Criminological Highlights Vol. 4, No. 4 – December 2001

View this issue as a PDF: PDF iconCrimHighlightsV4N4.pdf

This issue of Criminological Highlights addresses the following questions:

  1. Do people judge the justice system by its outcomes or by the fairness of its procedures?
  2. Smoke-Free Problems: What are the effects of a total ban on cigarettes in prisons?
  3. Which factor has the most impact on the sentences for young people in Canada – the previous criminal behaviour of the youth or the previous sentencing behaviour of youth court judges?
  4. Are Two Better Than One? - Are two-officer police patrols more effective than single-officer patrols in apprehending burglars?
  5. Working mothers: Are children of working mothers more likely to become delinquent than children of stay-at-home mothers?
  6. Is the large increase in school suspensions and expulsions a result of increased violence in the schools?
  7. Do high rates of firearm ownership lead to increases in homicide rates?
  8. Are decisions to prosecute in a specialized domestic violence court still dependent on the victim’s willingness to cooperate?

The justice system is judged largely on whether it is perceived as being fair in the manner in which it uses its authority.  Drawing from a number of different surveys, it appears that procedural fairness is more important than specific outcomes. 

Background. “People often assume that the outcomes received when dealing with specific police officers and judges shape reactions to those encounters. In contrast… research consistently suggests that people actually react to their personal experiences primarily by judging the procedures used by the authorities” (p. 215).   The manner in which people are treated, as well as whether they feel that decisions are made fairly appear to be of crucial importance.  “People are willing to accept the decisions of police officers, judges, mediators, and other third party authorities when they think that those authorities are acting in ways they view as fair” (p. 216).   Hence, the public’s views of criminal justice institutions are linked more to perceived justice than to specific outcomes or utilitarian concerns.

This study suggests that confidence in the police and the courts is related less to judgments about cost, delay, and performance than it is to perceptions of procedural justice.  The findings are drawn from a number of different sources and can be summarized as follows:

  • A study of Chicago residents’ views of the police and the courts compared the importance of the quality of services (competence) of these institutions with the quality of the treatment that citizens were perceived to receive (fairness).  Both competence and fairness are seen as important, but “the primary influence [on the overall evaluations of the police and courts] is from the quality of the treatment” (p. 218).  One’s sense of obligation to obey the law is influenced by the perceived fairness of the institution, not by its performance.
  • A study of high crime areas - predominantly minority neighbourhoods in Oakland, California - during a period of aggressive policing showed, once again, that the quality of police treatment of citizens (e.g., judgments about police honesty and respect for rights) rather than law enforcement performance (e.g., the impact of the police on crime) dominates the evaluations of the police, as well as residents’ willingness to pay more taxes for increased police services. 
  • A (U.S.) national study of people’s views of the courts found that “the primary influence on overall evaluations and overall ratings of performance [of the courts] come through judgments about the fairness of the outcomes… and the quality of the treatment they provide to members of the public” (p. 226).  
  • In another national study in the U.S., respondents who had been to court in the previous year were asked whether they felt that they would get a fair outcome and be treated justly if they were to go to court in the future. Ratings of the procedural fairness of their own experience were, in all cases, more important than their perception of having received the desired outcome. 

Conclusion. In four different studies, it was found that the quality of the treatment which people receive, or perceive in the community, is the most important factor in determining people’s views of criminal justice institutions.  Although specific outcomes are important, they are not as decisive as procedural fairness. These findings were confirmed for both white and minority groups. Results such as these serve as a reminder that it is not just what the criminal justice institutions do that is important but how they are perceived as doing it.

Reference: Tyler, Tom R. Public Trust and Confidence in Legal Authorities: What do Majority and Minority Group Members want from the Law and Legal Institutions? Behavioural Sciences and the Law, 2001, 19, 215-235.

