Criminological Highlights Vol. 11, No. 2 – May 2010

View this issue as a PDF: PDF iconCrimHighlightsV11N2.pdf

This issue of Criminological Highlights addresses the following questions:

  1. Why are young immigrants to Canada less likely to commit crimes than non-immigrant youths?
  2. What accounts for racial differences in arrest rates of youths?
  3. How can police improve the way in which they are seen by victims?
  4. How can guidelines create disparity and uniformity simultaneously?
  5. Why is the selling of small amounts of marijuana in Dutch ‘coffee shops’ allowed?
  6. Can the recidivism rates of sex offenders released from penitentiaries be reduced?
  7. How are U.S. states addressing the problems of high imprisonment levels?
  8. Does going to prison change a person’s “race”?

Youths who immigrated to Canada are less likely to engage in crime than non-immigrants in large part because of their increased commitment to family and education.

In the last quarter of the 20th century, immigrants to the Toronto area came primarily from Asia, Africa, and the Caribbean. Prior to that, immigrants were primarily from Europe or North America.  In Canada, as in other countries, many people apparently believe that immigrants are disproportionately responsible for crime even though research suggests that immigrants generally are less likely to  be involved in crime than their native born counterparts (see Criminological Highlights¸V10N6#7, V8N3#5, V11N1#4).  In 1929, after reviewing the birth place of Canadian prisoners, a well-known Canadian lawyer wrote that “the foreigner is not doing so badly in Canada, and these figures unquestionably persuade us that it is not the foreign born citizen that is responsible for the Crime Problem in Canada… the real problem is inherently Canadian in its origin” (J.C. McRuer, K.C. General Session of the American Prison Association Annual Congress, Toronto, 1929).   This study reports findings on the relationship between immigration and crime at two somewhat later points in time – 1975 and 1999 – among youths in a suburban community of Toronto.

Secondary school students were surveyed in the same schools in 1976 and 1999.  Youths were defined as immigrants if they came to Canada after age 12. The proportion of immigrant youths increased during this period, but more dramatic was the shift in origin of the immigrant youths.  Ethnicity was defined by the father’s origin.  In 1976, 60% were Anglo or North American and 30% were from elsewhere in Europe, leaving only 10% from Africa, the Caribbean, or any part of Asia.  In 1999, the immigrant community had changed dramatically: 24% were Asian, 13% were of African/Caribbean background, and 30% were from Southeast Asia or the Middle East, with only 34% from Anglo or European backgrounds.   Educational commitment was a combination of “engagement” (how important grades were to the student and how much time they spent on homework) and “expectations” (how far they expected to go in school).  Youths reported their use of illegal drugs as well as the frequency with which they committed theft, vandalism, car theft, and assault. 

Controlling for sex and socio-economic status, immigrant youths were less likely to engage in illegal activities than non-immigrant youths.  However, this effect disappeared when educational commitment, parental bonds, or grades in school were controlled for suggesting that the difference between immigrants and non-immigrants lies in the immigrants’ greater commitment to education and higher level of parental bonds.  None of the geographically-defined ethnic groups was more likely to be involved in delinquencies than non-immigrant groups. Furthermore, each of these ethnicity-specific effects appeared to be mediated by educational commitment or parental bonds. 

Conclusion:  “With investment in education comes a sense of commitment and a resulting stake in conformity [to societal norms] that makes these [immigrant] youths averse to [involvement in crime]” (p. 361).  The more recent immigrants – predominantly non-European - appear to have stronger family bonds and higher levels of commitment to education than non-immigrants.  The findings underline the importance of the conclusion that “immigration ought not be treated as causally determinative [of lower levels of crime] in isolation, but that it must instead be contextualized within the experiences, attachments and practices developed within families and schools” (p. 362). 

Reference: Dinovitzer, Ronit, John Hagan, and Ron Levi (2009). Immigration and Youthful Illegalities in a Global Edge City.  Social Forces 88(1), 337-372.

Back to top

For violent and theft-related offences, but not drug offences, White-Black differences in male arrest rates are accounted for by differences in exposure to risk factors years earlier.

