Criminological Highlights Vol. 7, No. 3 – June 2005

View this issue as a PDF: PDF iconCrimHighlightsV7N3.pdf

This issue of Criminological Highlights addresses the following questions:

  1. Why are people opposed to having prisons built in their communities?
  2. Does intensive probation for youths reduce recidivism?
  3. Do privately run prisons have lower recidivism rates than publicly run prisons?
  4. How do state elections in the U.S. contribute to high imprisonment rates?
  5. Is joining a gang a good way for youths to protect themselves from being attacked?
  6. Are three strikes sentencing laws responsible for the drop in crime in U.S. cities?
  7. What factors affect the likelihood that a victim of domestic violence will support the prosecution of the accused?
  8. Who gets chosen to participate in restorative justice programs?

Much of the opposition to having a prison in one’s community is related to concern about problems associated with visitors to the prisons, not inmates.

It is estimated that in about 80% of proposals for new prisons in the United States, there is community opposition (p. 145).  What is less well known, however, is exactly why people oppose having these institutions built in their neighbourhoods.  This study reports on a survey of residents in an area of Pennsylvania in which a new 550-bed facility was to be constructed for juveniles who were convicted and sentenced as adults.  Those surveyed included people in the immediate neighbourhood (within a 2 mile radius of the proposed prison site), those in the same borough/township, and those somewhat further away but in the same county.  

Only about 15% of those surveyed thought that crime would increase as a result of the placement of the prison in their community.   Those living very close to the proposed prison site were more likely than those living further away (but in the same county) to believe that crime would increase as a result of the new prison.  But the largest single predictor of concern about increased crime from the new prison was concern about visitors: those who thought that “people coming into the community to visit inmates would cause problems in the local community” (p. 152) were more likely to think that the new prison would be a cause of increased crime. 

Safety concerns were assessed through questions relating to concerns about inmate escapes, as well as safety of the respondent and the respondent’s family. People from wealthier households had fewer safety concerns related to the new facilities, with those in the immediate vicinity having the highest level of safety concerns.  Finally, the belief that visitors would cause problems was the strongest predictor of safety concerns. 

Conclusion: Not surprisingly, those living closest to the proposed site of the large youth custodial facility showed the highest level of concern about their own safety and increased crime.  The most intriguing finding, however, is that concerns about safety and crime seem to be most highly associated with concerns about problems created by visitors to the prison (e.g., family members).  It would appear that “fear of visitors and the impact they could have on the local community may be the driving force behind the opposition [to having a correctional facility in one’s neighbourhood]” (p. 161).  Clearly, if one were interested in addressing the concerns of those living in close proximity to a new correctional facility, addressing their concerns about visitors to that facility would be a good place to start.

Reference: Martin, Randy and David L. Myers. (2005) Public Response to Prison Siting: Perception of Impact on Crime and Safety.  Criminal Justice and Behaviour, 32(2), 143-171.

Back to top

An intensive juvenile probation program providing many services to young offenders had no impact on recidivism.

It is often assumed that “more” is better in providing services to youth in the community. In particular, since many of the youth who are involved with the (formal) youth justice system tend to have various identifiable social, educational, and psychological “needs,” the idea that efficient delivery of services would reduce future offending is very attractive. The evidence supporting this assumption is, at best, mixed. For example, one program that provided “at risk” but ordinary youths with comprehensive services showed long-term negative impacts (See Criminological Highlights, 5(4), #1). 

This study, in California’s Ventura County, examined local youths (age 12-18) who were apprehended either for an offence or for a violation of probation and who were deemed to be of “moderate” risk to reoffend.  They were randomly assigned either to receive ordinary probation supervision or a form of “intensive” treatment program. The ordinary probation involved minimal intervention: a monthly contact with the youth, referrals to outside agencies, and no particular special services.  

The intensive program involved at least one hour of face-to-face contact with a probation officer per week, at least two family contacts a month, a focus on helping the family, including programs to address various “risk factors” within the family (e.g., alcohol and drug programs, anger management, parenting skills, etc.), as well as restorative programs with the victims and contact with community programs and activities.  It was estimated that the youths receiving “standard” treatment had an average of 6.2 minutes of contact per month with various service providers compared to 6 hours and 11 minutes of contact for the youths in the intensive program.

