Criminological Highlights Vol. 9, No. 2 – December 2007

View this issue as a PDF: PDF iconCrimHighlightsV9N2.pdf

This issue of Criminological Highlights addresses the following questions:

  1. Do victim impact statements serve any important functions?
  2. What are the components of an effective ‘after care’ program for offenders?
  3. Can a community make itself safer by ensuring that more offenders are imprisoned?
  4. Can we assume that sensible sounding rehabilitation programs can do no harm?
  5. Do juvenile sex offenders grow up to be adult sex offenders?
  6. Why do ordinary people favour laws that allow for the civil commitment of sex offenders?
  7. What are the effects of ‘community notification’ schemes on sex offenders?
  8. How is the effect on violent crime of the racial makeup of a community affected by employment factors?

Although the majority of victims of crime may prefer not to file victim impact statements, such statements may be of value to those who decide to give their views of their own victimization. 

In the past few decades, the role of victims in the criminal process has become politicized. The suggestion is often made that ‘giving to victims’ means ‘taking from offenders’ –  as if being respectful and responsive to one group necessitated being rude and unfair to another.  The British government, for example, announced recently that it was going to “rebalance the criminal justice system in favour of the victim and the law abiding majority” (p. 361) implying that something needed to be taken from the accused in order to ‘give’ to the victim.  

In contrast to this approach, in 2000, the Scottish executive focused its concern about victims on “better access to information for victims, improved support and greater participation in the criminal process” (p. 361).  On a trial basis (in some locations), victims of all offences against the person (including all assaults) and theft by housebreaking were offered the opportunity to make a statement about the impact of the crime, not an opinion about what the outcome (e.g., the sentence) should be.  This paper summarizes the findings from a study of the Scottish experience based on interviews with victims after the case was over. 

  • Of the victims who were sent the materials necessary for filing a victim statement, only a minority of the victims (15%) took the opportunity to do so.  
  • Of those who did make a statement, the most commonly stated reason for doing so (some gave multiple reasons) was that they wanted to express their feelings about the offence (34%).  23% indicated that they wanted, in general, to influence the outcome.  But only 5% said that they wanted to ensure a conviction; and just 5% indicated that they wanted to influence the sentence. 10% said that they made a statement for “therapeutic reasons” and 14% wanted to make the accused think about the effect of the offence. 
  • The main reason for not making a victim impact statement was that the crime was not serious (53%).  8% didn’t file a statement out of fear. 
  • At the end of the case, 86% of those who filed statements thought that they had made the right decision to do so.  Only 7% indicated that they wouldn’t make a victim statement in the future. 
  • 61% of the statement makers indicated that making a statement made them feel better, though 38% indicated that it made them feel worse.   The more serious the offence, the more likely it was that victims found making the statement to be upsetting. 
  • Most of the actual statements followed the program’s instructions to focus on how the crime affected the victim, though 28% made one or more comments about the accused. 
  • There were very few instances in which information in the statement was challenged by the defence. 

Conclusion:  It appears that victim impact statements can be integrated into the justice system in such a way that most victims who file statements see a value in doing so. Costs were relatively low because the statement information was mailed to the victim and returned by mail. Criminal justice professionals did not perceive unfairness to the accused as the result. “There is no evidence from the Scottish evaluation that the truth seeking function of the criminal justice system was compromised by victim statements” (p. 379).  Although relatively few victims may take advantage of the opportunity to have their experiences described to the court, those who do appear to appreciate the opportunity. 

Reference: Chalmers, James, Peter Duff, and Fiona Leverick (2007). Victim Impact Statements: Can Work, Do Work (For Those Who Bother to Make Them). Criminal Law Review, 360-379.

Back to top

Boot camps might not work, but studies of programs that take place after inmates finish boot camp suggest that the aftercare component of these programs can reduce subsequent offending.