Racial discrepancies in the likelihood of being arrested have been found in many studies. Finding differences, however, does not explain why those differences exist.   This paper examines whether exposure to early “risk” factors – measured when male children were in Grade 2 – predicted juvenile arrests between 10-17 years old.

Youths, parents and teachers in Pittsburgh, Pennsylvania, provided information about the youths and their circumstances when the boys were about 8 years old.  These measures included measures of the youths’ conduct and psychological well-being, as well as measures of social disadvantage (e.g., neighbourhood problems, living in poverty). Contacts with the youth justice system were obtained from the local courts as well as the state court system.

As in other studies, Black youths were more likely to be arrested for violent, property, and drug offences than were White youths. (There were too few other-race youths to be included in the sample.)  Arrest for a violent offence was generally related to all of the ‘risk factors’ experienced by the boy.   Theft-related arrests were less strongly related to the risk factors than were arrests for violent offences, and drug arrests were only marginally related to the risk factors. 

However, when the various risk factors were taken into account, race was no longer a predictor of violence-related arrests.  Factors such as low academic achievement, communication problems, peer delinquency, and neighbourhood problems accounted for the relationship between race and violence-related arrests.  Similarly, race did not predict theft-related arrests, once the various ‘risk’ factors were taken into account.  However, race still predicted drug arrests, even when ‘risk’ factors were taken into account.  

There are understandable reasons for the Black-White differences in drug arrest rates. “Black youths are generally more likely to obtain and use [illegal] substances in more public places” (p. 924).  In addition, police officers may be more likely to stop and search Black males. It is also possible that Black youths are more likely to be exposed to drug use in their homes and families. 

Conclusion:  It would appear that if a community is concerned about high rates of violent offending by Black male adolescents, there are things that can be done.  Interventions that target problems that are evident when boys are 8 years old would be likely to reduce arrests for violent offences, and to a lesser extent, theft-related offences.  For example, early targeting  of conflicted parent-child interactions and delinquent peers might well be effective ways of addressing arrests for violence later in life.  

Reference: Fite, Paula J., Porche’ Wynn, and Dustin A. Pardini (2009).  Explaining Discrepancies in Arrest Rates Between Black and White Male Juveniles.  Journal of Consulting and Clinical Psychology, 77(5), 916-927. 

Back to top

The police have direct control over how favourably they are seen by crime victims.  Although victims generally think less favourably about the police than non-victims, the police can mitigate this effect by taking victims’ concerns seriously.  

It has been suggested that there are at least three somewhat distinct components of the community’s evaluation of the police: effectiveness in dealing with crime, fairness or integrity of the police, and police engagement with the community.  Using measures of each of these somewhat separate components of the public’s view of police, this paper examines the impact of different types of police-citizen contact on each of these constructs in a sample in London, England.  

One of the most common reasons for citizen-initiated contact with the police is that the citizen was a victim of crime.  The most important single determinant of citizens’ assessment of the quality of the contact with the police was whether the police appeared to take the citizen’s concerns seriously. Two other factors predicted citizen satisfaction with the specific contact they had with the police: whether the citizen believed that the police followed up on the call and whether the citizen thought that the time he or she had to wait for the police was reasonable. 

Both citizen- and police-initiated contact with the police were related to lower ratings of police effectiveness, even when the citizen was, overall, satisfied with the quality of the particular encounter.  Not surprisingly, people who had unsatisfactory recent contacts with the police were more likely to rate the police, generally, as being unfair and not involved with the community.   But victims’ contacts with police that were seen as favourable did have positive impacts on ratings of fairness and engagement of the police (compared to people who had not had recent contact with the police). 

Perhaps the most important findings are those that suggest that individual police officers can enhance the overall ratings of the police.  When crime victims believe that their concerns are being taken seriously by the police, they see police as not only being more engaged in the community, but also as more fair and effective.   When the police follow up in any way with the crime victim, ratings of effectiveness and community engagement are higher.  