Back to top

Banning cigarettes in prisons may be a good health practice. However, non-smoking policies may have unforeseen negative effects as cigarettes move from the “largely benign gray markets… into more problematic black markets, where [they] are a highly priced commodity” (p.142).

Background. Many jails and prisons currently prohibit inmates from smoking.  Until bans were imposed, cigarettes were traditionally used “as a standard form of currency in informal prison economies” (p.144). This practice is explained by the fact that they are relatively small, durable, and are not considered serious contraband unless held in large quantities.  When cigarette smoking was merely controlled, various methods were developed to conceal possession and use of tobacco. 

This study examined the impact of bans in 16 correctional institutions in the U.S.. Researchers conducted formal, as well as informal discussions with prisoners and staff.  When tobacco products were banned completely in prisons, one of the consequences was a dramatic increase in the cost of cigarettes. In fact, the value of a carton of cigarettes in correctional facilities ranged from approximately $200 to $500. Various techniques were used to smuggle tobacco into the institution. These strategies involved some risk of apprehension and, in some cases, payoffs to correctional officers.  Since most correctional officers did not see cigarettes as immoral or dangerous, it was not surprising that a portion of them were involved (either directly or indirectly) in the smuggling of cigarettes into prisons. For instance, packs of cigarettes were purchased from correctional officers in one institution for $20 (at least five times the street value) and the individual cigarettes were subsequently sold to inmates for $5 each.  Various practices – some benign, some dangerous – developed to avoid detection.  For example, ingenious methods of hiding cigarettes in toilets were found (p. 150) as well as lighting cigarettes without matches (using pencil lead and toilet paper to start a fire with the electricity from an ordinary electrical outlet, p.157).  For clever inmates, cigarettes clearly provided opportunities. As one maximum security inmate serving a 90-year sentence for drug trafficking lamented, “I would’ve never messed with coke on the street if I knew how much money I could’ve made selling cigarettes here in the joint” (pp. 158-9). 

Conclusion. Although discouraging smoking may be good health policy, most inmates interviewed continued to smoke in the institution. Unfortunately, the financial costs of maintaining this habit were considerable and several individuals believed that robbery within the prison had increased as a result of the ban.  Moreover, health risks may have increased in some situations (e.g., various inappropriate papers – for instance, pages from a Bible that contained ink that is, itself, harmful when burned - were being used as well as the consumption of higher tar tobacco).  Further, the ban resulted in increased tensions between staff and inmates.  Finally, criminal justice consequences clearly exist for the small portion of inmates who are apprehended and who, as a consequence, are likely to spend more time in custody.

Reference: Lankenau, Stephen E. Smoke ’em if you got ’em: Cigarette Black Markets in U.S. Prisons and Jails. The Prison Journal, 2001, 81, 142-161.

Back to top

Canadian youth court judges are influenced more by the previous disposition that a young person received than by his/her past criminal behaviour.  Youths tend to be handed down either the same type of sentence that they received the previous time or a more severe sanction, independent of the present crime that they committed.  In other words, it is not just what you did, or who you are, it is what the judge did the previous time that makes the difference.

Background.  Typically it is thought that legal variables – the current offence and the youth’s history of offending – are the most important factors in determining the sentence.  However, research also suggests that “the severity of earlier sanctions is an important factor affecting current disposition – specifically the decision to impose a prison sentence” (p. 170). 

This study examines the effects of previous youth court dispositions on sentences handed down in Canadian youth courts in 1993-4.  Attempts were made to control for the seriousness of the current and most recent prior charges of conviction, as well as age and gender.  Data on both the most recent prior disposition and the second most recent prior disposition were examined.

The results show that the most recent prior disposition and the second most recent disposition were important in determining the sentence handed down.  For example, if a youth received secure custody on his/her most recent as well as the second most recent dispositions, his/her probability of being given secure custody for the “current” offence was .65.  On the other hand, if the most recent disposition was secure custody, but the second most recent disposition was probation, the probability of being handed down secure custody at this time was only .45. Clearly, remote dispositions appear to have an influence on the current sanction. 