Most offenders who enter prisons return to their communities.  In countries with high incarceration rates, this simple fact raises an important issue: what can be done with ex-prisoners when they return to the community to help them desist from crime? Ex-prisoners face a number of special obstacles to abstaining from crime. Most ‘after-care’ programs for ex-prisoners in the U.S. emphasize surveillance and public protection rather than service for the offender. 

This paper examines a post-release ‘re-entry’ program for inmates who had served 6 month sentences in a Pennsylvania boot camp.  The program involved 90 days of residential aftercare which included cognitive behaviour therapy, job readiness and job acquisition skills, and substance abuse counselling. In most instances, individual treatment plans were developed and executed.  In March 2002, a new policy mandated that all of those released from the boot camp would receive this program.  Those who went through the program immediately prior to March 2002 served as a comparison group (n=383) and were compared to those who had the new treatment immediately after the change in policy (n=337). The two groups did not differ on any major demographic variables (e.g., age, education level, offence, prior arrests).  There were some minor differences between the two groups on some attitude measures (taken as they completed their term of incarceration).  These differences were controlled statistically in the recidivism analyses. 

Aftercare services were provided by 23 different accredited providers, and program content varied somewhat across providers. Nevertheless, there were no differences in recidivism rates across program providers; hence results cannot be attributed to special characteristics of one provider. Recidivism was measured by arrest for a new crime within two years.  At the end of a two-year period following release from the boot camp, 33% of the untreated control group had been arrested in comparison with only 22% of the treatment group. It could be argued that those in the aftercare program were not as much at risk during the program because they were in a rather structured environment for 90 days of the 2-year follow-up.  The analysis was repeated, therefore, using release from the boot camp as the starting point for the control group and release from the aftercare program for the ‘aftercare’ group.   The results were essentially the same. 

Conclusion:  Structured aftercare, focusing on the needs of offenders on release, clearly had an impact on subsequent offending.  Although these (effective) programs differed from one another, “they all were accredited programs, all but one used individual treatment plans and all provided a vast array of services from employment and social skills to drug and alcohol rehabilitation and counselling.  Therefore, this finding does not imply that simply extending the period of supervision for ex-inmates without providing extended services should reduce recidivism” (p. 380).

Reference: Kurlychek, Megan and Cynthia Kempinen (2006).  Beyond Boot Camp: The Impact of Aftercare on Offender Re-entry.  Criminology and Public Policy, 5 (2), 363-388.

Back to top

Counties in Florida that increased the size of their prison populations received no benefit in terms of a reduction in local crime rates. 

Studies which try to estimate the impact of imprisonment on crime rates have used different methods and have arrived at somewhat different findings. A major problem these studies face is estimating the annual number of crimes an offender would commit if he or she remained on the street.  Another problem is that imprisonment policies in neighbouring jurisdictions could have an impact on a jurisdiction’s crime rate if offenders move around a lot.  Data on arrests, however, suggest that few offenders move very far from home when they offend. 

This study looks at variation in incarceration rates across counties in Florida. Changes in crime rates and in imprisonment rates in Florida are similar to those in the U.S. as a whole.  The ‘imprisonment’ measure was the number of people sentenced to prison for a year or more in the county (for crimes committed in that county).  County crime rates were measured by the seven ‘index crimes’ (homicide, rape, robbery, assault, burglary, larceny, auto theft) each of which was analyzed separately. Various factors were controlled statistically: the age distribution of the county, % Black, poverty rate, per capita income, % unemployed, % divorced, and % female-headed households with children. 

“The basic finding… is that county-level prison population growth seems to have little or no significant relationship with county-level crime rates, at least not in Florida” (p. 227).  Although none of the effects of imprisonment rates on crime are significant, they are all in the predicted direction (more imprisonment, less crime).  “Most crime reduction occurs for property crimes” (p. 229), but even then the effect is rather small – “slightly over one index crime per year per additional prisoner” (p. 229) – and as noted not statistically significant. 