Conclusion:  The data suggest that individual officers can either enhance or damage perceptions that the public holds of the police.  “While opinions about police effectiveness may be challenged by any contact – whether it is satisfactory or unsatisfactory - ideas about fairness and community engagement appear to be amenable to change in either a positive or a negative direction” (p. 41).  “Fairness and community engagement … are the aspects of overall confidence [in the police] that are most related to personal treatment during the [police-citizen] encounter” (p. 42).  Effectiveness in dealing with crime, on the other hand, is largely out of the control of the individual officer who interacts with the public, although police officers who communicate that the citizen’s victimization is being taken seriously can have a positive impact even on this dimension of effectiveness. 

Reference: Bradford, Ben, Jonathan Jackson, and Elizabeth A. Stanko (2009). Contact and Confidence: Revisiting the Impact of Public Encounters with the Police.  Policing & Society, 19 (1), 20-46. 

Back to top

Uniformity of sentences is not an appropriate goal for the sentencing process.

Excessive structuring of sentences can occur in various ways.  Legislatures can mandate fixed or mandatory minimum sentences for certain offences. Alternatively, rigid guidelines can be constructed that do not take into account the complexity of the behaviour that needs to be considered in determining a proportional sentence.   Sentences, therefore, can be excessively similar.  Guidelines are often justified as mechanisms designed to reduce unwarranted variation in sentences.  In doing so, they may create unwarranted uniformity. 

This paper examines drug sentencing under the U.S. federal guidelines.  Those guidelines were made somewhat more restrictive by the fact that the U.S. Congress imposed a number of mandatory minimum sentences on drug offences at the time the guidelines were developed.  The guidelines used the mandatory minimum as the starting point.  In addition to the criminal record of the accused, the main basis for sentences harsher than the minimum is the quantity of the drug that is the focus of the offence.  This means that the role the offender played in the drug process is given no explicit importance in determining the offence.  Different drugs are made “equivalent” by conversions into “marijuana-equivalent” amounts of each drug.  Furthermore, sentences can not be reduced substantially by mitigating factors, and, often, it is only the more important people in the overall drug-selling process who have the information that allows for a lesser sentence (on the basis that they had relevant information to provide to prosecutors in return for more lenient treatment).  

This study examines sentences of 1259 inmates in U.S. federal correctional facilities who were imprisoned for drug offences (most commonly powder and crack cocaine, marijuana, and methamphetamine).  Drug quantity was the strongest predictor of the sentence, as would be expected from the guidelines.  Notwithstanding the drug ‘equivalence’ calculations included in the guidelines, offenders serving sentences for marijuana offences received somewhat shorter sentences.  Those who went to trial got harsher sentences, especially those found guilty by a jury.  In addition, even after controlling for criminal record, drug quantity, drug type, and other legal factors, Black offenders received harsher sentences, as did males, older offenders, and those with lower levels of education. 

The most important finding, however, was that the offender’s role in the offence – simple possession, street selling, wholesaling, producing, importing, or laundering money related to drugs, had no impact on sentences. Said differently, an offender who was merely in possession of a certain amount of drugs for his or her own use was treated as equivalent to someone with a more central role in the drug trade who happened to be sentenced for the same amount of drugs.  The quantity of drugs that was the subject of the sentence was, interestingly enough, independent of the person’s role in the drug trade. 

Conclusion:  The study demonstrates what can happen to sentencing when rules become excessively rigid.  In this case “quantity-driven sentencing, coupled with culpability-based adjustments that are too limited in scope, leads to excessively uniform sentences for offenders of widely differing culpability and responsibility for the drug trade” (p. 172).  This fact, combined with the finding that other legally irrelevant factors such as race still affect sentence length, suggest that there are serious problems that need addressing in these guidelines and, perhaps, in other rigid guideline systems. Alternatively, of course, it could be that these were the effects desired by those who designed the guidelines. 

Reference: Sevigny, Eric L. (2009). Excessive Uniformity in Federal Drug Sentencing. Journal of Quantitative Criminology, 25, 155-180.  

Back to top

The selling of marijuana in Dutch “coffee shops” reflects a local, but somewhat unstable, official tolerance that allows a focus on “safety and security instead of moral differences” (Buruma, p. 108). 