Not surprisingly, the most recent disposition also has an influence. For instance, this article examined the sentences given to youths whose second most recent previous disposition was open custody. For these youths, the probability of being handed down secure custody for the current offence after having received secure custody as the most recent previous disposition was .48. In contrast, the probability of being given secure custody for the current offence if the most recent previous disposition was open custody was .28.  Further, in those cases where the most recent previous disposition was probation, the probability of receiving secure custody dropped to only .13.  Statistical analyses demonstrated that these effects were independent of the current offence. 

Generally speaking, there is evidence for stability and escalation.  “Consistency over the second prior [disposition], first prior [disposition] and the current disposition is more likely than switching between dispositions… Where there is a change from one disposition to another, it is more likely to be escalation than de-escalation” (p. 195) for those cases involving probation, as well as open and secure custody.

Conclusion. The disposition handed down to a young person in Canada’s youth court is “strongly influenced by prior dispositions, rather than being entirely guided by the nature of the current offence” (p. 195).  “Prior dispositions are ‘sticky’ labels which tend to last indefinitely” (p.195).  “The nature of the most recent prior offence and the age and gender of the offender had only a weak association with the current disposition when relevant case characteristics were controlled…. The finding that the effect of a prior offence is mediated by the disposition awarded for that offence has the intriguing implication that it is the dispositional history rather than the history of offending that is salient in later dispositions… Thus a criminal history is a history of judicial reactions to past behaviour, rather than the behaviour itself” (p. 196).   “For young offenders with prior records, it will be difficult for [them] to receive less severe court responses that match any possible reduction in the seriousness of the reoffending” (p. 197).

Reference: Matarazzo, Anthony, Peter J. Carrington, and Robert D. Hiscott.  The Effect of Prior Youth Court Dispositions on Current Disposition: An Application of Societal-Reaction Theory.  Journal of Quantitative Criminology, 2001, 17, 169-200.

Back to top

Two-person police patrols are no better than one-person patrols in catching burglars in the act.  Burglars are most likely to be caught if the police are called at the beginning of the burglary, not when the burglar is on his/her way out of the building.  For these burglaries, speed of response – not the number of police in the speeding car – makes a difference. 

Background. Solving crimes after the offender has left the scene and without a witness who can identify the assailant is very difficult.  Therefore, it has been suggested that the police not only need to arrive quickly at the scene of a crime but they also need to have sufficient force to apprehend the offender.  Although two-person patrols have been promoted for various reasons (e.g., officer safety and the ability to apprehend offenders more effectively), research suggests that increased patrol strength is not particularly effective in deterring crime or making arrests.  Moreover, “studies indicate that single-officer patrols tended to be safer… and [officers] are no more likely to be injured, irrespective of type of shift or police precinct…” (p. 382). 

This study was carried out in the U.K. and examined a sample of 441 burglaries which were drawn from the 9% of burglaries that were reported in 1996 while “in progress” (p. 383).  [Clearly, having two police officers attend a burglary in which the offence took place hours or days earlier would not be expected to have any impact.]  

The findings suggest that there are two main determinants of whether a burglar is caught at the scene of the crime: the moment at which the police are called, and the length of time that it takes them to arrive.  When the police were called as the burglar was entering the dwelling, 19% were caught. In comparison, there was an 11% apprehension rate when the police were contacted after the burglar was already inside the building, and a 2% success rate when the burglar had already left the premises. If the police arrived at the scene within 4 minutes of being called, arrests were made in 15% of the cases. Conversely, the success rate dropped to approximately 8% after 6 minutes.  