The study considered the possibility that the effect of imprisonment on crime within counties is not strong in part because it does not take into account imprisonment in nearby counties.  An analysis including imprisonment rates in nearby counties suggests that “nearby prison population has no impact on in-county crime” (p. 234). 

Conclusion: “This study finds no support for the ‘more prisoners, less crime’ thesis” (p. 234).  One possible explanation for this could be that “As the prison population [of a community] expands, its potential impact on crime may decrease as lower-rate offenders are included in the expansion.  Thus, incarcerating serious high-rate offenders may reduce crime, but expanding incarceration to include less-serious lower-rate offenders will produce small reductions in crime” (p. 235).  

Reference: Kovandzic, Tomislav V. and Lynne M. Vieraitis (2006).  The Effect of County-Level Prison Population Growth on Crime Rates.  Criminology and Public Policy, 5(2), 213-244.

Back to top

Even sensible sounding programs taking place in prisons can be harmful: An innovative, short-term, prison-based re-entry program increased recidivism.

Given that most prisoners eventually are released, it makes intuitive sense to focus at least part of prison treatment program resources on helping offenders re-enter society. On the other hand, it is not clear what the essential elements of a prison-based re-entry program should be.  

Generally, however, there seems to be agreement that effective prison treatments would have the following characteristics: they should address dynamic risk factors (i.e., factors in the prisoners’ lives that can be changed); they should employ cognitive-behavioural, skills oriented, or multimodal treatment approaches; they should focus on the needs of the participants; and the implementation should be well supported and professionally administered. 

Project Greenlight was designed to incorporate various programs while the offender was still in prison and to provide links to families, service providers, and parole officers after release.  As with many ‘modern’ correctional programs, there was a cognitive skills component. “A substantial portion of the intervention was focused on addressing key measures such as employment, housing, drug relapse prevention….  Inmates also participated in sessions that focused on practical living skills…[such as] how to use public transportation, budgeting,… where to get emergency cash or non-cash assistance when money is scarce” (p. 308). 

Eligible inmates were identified 75-105 days prior to release and were transferred to a particular facility in which the program was administered.  Two comparison groups were identified: those who met the same criteria but were not transferred to the facility (because of inadequate space), and those transferred but for whom space in the ‘Greenlight’ program did not exist.  The three groups did not differ on demographic or criminal justice history measures. 

Inmates were followed for two years after release.  Whether one looks at total arrests, felony arrests, or revocations of release for any reason, the results are quite clear: the Greenlight program was associated with an increased likelihood of recidivism, not decreased recidivism as had been hoped and expected. 

Conclusion:  Obviously it is surprising that a program designed to ease offenders back into the community would be associated with increased recidivism.  Why this is the case is not clear.  Explanations such as the program sessions having more than ideal numbers in them might explain effectiveness, but do less well explaining the harmful effects.  It is possible, however, that the actual implementation of the program was poorly carried out. In addition, there was no transitional programming to create a firm plan for post-release services.  Clearly there are lessons to be learned: most obviously it is important for program providers to remember that no matter how good the intentions are, there is a realistic possibility that programs can be harmful to offenders (see Criminological Highlights, V5N4#1, V6N2#4). As one researcher suggested, “Social programs deserve to be treated as serious attempts at intervention, with possibly toxic effects, so that a science of intervention can prosper” (p. 331).  

Reference: Wilson, James A. and Robert C. Davis (2006).  Good Intentions Meet Hard Realities: An Evaluation of the Project Greenlight Reentry Program. Criminology and Public Policy, 5 (2), 303-338.  See also “Smoothing the Path from Prison to Home: A Roundtable Discussion of the Lessons of Project Greenlight.”

Back to top

Youths who commit sexual offences are not very likely to commit sex offences as young adults. 

Special laws for sex offenders – e.g., registration, notification, special ‘peace bonds’ – are based in part on the assumption that once a person commits a sexual offence, that person will continue to commit such offences.  As these policies are extended to include youths who commit sex offences, it is important to consider whether there is any special reason to target juvenile sex offenders.