The selling of small amounts of marijuana in “coffee shops” has made the Netherlands a popular destination for certain tourists.  Though the law on the books does not allow possession or sales of marijuana, the sale and consumption of small amounts is tolerated in these shops as part of an overall orientation that focuses more on pragmatic than on moral concerns. The issue of how the selling of marijuana can simultaneously be illegal and officially tolerated in certain circumstances raises interesting questions about the meaning of law and the enforcement of it.

In 1976, Dutch law first differentiated ‘hard’ from ‘soft’ drugs.  This distinction was based, in part, on the policy objective of ‘harm reduction.’  Users of marijuana were not seen as being at risk of social exclusion nor were they seen as being a particular threat to anyone.  The focus was on regulation rather than prohibition. Since the 1980s, marijuana was sold in “coffee shops” in certain parts of certain cities.  There were, however, important restrictions.  Only small amounts could be sold and at least theoretically only relatively small amounts could be kept in these locations.  The view was that “the criminal justice system should be used only against criminal entrepreneurs” (Buruma, p. 90). 

This is not to say, however, that the selling and consuming of marijuana is generally tolerated in all parts of the Netherlands.  It is estimated that coffee shops are not tolerated in 78% of Dutch municipalities. Regulation appears to be local and, to some extent, regional, even though the laws governing drugs are national.   It is understood that the existing coffee shops will be left alone if they don’t advertise, don’t sell hard drugs, have only small quantities of marijuana on the premises, don’t sell to minors, and generally aren’t seen as a problem.  Nevertheless, it is estimated that between the early 1990s and 2007, the number of coffee shops dropped from about 1500 to about 700.  

Estimates of the size of the ‘drug problem’ would suggest that the Netherlands has less of a drug problem than other European countries and “The Netherlands does not have significantly higher cannabis use than other European countries….The estimated prevalence of problem users of hard drugs… is the lowest per thousand inhabitants in western Europe” (Buruma, p. 91). 

Though there may well be changes in the number of coffee shops across the Netherlands in the next few years, the more complex issue that their existence raises has to do with European drug policy.  Coffee shops are seen as an obstacle to future European convergence in this policy area.  Even now, the border towns create very different problems from those in the large cities. But in addition, local policies on enforcement are changing.  Some of the large cities, for example, are planning on closing down coffee shops near schools. 

Conclusion: “To speak of Dutch coffee shop policy is increasingly a misnomer: policy is regional and locally coloured” (Pakes, 256) even though the law governing drugs is national. Coffee shop policy (or at least practice) is affected by international (largely European) concerns, national policies and traditions of tolerance, and local concerns. 

References: Buruma, Ybo (2007). Dutch Tolerance: On Drugs, Prostitution, and Euthanasia. In Tonry, Michael and Catrien Bijleveld. Crime and Justice in the Netherlands. Crime and Justice: A Review of Research. Volume 35. University of Chicago Press.  Pakes, Francis (2009). Globalisation and the Governance of Dutch Coffee Shops.  European Journal of Crime, Criminal Law, and Criminal Justice, 17, 243-257. 

Back to top

A multi-site replication of a high intensity community program reduces recidivism for sex offenders after release from penitentiaries.  

Sex offenders are frequently said to have high rates of recidivism, especially for sexual offences.  The major problem with that belief is that it is not supported by available data  (see Criminological Highlights, 9(2)#5, 8(3)#8, 6(6)#8, 6(3)#3, 5(1)#4, 3(3)#3), 9(3)#6).  Furthermore, even though many people appear to believe otherwise, certain kinds of treatment programs for sex offenders appear to be effective (Criminological Highlights 9(5)#7).  One such program is “Circles of Support and Accountability” [COSA] (Criminological Highlights 9(3)#6). This paper reports on a 7-location replication of an evaluation of this program.  

The COSA program is very intensive. A group of 4-6 ordinary citizens (who have received special training) volunteers to work with a ‘core member’ – a sex offender released from a penitentiary at the end of his sentence.  During the first 2-3 months after an offender is released from penitentiary, the offender meets with at least one volunteer circle member every day. Other circle members meet individually with the offender at least once per week.  In addition, they meet as a group at least once per week. “A COSA is a relationship scheme based on friendship and accountability for behaviour” (p. 415).  Circles continue to meet regularly with their “core member” for months or, in some cases, years after his release from penitentiary.