Youths were followed for 18 months after the end of the intervention period. The proportion of youths in the two groups who apparently offended during the follow-up period was almost identical: 59% for the intensive probation group and 58% for the standard probation group came in contact with the police for a new offence.  Members of two groups were almost equally likely to be taken to court and to end up being incarcerated.  Not surprisingly, given their levels of contact with various service providers (and monitoring by these providers), the youths who received the high level of treatment were more likely to have been found to have used drugs. Although the intensive intervention program was designed for relatively high risk youths, and a high proportion of the youths in the program were apprehended for subsequent offences, it is possible that these were not high enough risk youths to benefit from the high intensity intervention (p.43). 

Conclusion. It is often believed that a youth justice system without a full array of services that are delivered when they are deemed to be needed constitutes an unfulfilled promise of special treatment for youths.  This study, consistent with many others, suggests that although youths in the experimental program “received a more intense program in terms of amount and length of contacts and types of services given,” (p. 42) there were no differences in recidivism and few other differences between the experimental group and the randomly assigned control group which received “standard” probation programming.  Providing “services” does not guarantee that these services will be effective.

Reference: Lane, Jodi, Susan Turner, Terry Fain, and Amber Sehgal (2005). Evaluating an Experimental Intensive Juvenile Probation Program: Supervision and Official Outcomes. Crime and Delinquency, 51(1), 26-52.

Back to top

The private sector is no better than the public sector at running prisons that reduce recidivism. 

When private prisons first came on the scene, it was suggested that they would outperform publicly run prisons on many dimensions.  However, over the years, it has been shown that they are not less expensive (Criminological Highlights, 2(5)#2); they appear to have at least as many escapes and other disruptions (Criminological Highlights, 5(2)#1); and the operation of private prisons raises fundamental questions about their relationship to government (Criminological Highlights, 3(6)#1; 3(6)#2).  This paper examines the hypothesis that privately run prisons will be more effective than publicly run prisons at reducing recidivism.  This hypothesis is based on the assumption that private corporations will have a greater motivation than the public sector to show positive results.

This study compares recidivism rates of privately and publicly run prisons in Florida.  Various comparisons relating to the amount and type of exposure to private prisons were used to take into account different theories concerning the impact of prisons on individuals. Rather than simply comparing recidivism rates of those released from public and private institutions, inmates were grouped in a number of different ways in order to make public-private comparisons. These included, for example, measures of the proportion of time spent in the two types of prisons, and comparisons involving inmates who had spent at least 6 months in one or the other type of prison. The records of inmates were examined for a 5-year period following release.

An initial analysis shows, not surprisingly, that inmates (male adults, female adults, and young males) who served time in publicly run prisons (by various operationalizations of “serving time”) tended to be somewhat different from inmates who served time in “private” prisons on a number of dimensions known to relate to recidivism (e.g., offence, level of custody). These results suggest that simple comparisons of the impact of “public” and “private” prisons that do not control for inmate differences are likely to show effects that have nothing to do with the public/private comparison. 

The results – when pre-existing factors known to relate to recidivism are controlled for – are easy to describe.  For male adult offenders, female adult offenders, and for male young offenders (there were no young females placed in private facilities) there were no effects of the institution type (public vs. private) on recidivism (reoffending or reimprisonment).  

Conclusion:  In this study of the impact of private prisons on recidivism, “Six separate comparisons of the levels and types of exposure to public and private prisons were quantified and analyzed for three inmate types…: adult males, adult females, and youthful male offenders.  In total, 36 distinct comparisons of recidivism rates between public and private prisons were analyzed, each controlling for numerous factors known to affect recidivism rates after prison release.  For adult males, adult females, and youthful male offenders, no statistically significant differences in recidivism rates were found between public and private inmate groups” (p. 78).

Reference: Bales, William D., Laura E. Bedard, Susan T. Quinn, David T. Ensley, and Glen P. Holley.  (2005). Recidivism of Public and Private State Prison Inmates in Florida. Criminology and Public Policy, 4(1), 57-82.

Back to top

In recent years in the United States, Republican strength in state legislatures following closely contested state elections is a good predictor of increased prison admissions.