Adult sex offenders are not especially likely to commit further sexual offences and do not look like the ‘specialist’ offenders they are sometimes believed to be (see, e.g., Criminological Highlights V3N3#3, V5N1#4, V6N3#3, V6N6#8, V8N3#3).  Nevertheless, they are often subject to special conditions after serving their sentences.  There seems to be an assumption that certain people are ‘born’ to be sex offenders and will not change. 

This study looks at the criminal careers of males who were born in 1942, 1949, and 1955 in Racine, Wisconsin. Their involvement in the criminal justice system was tracked until they were 32, 25, and 22 years old, respectively. The first part of the study looked at these cohorts when they were juveniles. Compared to youths whose list of offences did not include sex offences, boys who committed sex offences in Racine were much more likely to have large numbers of police ‘contacts’  (for various kinds of offences).   However, the vast majority of these contacts were not for sex offences.  Forty-three percent of the youths with a juvenile sex offence record had 9 or more contacts with the police and an additional 23% of juvenile sex offenders had 4-8 contacts with the police. However, more than three quarters of the youths with sex offence contacts only had one such contact.  Juvenile sex offenders, it seems, are high rate offenders who commit various offences. For the most part, however, they commit few sex offences (typically only one). 

The second part of the study tracked these cohorts into adulthood. Only 8.5% of the boys who had a juvenile sex offence record had any contact with the police, as adults, for sex offences. This was not significantly higher than the adult sex offence rate (6.2%) for boys who had juvenile records that did not include sex offences.  Both of these rates were, however, higher than the rate of adult sex offences for those with no juvenile record (1.5%).   It appears that “juvenile sex offending does nothing to predict the type of adult record, specifically adult sex offending, above and beyond the frequency of [overall] offending…. Sex offenders are frequent offenders who roll the dice more often and increase their chances of accumulating a sex offence in their career” (p. 526-7, emphasis added).   When one looks at the backgrounds of males who have had contact with the police, as adults, for sex offences, it appears that only 4% of them had a juvenile sex offence.  What does appear to predict adult sex offending is simple: it is high rates of juvenile offending of any kind, whether that offending included a sex offence or not.

Conclusion: The data demonstrate quite clearly that addressing adult sex offending by concentrating one’s efforts on juveniles who have committed sex offences is a foolish strategy: it will miss most adult sex offenders and will mis-identify most of the targeted group.   These findings are particularly relevant given that “most [U.S.] states currently require juvenile sex offenders to register…. These registries are inappropriate because those on the list may not be any more likely to commit another sexual offence as [adults than are those] who are not on these lists” (p. 530).

Reference:  Zimring, Franklin E., Alex R. Piquero, and Wesley G. Jennings (2007). Sexual Delinquency in Racine: Does Early Sex Offending Predict Later Sex Offending in Youth and Young Adulthood.  Criminology and Public Policy, 6(3), 507-534.

Back to top

Ordinary citizens want sex offenders to be subject to civil commitment procedures largely in order to ensure that they are punished sufficiently.  

It is well established that recidivism rates for sex offenders do not differ substantially from the recidivism rates of other offenders (see Criminological Highlights 3(3)#3, 5(1)#4, 6(3)#3, 6(6)#8),  8(3)#8).  Nevertheless, there are many policies that appear to assume that special programs are needed to control sex offenders when they are released into the community (see Criminological Highlights 8(6)#5).  One such procedure is the civil commitment of offenders after they are released from prison because they are thought to be likely to commit another offence. Even though the identification of repeat sex offenders has been shown not to be reliable (see Criminological Highlights 8(6)#5),  such practices appear to be popular.  

This paper tries to understand the motives that underlie popular support for the civil commitment of those who have already served their sentences for a sex offence.  One possibility is that the public believes that such procedures are necessary for public safety.  A second possibility is that the public simply wants an opportunity to increase the punishment that these offenders receive. 