The fact that two-officer patrols tended to reach the scene of the offence more quickly than single-officer units could easily make it appear that the former strategy was more effective.  However, when various characteristics of the two-person patrols were controlled for, the presence of two police officers in the car did not seem to make a significant difference.  More specifically, two-officer patrols appeared to do better because they had certain advantages (e.g., they tended to be driving faster cars or were more likely to respond to incidents in which the burglar had been spotted earlier).  Overall, the findings are simple to describe: “Chances of capture were significantly better when the burglar was spotted entering or inside the building rather than leaving, when someone other than the victim reported the incident, and when the police reached the scene quickly.  When these factors are taken into account, patrol unit staffing is of little or no importance” (p. 393). 

Conclusion. Catching offenders “red handed” is obviously advantageous for everyone.  The one factor largely under police control which affects apprehension rates is the speed at which the police arrive at the scene.  Hence, an increase in single-officer patrols would likely be more effective than two-person patrols.  However, these results need to be understood in the context of the finding that most burglaries are reported too late for this to be of any real importance.

Reference: Blake, L., and R. T. Coupe.  The Impact of Single and Two-Officer Patrols on Catching Burglars in the Act. British Journal of Criminology, 2001, 41, 381-396.

Back to top

Maternal employment has no direct impact on the delinquency of children in the family. Children of working mothers are no more likely to be involved in delinquent behaviour than are children whose mothers are not in the work force.  

Background. One of the largest structural changes in American society in the past 50 years has been the increase in female labour force participation.  While only 16% of all American children had working mothers in the 1950s, the number rose to almost 70% in the 1990s. Although there are numerous claims that maternal employment “causes” delinquency, few data exist to support this notion. Part of the explanation for the lack of effect of maternal employment on juvenile crime may be that it is “most beneficial for children when the alternative is poverty…” (p. 21). 

This study looked at a group of 707 adolescents between the ages of 12 and 14 years.  The subjects of this sample tended to be born to mothers who were younger, less well educated, and more likely to be members of minority groups than the population at large. As such, they may constitute a “high-risk” group.  Self-report delinquency measures were obtained on these youths.  Data were collected on the mother’s employment at the time that the child was 12-14, as well as when he/she was 6-8 years old. 

The findings are straightforward.  Mothers’ employment (either when the child was 6-8 years old or when the child was 12-14 years old) had no impact delinquency when the child was 12-14 years old.  However, low family income when the child was an adolescent was related to delinquency. Delinquency was associated with reports by the child that his or her mother did not know whom he was with, the presence of delinquent friends, and dissatisfaction with one’s school experience.  The only indication that maternal employment may have some small and indirect impact on delinquent behaviour was the fact that maternal employment was one of the predictors of the child’s report of whether his or her mother knew whom he was with. However, maternal employment was not found to be directly associated with delinquent behaviour.  Put simply, children whose mothers were employed (either when they were 6-8 years old or in early adolescence) were no more likely to be delinquent teenagers than were children whose mothers were at home. 

Conclusion. “The widespread concern over the fates of working women and their children is largely unsupported” (p. 252).  The study’s results are consistent with other research on delinquency: supervision, association with peers, and attachment to school - factors that are directly related to the youth’s own experiences – predict early adolescent delinquency.  “If improving family life is a goal of crime control policy, it would make good sense to aim at addressing the structural factors that limit maternal and family resources and that contribute to community disorder. [This] study suggests that policy debates should avoid ideological attacks on working mothers, which portray them as leaving their children “home alone” and concentrate instead on the economic and educational inequalities that weaken families and neighbourhoods” (p. 154).  Mothers in the work force are not a cause of crime.

Reference: Vander Ven, Thomas M., Francis T. Cullen, Mark A. Carrozza, and John Paul Wright. Home Alone: The Impact of Maternal Employment on Delinquency.  Social Problems, 2001, 48, 236-257.

Back to top

Violence in schools in the U.S. has not increased in frequency or seriousness in the past quarter century.  Although the youths have not become more violent, the schools have become more punitive whereby suspensions and expulsions have doubled. Hence, it is the school’s response to the youth and not the behaviour of the young people that may have the greatest negative impact.