The rate of imprisonment in the United States is at an all time high – approximately 5-9 times those rates in European Union countries and in Canada.  Most commentators suggest that crime first became a national election issue in the U.S. in 1964, and since then its prominence as an election issue has increased.  This paper examines the cynical hypothesis that the “tough on crime” approach in state election politics translates into higher rates of imprisonment not just when there is Republican strength in state legislatures, but in particular immediately following a closely contested (state) electoral race.  

This study examines the impact of state elections (from 1977 until 1996) on prison admissions, arguing that prison admissions constitute the measure most immediately sensitive to actions of state legislatures.  The measure of electoral competition was an index combining the closeness of the electoral races for legislative seats, the proportion of highly contested seats in the election, and the proportion of seats that were contested at all. 

The findings suggest that the impact of Republican strength in the legislature on prison admissions increased as the level of competition in the preceding election increased. In addition, it was found that for a given level of competition, the impact of Republican legislative strength on prison admissions increased over time. Over the 20 year period examined in this paper, the impact of a Republican dominated legislature on prison admissions was negligible in the 1970s, but strongly associated with increasing prison admissions in the 1990s.  More interesting, however, is the fact that the importance of Republican legislative strength (especially in recent years) increased when there was a close election. 

There were other predictors of the prison admission rates. For example, high welfare rates were negatively associated with prison admissions, supporting the conclusion that “penal and welfare institutions [in the U.S.] have come to form a single policy regime aimed at governance of social marginality” (p. 233).  Above and beyond the political effects, however, there was no impact of unemployment rate or of the racial makeup of the state on the overall rate of prison admissions. 

Conclusion. “When [electoral] district competition is low, increases in Republican strength in the legislature do not have much effect on prison admissions.  By contrast, when [electoral] district competition is high, increases in Republican legislative strength are accompanied by significant increases in prison admissions” (p. 232).  Law-and-order politics resulting in increased prison admissions appear to be most likely to be used in circumstances when right-of-centre politicians believe that they are under electoral threat. 

Reference: Stucky, Thomas D., Karen Heimer, and Joseph B. Lang. (2005). Partisan Politics, Electoral Competition and Imprisonment: An Analysis of States Over Time. Criminology, 43(1), 211-247.

Back to top

Contrary to what many youths believe when they join gangs, gang membership offers no protection from violent victimization.

There is a fair amount of evidence that “being involved in a delinquent lifestyle increases the risk of personal victimization” (p. 794).  Clearly gang membership is associated with increased offending. It would appear that, in general, “gang members may exhibit extensive involvement in delinquency before joining a gang, increase their delinquency even more upon entering the gang, and then decrease their level of delinquency after exiting the gang” (p. 796).   It appears that gang membership is typically rather short lived: One study found that most gang members were in a gang for a year or less. 

This study examined not offending, but violent victimization among youth. Specifically it examined the impact of gang membership on victimization in two samples: a representative group of Grade 8 students and a group of 11-12 year olds who were followed for 5 years after their initial contact with the researchers.  

Youths who indicated that they were members of gangs were considerably more likely to report that they had been victims of both minor and more serious assaults as well as robbery. More interestingly, “consistent with prior findings regarding delinquent behaviour, [the] findings provide evidence of an enhancement effect of gang membership on victimization.  Rates of violent victimization are higher for gang members than for nongang members both before and after their gang membership” (p. 808). Generally victimization peaks when the youth is a member of a gang.

Many youths indicated that one of the reasons they joined a gang was for protection. However, the “results indicate that it appears not to be the case that youths who join [gangs] for protection have higher victimization rates prior to gang joining than youths who do not join for this reason, nor are their victimization rates lower once in the gang compared to youths who join for other reasons” (p. 809). 

Conclusion: Generally speaking, it would appear that the likelihood of a youth being a victim of violence is related both to violent offending by that youth and to gang membership. “Gangs do not provide protection for their members. In fact, youths who join gangs experience greater violent victimization while in their gangs than they do either before they join or after they leave” (p. 813). 

Reference: Peterson, Dana, Terrance J. Taylor, and Finn-Aage Esbensen. (2004). Gang Membership and Violent Victimization. Justice Quarterly, 21(4), 793-815.

Back to top

A comprehensive study of crime in 188 U.S. cities demonstrates that three strikes laws did not lower crime rates.  