The assumption behind the study was that if members of the public were concerned about public safety, then professional estimates of the likelihood that a particular prisoner would re-offend would affect their judgements of whether that offender should be civilly committed. On the other hand, if those same members of the public supported civil commitment because they considered the sentences that these offenders received to be insufficient, the only thing that should matter in their decision making would be the severity of the sentence that the offender was serving. 

Two groups of people (jury eligible Americans with a median age of 47 and university students) were given vignettes describing a sex offender who was completing a sentence for two sex offences.  For approximately half of the participants, the sentence was described as harsh (25 years); for the others it was described as relatively lenient (3 years in a minimum security institution).  Different groups of participants were told that a careful assessment of the offender estimated his likelihood of re-offending as 0%, 4% or 70%.   Participants were then asked whether they supported or opposed civil commitment of the offender after the offender had served his complete sentence. 

Generally speaking, support for civil commitment was higher when the offender was described as having received a lenient sentence.  Support for civil commitment was affected by the professional estimate of the likelihood of re-offending, however, only when the original sentence was seen as sufficient. When the sentence he was completing was seen as lenient, the offender’s probability of recidivism had a much smaller impact.  Further support for the hypothesis that the desire for civil commitment was really a desire for a harsher sentence came also from another finding: substantially more people thought the offender should be civilly committed to a psychiatric hospital rather than a prison when he was seen as having been adequately punished.

Conclusion:  The study provides evidence suggesting that public support for the civil commitment of sex offenders after they serve their sentences comes largely from the belief that they have not been punished enough.   The public safety goal of avoiding further offending through incapacitation appears to be relevant largely for those who are seen as already having been adequately punished.

Reference: Carlsmith, Kevin M., John Monahan, and Alison Evans (2007).  The Function of Punishment in the “Civil” Commitment of Sexually Violent Predators. Behavioural Sciences and the Law, 25, 437-448.

Back to top

Community notification about the whereabouts of sexual offenders released from prison has a negative impact on the very factors that appear to be important for their peaceful reintegration into society.

The special procedures for sex offenders coming out of prison in many jurisdictions appear to be based on the false assumption that sex offenders are particularly likely to re-offend (see Criminological Highlights 3(3)#3, 5(1)#4, 6(3)#3, 6(6)#8), 8(3)#8).  Generally, there appear to be two related procedures imposed on this group when they are released from prison: legislative restrictions on such matters as where they can live,  and registries of ‘known sex offenders’ (see Criminological Highlights, 5(6)#17(4)#4, 8(6)#5) which may or may not include public notification procedures.  From what is known about sex offences, it is not surprising that these procedures seem to be of dubious value (see Criminological Highlights, 4(1)#2). 

Procedures such as restrictions on the locations where offenders are allowed to live have been shown to create difficulties in the reintegration of ex-prisoners.  This paper looks at the impact of community notification procedures on the sex offenders themselves. 239 sex offenders living in Connecticut and Indiana were interviewed. In both states, names, addresses, descriptions and colour pictures as well as some information about their criminal records are available to anyone with internet access and are searchable by address. 

A number of offenders mentioned that they felt that there were some positive consequences of the notification laws.  For example, a sizable number (74%) of offenders indicated that being publicly identified made them more motivated to avoid re-offending so as to prove to others that they were not bad people.   Some (34%) believed that communities were safer when people know where sex offenders live.  About a third (31%) indicated that they thought that the notification procedures helped them manage risk factors (because they believed that neighbours were watching).  There is, however, no evidence that notification laws actually reduce re-offending rates. 