This paper examines COSAs that were organized in seven Canadian locations in 6 provinces.  Forty-four offenders who volunteered to participate in COSAs were matched with other sex offenders who were also released at the end of their sentences (i.e., with no legally mandated supervision). Matching criteria included various recidivism-predicting measures, measures of sexual deviance, age, participation in programming in prison, and the date and location of the release.  Obviously this is not a perfect comparison group in that the comparison group members were not given an opportunity to volunteer to participate in COSAs.  However, their backgrounds, experiences, and treatment in penitentiary, etc., appeared to be relatively comparable to the COSA members. 

During the 3-year follow-up period, significantly fewer (5) of the 44 (11%) COSA offenders re-offended than in the comparison group (17 of the 44 or 39%).  Looking at violent (including sexual offences), a significantly higher proportion of the comparison group (34%) reoffended compared to the COSA treatment group (9%).  Very few offenders in either group committed sexual offences (1,  or 2%, in the COSA group and 6, or 14%, in the comparison group) a difference that was not significant at the 5% level. 

It is not known exactly why COSAs are effective, but it is likely that it relates to the social support and the positive social influences of the group on the offender. The success may also be a result of the help circle members give with fundamental problems such as housing and employment.  But in addition, “with its concurrent focus on accountability on the part of the offender, it targets issues related to distorted cognitions that support offending and minimize risk….” (p. 426). 

Conclusion. Circles of Support and Accountability, an intensive program for sex offenders released from penitentiary, appear to reduce subsequent offending. As is clear from numerous other studies, sex offenders can be treated and their reoffending rates – which initially were not very different from re-offending rates of other offenders – can be reduced significantly. 

Reference:  Wilson, Robin J., Franca Cortoni, and Andrew J. McWhinnie (2009) Circles of Support and Accountability: A Canadian National Replication of Outcome Findings.  Sexual Abuse: A Journal of Research and Treament, 21 (4), 412-430.

Back to top

Nineteen states in the U.S. have initiated policies designed to reduce imprisonment levels. Twenty states reduced their rates of imprisonment in 2008. 

Even before the beginning of the financial crisis in 2008, “Legislators in many states had become increasingly interested in adopting evidence-based policies directed at producing more effective public safety outcomes” (p.1). Motivation for these changes varied, but in many instances the change in orientation was driven by the high costs and little benefit of high imprisonment policies. The changes in policy focused largely on strategies to reduce the use of imprisonment and to reduce the time people sentenced to prison actually spend there. For the most part, states undertook a range of different strategies to use public safety resources more effectively.  In some cases, the reductions in imprisonment rates were dramatic. In the 10 years ending in 2009 there was a 20% reduction in the imprisonment rate in New York  and  a 19% reduction in New Jersey.  In Michigan, there was a 19% reduction in imprisonment in the 3 years ending in 2009. The following are examples of these how these changes were accomplished.

Sentencing reforms.  Some mandatory minimum sentences were eliminated (e.g., changes to the Rockefeller Drug Laws in New York in 2003). With apparent support of the public, minor drug offenders previously sent to prison in NY were diverted to community based drug rehabilitation programs. Offenders were sometimes offered shorter ‘boot camp’ programs as an alternative to the long prison sentences already imposed on them. A substantial portion of the drop in the number of drug offenders admitted to prison apparently came as a result of prosecutorial discretion rather than changes in the law.  In Michigan, the abolition of mandatory minimums occurred with the support not only of judges but prosecutors.  In 1998, 20% of Michigan’s drug offenders were sent to prison. By 2008, this had dropped to 11%.  Certain aspects of the drug laws in New Jersey were also made more flexible. 

The development of early release programs. In New York, many prisoners were allowed to be released after serving less than their minimum term of imprisonment.  In Michigan, release procedures were improved.  Those seeking early release had been required, since 1992, to prove that they were not risks.  Given that risk was assessed largely on the basis of offence and offence history – factors that are not amenable to change by the offender while in prison – it is not surprising that few were released. By shifting responsibility to local teams who were required to address the dynamic needs of those released, many more people were granted parole (19,000 in 2000 and 25,000 in 2008). Beginning in 2009, the state began re-considering, largely on the basis of risk assessment scores, some of those previously refused parole.  In New Jersey, when risk assessment instruments were introduced, staff were told that when something ‘bad’ happened (e.g., a serious offence by someone who was released), they would not be blamed if they had acted reasonably with the information that was available to them.  