Between 1993 and 1996, 25 U.S. states plus the U.S. federal government passed some form of “three strikes” laws – laws in which a person convicted for the third time typically gets a dramatically longer sentence than they would have received if they had no criminal record.  In some states, the second conviction also results in a sentence harsher than the offence itself would warrant.  What counts as a “strike” varies across jurisdictions.  The proponents of this form of sentencing law suggest that through general deterrence and/or incapacitation, these laws will reduce crime. 

The application of “three strikes” laws, however, has been dramatically uneven.  “Strikes” in some locations are not brought to the attention of the court (perhaps as part of plea bargains) and, as a result, offenders often do not receive the enhanced sentence prescribed by the law.  Much of the research has focused on California where three strikes provisions were aggressively administered. But even in California the evidence does not support the hypothesis that these laws reduce crime (See Criminological Highlights, 1(3)#4, 2(6)#3, 5(2)#2, 6(2)#1). Furthermore, there is evidence that they create other problems (See Criminological Highlights, 1(2)#5, 3(3)#5) for the administration of justice. 

This study examines the impact of the state level sentencing laws on crime in 188 U.S. cities with populations of 100,000 or more. Crime rates for homicide, robbery, assault, rape, burglary, larceny, and motor vehicle theft were examined.  A long list of variables known to relate to local crime rates – e.g., percent of households headed by females, percent living below the poverty line, percent Black or Hispanic – were controlled statistically. 

In general, crime in these cities began decreasing before the passing of three strikes laws in 1994 or 1995.  But, before controlling for any other variables, it appeared that crime decreased “slightly faster” (p. 221) in three strikes states. However, when the various control factors are included in the analysis, there was no evidence of a decrease in crime rates attributable to the 3-strikes laws.  There was, however, some evidence that “states adopting three strikes laws were the same ones relying more heavily on incarceration as a crime control strategy during… the 1980s and 1990s” (p. 225).  Only one reliable effect was identified: homicide rates increased by about 10.4% after the passage of three strikes legislation. The passing of three strikes laws did not appear to be consistently related to other crime rates.

Conclusion. When the data are examined on a state-by-state basis, one can find jurisdictions where crime reductions appeared to be associated with the passage of three strikes laws.  However, given that there are increases in the same crimes associated with the passage of similar laws in other states, it would appear to be most likely that the laws are not the cause of either the decrease in crime rate in some locations or the increases in others. “One cannot cherry-pick those states that appear to benefit from the passage of a three strikes law and ignore states where the laws appear to have a deleterious impact on crime” (p. 232). 

Reference: Kovandzic, Thomislav V., John J. Sloan, III, and Lynne M. Vieraitis. (2004). “Striking Out” as Crime Reduction Policy: The Impact of “Three Strikes” Laws on Crime Rates in U.S. Cities. Justice Quarterly, 21(2), 207-239. 

Back to top

In a study of 5,272 domestic violence cases, it was found that the more serious the charge that was laid, the less likely it was that victims would support the prosecution of the person charged with the assault. 

Police and prosecutors, on the one hand, and victims of domestic violence, on the other, often have conflicting goals.  Victims may initially involve the police in domestic violence incidents in order to stop the violence that is occurring, but may not want the offender to be arrested   (See Criminological Highlights, 3(5)#6).  Victim cooperation with the prosecution appears, even in specialized domestic violence courts, to be an important determinant of whether a prosecution goes forward (See Criminological Highlights, 4(4)#8). In the past 25 years or so, mandatory charge and “no drop” policies have become quite popular with police and prosecutors.  At the same time, however, there are data (See Criminological Highlights, 3(2)#7) suggesting that women are quite good at predicting whether they are going to be re-victimized by their partners. Thus it is not surprising that there is increased pressure for criminal justice decision makers to take into account victims’ preferences when deciding whether to prosecute an accused. 

This study investigated the determinants of victim support for prosecution in domestic violence cases in Sacramento County, California, from 1 July 1999 until 31 December 2000. The prosecutors’ files on each case served as the source of information about the nature of the case and the victim’s support for prosecution.  Continuous support for the prosecution was defined as originally wanting a prosecution and not withdrawing support at any point during the criminal process.  In about 29% of the cases it was clear that the victim supported prosecution, and in an equal number of cases it was clear that the victim did not give continuous support. In 43% of the cases, the level of victim support could not be determined.  