On the other hand, large numbers of offenders perceived there to be negative consequences that could interfere with peaceful reintegration into the community. More than half indicated that being identified as a sex offender had each of the following impacts: it increased stress; it kept them from participating in certain activities; it isolated them from others; and it gave them less hope for the future.  Almost half (46%) indicated that they feared for their safety, a feeling that was consistent with the fact that as a result of the notification laws 10% had been physically assaulted or injured, 21% had been threatened or harassed by neighbours, 18% experienced having their property damaged, and 16% reported that a person living with them had been harmed (as a result of their association with a known sex offender). One in five sex offenders (21%) reported that they had lost a job because a boss or co-workers found out about their past.  The results did not differ appreciably between the two states in which the research was carried out.  

Conclusion: Clearly there are negative consequences of efforts to publicize the identity of those who have been released from prison after serving time for sex offences.  Given the absence of convincing data on the efficacy of these procedures in reducing recidivism, it would appear that these broad notification policies “are more likely to undermine the stability of sex offenders than to provide the sweeping protection they intend to achieve” (p. 599).

Reference:  Levenson, Jill S., David A. D’Amora, and Andrea L. Hern (2007). Megan’s Law and its Impact on Community Re-Entry for Sex Offenders.  Behavioural Sciences and the Law, 25, 587-602. 

Back to top

The relationship between the racial composition of neighbourhoods in Seattle, Washington, and violent crime seems to be largely the result of two related factors: labour instability (unemployment, and high rates of undesirable jobs) and social disorder.

It has been argued that jobs vary not only in terms of wages, but also in terms of work conditions, possibilities of advancement,  and, most importantly, in terms of job stability. In many cities, especially (perhaps) in the U.S., neighbourhoods that have high concentrations of Blacks also have high concentrations of those who have undesirable and intermittent employment.

This paper examines the extent to which the link between the racial composition of a neighbourhood and violent crime can be explained by labour instability and social disorder. Labour instability was measured as a combination of the proportion of jobs that were low wage, low skill, with poor promotion prospects and the local unemployment rate.  Social disorder was estimated from reports by residents who were surveyed of troublesome neighbours and other local problems.  The focus of the paper was the violent crime rate (robbery, aggravated assault, rape, murder) in 123 census tracts in Seattle.   

It was hypothesized that race is associated with both labour instability and social disorder, and that labour instability and disorder combine to have an influence on violent crime in the community.  This was confirmed in part by demonstrating that the size of the effects of labour instability and income on crime were reduced when local social disorder was taken into account. 

In Seattle, as in many U.S. cities, violent crime rates are higher in neighbourhoods with high concentrations of people of colour.  It was expected that these high rates would be explained, at least in part, “by neighbourhood labour instability and low income because these factors create greater disorder, which in turn leads to crime” (p. 215).  

However, the effects of these variables differed across minority groups.  Once labour instability was taken into account, the relationship between high concentration of Asians (and immigrants) and high rates of violent crime disappeared. “It appears that the employment disadvantage of neighbourhood residence completely accounts for higher violent crime rates where more Asians and immigrants live.  The relationships between the percentage of neighbourhood residents who are African American or Latinos and violent crime remain when we take into account the labour instability of those living in the community, but the effects of both are substantially reduced” (p. 215).   When average income and level of social disorder are taken into account, the association between violent crime and percent African American or Latino was reduced even further. 

Conclusion: “Neighbourhoods with large percentages of Asian Americans and immigrants have higher rates of violence because they have higher levels of labour instability and social disorder. In contrast, social disorder and labour instability explain part, but not all, of the effects of the percentage of African Americans and Latinos on neighbourhood violence.  Even after taking into account labour instability and social disorder, African American and Latino neighbourhoods still have higher rates of violence than other neighbourhoods” (p. 217).  In effect, these findings remind us that the roots of violence may differ, somewhat, from group to group.

Reference: Crutchfield, Robert D., Ross L. Matsueda, and Kevin Drakulich (2006). Race, Labour Markets, and Neighbourhood Violence. In Peterson, Ruth D. et al. The Many Colours of Crime: Inequalities of Race, Ethnicity, and Crime in America.  New York University Press. 

Back to top

Publication Type

Volume Number


Issue Number