Reducing the use of revocations. Many of those who committed technical violations of their release conditions now are able to remain in the community.  In Kansas, for example, county agencies that supervised offenders were told that funds would be made available to their agency if parole violations were reduced by at least 20%.

Conclusion.  Prison populations can be controlled in large part by ensuring that sentences can be imposed that reflect the actual seriousness of the offence and by ensuring that evidence-based release and revocation decisions are made.  The elimination of mandatory minimums – especially for less serious but high volume drug offences – appears to be an important lever that can be used to reduce prison populations. However, states with relatively high incarceration rates have been successful in reducing prison populations in part because of the large number of people who, from almost any perspective, did not need to be in prison in the first place. 

Reference: Greene, Judith, and Marc Mauer (2010). Downscaling Prisons: Lessons from Four States.  The Sentencing Project ( 

Back to top

For some Americans, the experience of going to prison increases the likelihood both that they will identify themselves as Black and that others will see them as Black.

When O.J. Simpson was charged in 1994 with murder, Time magazine ran a doctored ‘mug shot’ of him on its cover in which his skin colour had been purposely darkened.  With Blacks in America vastly over-represented in prison, Time could be seen as having provided additional support for the association between crime and race. This paper examines the hypothesis that “incarceration affects both how respondents identify themselves and how they are perceived by others” (p. 92). Specifically, it examines whether those who had been to prison are more likely to identify themselves as black, and if others are more likely to see them as black.

This study took advantage of the manner in which ‘race’ data were collected in the U.S. National Longitudinal Survey of Youth and the fact that many respondents, during the period of the survey, were incarcerated at one or more points between their first interview in 1979 and 2002.  Respondents in the survey identified their own race in 1979 and then again in 2002 (when the form of the question changed as a result of federal standards for collecting information on race/ethnicity).   Interviewers, however, were instructed to classify respondent’s race at the end of each annual interview.  The independent variable – experience of being imprisoned – was captured according to whether the respondent was interviewed while in prison.  

Of those who described themselves as of European origin in 1979, 95% described themselves as white in 2002 if they were interviewed out of prison at that time.  However, if they were interviewed in prison in 2002, only 81% identified themselves as white.  Of those who identified themselves as non-European in origin in 1979, 74% of those interviewed in prison said that they were black in 2002.  In contrast, 59% of those who identified themselves as of non-European origin in 1979 but were not in prison in 2002 identified themselves as black.  Simply put, “incarceration leads to an increase in the likelihood of self-identifying as black and a decrease in the likelihood of self-identifying as white.”  

Similar effects were found for the interviewer’s assessment of the race of the respondent.  Looking first at those who were identified as white in the previous year, if the respondent was not interviewed in prison, 96% were classified by the interviewer as white.  If the interview took place in prison, however, only 90% were classified as white.   Interestingly enough, the effect was not due solely to the location of the interview.  Looking only at those interviewed outside of prison, of those who were classified as white the previous year, 96% of those who had never been incarcerated were classified by the interviewer as white.  However, of those who, at some point in the past had been incarcerated, only 92% were classified as white.

Conclusion.  It is not clear what the mechanism is for either of these findings.  It may be that with certain prison systems having large numbers of black inmates, ‘being black’ is, in fact, a more comfortable identity for reasons such as personal safety. Inmates who might ‘pass’ as either white or black choose black.  The effect of incarceration on classification by the interviewer could be a result of changes in the respondent (e.g., in appearance or presentation by the respondent) or it could be that interviewers, themselves, associate race with imprisonment. It would appear, therefore, that racial identification and classification are not static, but are “continually negotiated in everyday interactions” (p. 109). 

Reference: Saperstein, Aliya and Andrew M. Penner (2010).  The Race of a Criminal Record: How Incarceration Colors Racial Perceptions.  Social Problems, 57(1), 92-113. 

Back to top

Publication Type

Volume Number


Issue Number