Background. Homicides by youths in the U.S. (and in Canada) are not increasing.  At the same time, most Americans and Canadians believe that they are, in fact, on the rise.  This is not surprising.  For example, a study of local TV coverage in California found that 70% of news stories about violence involved youths. In contrast, only 14% of the arrests for violence involved young people. These results were corroborated by other research which showed that half of the stories (of any kind) that concerned children or youth involved violence.

The data on youth and the schools’ response to violence suggest that school policies are responding to beliefs rather than facts about school violence.  Between 1976 and 1998, there was no change in the number of youths who reported, in large scale surveys, having been victimized on any of the following dimensions: being the victim of theft or vandalism, being injured with a weapon, being threatened with a weapon, being injured without a weapon or being threatened without a weapon.  “Despite remarkably stable rates of student victimization over the past 23 years, suspensions and expulsions have increased… from 3.7% of students in 1974… to 6.8% of students in 1998 (3.2 million students suspended)” (p. 3).  Given that the school is one of the few institutions in which positive interventions into the lives of young people can occur with little difficulty (i.e., without the obstacles inherent in mediation involving the family or peers), these large percentages translate into lost opportunities for positive intervention by the educational system. 

“Code of conduct” policies such as those of the Province of Ontario require suspensions for such behaviour as uttering a threat, possessing illegal drugs, or providing alcohol to minors (e.g., a 19 year old giving a can of beer to an 18 year old).  Hence, despite the fact that Canadian data also do not suggest an increase in youth violence, our policies appear to be based on a similar demonization of young people.  

Conclusion. The “mass exclusion of … children from the educational process” (p.4) has been criticized on various grounds, including the fact that it appears to increase the likelihood of troublesome behaviour by these youths, as well as augment the chances that young people will drop out of school. In the U.S., suspensions are strongly associated with race: “African Americans are approximately 2.6 times as likely to be suspended from schools as whites” (p. 4).  Hence, consideration of these outcomes requires one to remember that the data neither support the view that school violence is increasing nor that schools are a particularly dangerous place for youths.  In fact, the data suggest that schools are a safe place to be when compared with the places in which youths spend the other 80% of their waking hours.

Reference: Schiraldi, Vincent and Jason Ziedenberg.  Schools and Suspensions: Self-reported Crime and the Growing Use of Suspensions.  Justice Policy Institute Policy Brief. 1 September 2001. [Online]. Available:

Back to top

Homicide rates across 36 nations are related to private firearm availability.  It is not the case that high homicide rates lead to high levels of firearm availability.  

Background.  In comparison with most industrialized nations, the rate of lethal violence in the U.S. is considerably higher, even though the rates of other criminal offences are generally similar.  Gun homicide rates, in particular, are much higher in the U.S. than in other countries. One explanation is that the rate of “gun possession in the U.S., particularly handgun possession, exceeds that of every other country for which data are available” (p.570). Research has found that, across 21 countries, a substantial relationship exists between the rate of homicides - particularly of women - with a gun and the rate of handgun ownership (See Killias: Canadian Journal of Criminology, October 2001).  Alternatively, some have suggested that firearm ownership, particularly handgun ownership, is a consequence of high rates of lethal violence, rather than a cause of it.

This study examined the relationship between gun ownership and homicide rates in 36 nations in an attempt to determine the direction of the relationship (high gun possession leading to vs. resulting from high homicide rates).  Because rates of gun possession are only available from surveys carried out in 21 countries, a proxy measure – gun suicide rates – was used as an indicator of firearm availability.  It turns out that the correlation between the percentage of suicides involving a gun and the survey data on firearm possession is very high – a correlation coefficient of +0.90 (where a correlation of 1.0 would mean a “perfect” relationship).  The proportion of suicides involving firearms varies considerably from less than 1% (South Korea) to 61% in the U.S.. Various control variables were included in the analysis: population density, proportion of the population that are young males, ethnic heterogeneity, the generosity of the welfare system, and income disparity. 