Not surprisingly, those who suffered severe attacks or needed medical attention were more likely to give full support to the prosecution, as were those for whom there had been previous incidents of domestic violence.  Similarly, those victims who had previously received “protective orders” from the court were more likely to support the prosecution of the offender.  Victims, especially Black, Hispanic, or other minority races or ethnicities, who were cohabiting with the offender were, however, less likely to support prosecution.  The 12% of victims who were male were also less likely to support prosecution. Prosecutors in many of these domestic violence cases had some choice on how to proceed.  They could file the case as a felony, as a less serious misdemeanour, or in some cases as a violation of probation.   When other factors (e.g., the seriousness of the offence) were statistically controlled, victim support for prosecution was lower when the accused was charged with a felony.  In this jurisdiction, a felony conviction would likely result in an average sentence of 240 days as compared to 30 days for a misdemeanour or a probation violation.  It is possible that potentially harsher sentences for the offender were seen by victims as having excessive “costs” for the victim as well.

Conclusion: Offence characteristics (e.g., whether there was injury), victim characteristics (e.g., race/ethnicity) as well as relationship characteristics (e.g., that the offender and victim were cohabiting) were important determinants of support for prosecution.  However, one variable – the severity of the charge that was laid – had an impact on support for the prosecution above and beyond these other factors.  It would appear that a strong initial response (in the form of a more serious charge) from the criminal justice system may not necessarily bring support for a prosecution in domestic violence cases.

Reference: Kingsnorth, Rodney F. and Randall C. Macintosh.  Domestic Violence: Predictors of Victim Support for Official Action. Justice Quarterly, 21(2), 301-328.

Back to top

The likelihood that a youth will be chosen to participate in a restorative justice program varies with the type of community the youth comes from, the offence that was committed, and characteristics of the youth.

Restorative justice programs for youths who offend are increasingly popular in many parts of the world.  However, because communities are almost invariably involved in the implementation of restorative justice programs, the nature of the community in which a youth resides is likely to be seen as being important in determining whether a young offender will be given an opportunity to participate in a restorative process. “Restorative justice assumes a community responsible for criminal activity, while addressing the harm caused by the offender.  However, community aspects can facilitate or hinder offenders’ reintegration process” (p. 106, emphasis in original).

In this study, youths who were eligible for diversion from an Arizona youth court were identified and followed through the justice process. One group of youths was referred to a restorative justice program in which decisions about what would happen to them were made using a family group conferencing model. Resolutions typically involved some combination of restitution, community service, fine, counselling, etc. A second group of youths was diverted from the court using “standard diversion mechanisms” (p.111). The decision on whether the youth was offered the restorative program or the standard program was made by probation staff and the prosecutor’s office. Generally, diversion from full prosecution was offered to first or second time offenders except those charged with sex offences or serious violent offences. Characteristics of the youth as well as characteristics of the youth’s community were examined.  In considering the selection process, it is important to keep in mind that the choice for these youths was either a restorative justice program or informal supervision by court personnel. 

Youths selected for the restorative justice program tended to be much more likely to have been apprehended for a property offence, and much less likely to have been apprehended for a status offence than youths who were diverted from the court in the ordinary manner.  The restorative justice youths were slightly more likely to be in school and were more likely to be white.  Youths from communities with high unemployment rates were more likely to be chosen for the restorative justice program, perhaps because they are seen as most in need of the reintegration process (p. 120).  Youths from communities with a high level of Spanish speaking households and from racially/ethnically heterogeneous neighbourhoods were less likely to be chosen for the restorative justice program. 

Youths were followed for 24 months after they had successfully completed their program requirements. There was some indication that youths sent to the restorative program were less likely to reoffend than were youths who were sent through the standard program, though it was not clear whether this difference could be attributable to the different treatment that the two groups received or some other difference between the groups.

Conclusion: In this particular program, it would appear that justice officials take into account both individual and community factors in deciding who is “appropriate” for restorative justice programs.  Even within a group of youths who were all being diverted from the formal court system, characteristics of the offender, offence, and the community made a difference.  Access to restorative programs, even when they are available, then, depends to some extent on who you are and where you live.

Reference: Rodriguez, Nancy (2005). Restorative Justice, Communities, and Delinquency: Whom Do We Trust? Criminology and Public Policy, 4(1), 103-130.

Back to top

Publication Type

Volume Number


Issue Number