 15).
- Distinguishing between aggregate values and those values that are realized in individualized transactions. Being generally responsive to the community may be in conflict with giving the attention to a particular citizen.
The paper suggests the following as a starting point:
- What Citizens Should Value:
- Dimensions of Police Performance
- Principled Values
- Social Perspective:
- Enforce Law Faithfully and Impartially
- Call Offenders to Account
- Reduce Corruption
- Reduce Brutality/Excessive use of force
- Fair Allocation of Police Resources
- Fairly Distribute Burden of Protection between public and private
- Enforce Law Faithfully and Impartially
- Individual/group perspective:
- A sense of fair treatment…
- Among those obligated to the police
- Among particularly situated groups
- A sense of fair treatment…
- Social Perspective:
- Instrumental Values
- Social Perspective:
- Enhance Safety & Security
- Reduce Crime & Victimization (through various means, including those not related to arrest)
- Increase Traffic Safety
- Reduce Public Disorder
- Provide emergency medical/social services
- Increase efficiency and cost effectiveness
- Enhance Safety & Security
- Individual/group perspective:
- High quality of service…
- As seen by individuals and groups who call or make requests of the police
- High quality of service…
- Social Perspective:
Measures for each of these can be created. Such a set of values, of course, puts pressure on police to “pay closer attention to the value… [that is derived] from sustaining particular capabilities and specialized product lines” (p. 17). Investments in new technology and programs would need to be capable of demonstrating improvements in highly valued areas.
Conclusion. The suggestion made here is that a combination of principled and instrumental values that can be measured need to be identified. The perspective that needs to be taken is that of members of the community who do not know whether they are or will be simply taxpayers, citizens making requests of the police, or citizens subject to police powers. Such an explicit set of values, associated with police policies and the allocation of resources, would allow the community to evaluate whether “the police have been fair in the ways they allocate their resources and impose the burdens of their investigations…. To see whether a police department is getting better…. one must first have conception of what is valuable, and then find a way to measure it” (p. 18).
Reference: Moore, Mark H. and Anthony A. Braga. (2004) Police Performance Measurement: A Normative Framework. Criminal Justice Ethics, 23 (1), 3-19.