The analysis involved a statistical technique designed to help determine the direction of the relationship.  Controlling for other variables, results show that the homicide rate does not predict firearm availability. Conversely, firearm availability does predict homicide rate, once control variables are entered into the equation.

Conclusion. The findings are consistent with the hypothesis that “greater gun availability increases the likelihood that a gun will be used in an attack, which in turn increases the risk that the victim will die of his or her injuries” (p. 587). Hence, the “contention that widespread gun ownership deters homicide is not supported in the present analysis” (p. 588). “Results do not support the hypothesis that high levels of lethal violence cause people to purchase firearms in order to protect themselves. Homicide rates had no effect on the measure of firearm availability” (p. 588). 

Reference: Hoskin, Anthony W.  Armed Americans: The Impact of Firearm Availability on National Homicide Rates. Justice Quarterly, 2001, 18, 569-592.

Back to top

Despite the purposeful design of a Toronto “domestic violence” court to allow the prosecution of cases in a manner that does not rely solely on the victim’s testimony, the likelihood of a prosecution going forward is considerably higher if the victim cooperates. Even in those cases in which the victim’s collaboration is not supposed to matter, it does.  

Background. With the popularity of mandatory arrest policies for domestic violence, one might have expected high levels of prosecution of these cases.  However, studies in some jurisdictions suggest that as few as 10% of these cases are being prosecuted.  The lack of cooperation of the victim appears to be a major impediment. Victim cooperation, itself, is likely to be partially a result of a justice system that ignores the reality of the situation of women (91% of the victims in this study were women) who have been victimized.  Existing research suggests that “one of the major obstacles to cooperating is intimidation by the accused or fear of reprisal if they cooperate” (p. 599).  Other factors include lack of responsiveness of the criminal justice system to the victim’s needs and a desire on the woman’s part to continue or repair the relationship, as well as women’s acceptance of violence from their partners.

This study looked at the 474 cases brought to a specialized Toronto court in 1997-1998. This court was established to improve the response of the criminal justice system to domestic violence. According to the Ontario Crown Policy Manual, charges were not to be withdrawn in these cases except in exceptional circumstances.  The police were to gather evidence “so that the prosecution need not rely solely on victims’ testimony” (p. 603).  Statements of the victims were to be videotaped and a victim/witness assistance program made attempts to contact all victims in this court.  This program was designed to keep clients aware of the court proceedings and make the victims comfortable with the legal process in various ways.

The results showed that when the victim cooperated, the chances of the prosecution taking place were dramatically higher than when the victim did not.  Hence, even in a situation in which the court and prosecution service were structured to minimize the importance of victim cooperation, their collaboration was, by far, the most important variable determining whether the case was prosecuted.  Not surprisingly, victims who had already been videotaped and who had met with the victim/witness assistance personnel were more likely to cooperate.  In other words, early cooperation predicted subsequent cooperation which, in turn, predicted prosecution.

Conclusion. From interviews with victims, it appears that support given by the victim/witness program may be important in encouraging cooperation.  If successful prosecution is one of the goals of the criminal justice system, ways of creating cooperative victims need to be examined. These programs appear to be promising in this regard.  At the same time, the interview data highlighted the importance of the police as well.  Hence “to be effective, policies that seek to improve the criminal justice response to domestic violence must be informed by the victims’ experience throughout the process.  Such an approach will help us to address more effectively the issue of victim cooperation” (pp. 619-620) which will, in turn, lead to more successful prosecutions.

Reference: Dawson, Myrna, and Ronit Dinovitzer.  Victim Cooperation and the Prosecution of Domestic Violence in a Specialized Court. Justice Quarterly, 2001, 18, 593-622.

Back to top

Publication Type

Volume Number


Issue Number