Criminological Highlights Vol. 12, No. 2 – October 2011

View this issue as a PDF: PDF iconCrimHighlightsV12N2.pdf

This issue of Criminological Highlights addresses the following questions:

  1. What’s the best time to ask a judge to make a decision?
  2. How should you interpret statements about criminal justice interventions that appear to be too good to be true?
  3. When members of the public say that they want harsh penalties for youths, what do they mean?
  4. Should the whereabouts of sex offenders be monitored with GPS devices?
  5. Why are many ‘residency restrictions’ for sex offenders impossible to implement effectively?
  6. What kinds of drug courts are most likely to have positive impacts?
  7. How can symptoms of depression be reduced in youths who are incarcerated?
  8. Why don’t some people call the police when there is trouble?

Parole decisions made by judges at the beginning of the day or immediately after a food break are dramatically more likely to be successful than those made immediately before a break.

Making a series of complex decisions in a short period of time appears to “deplete individuals’ executive function and mental resources which can, in turn, influence their subsequent decisions” (p. 6889).  Studies on various forms of judgments would suggest that “making repeated rulings can increase the likelihood of judges to simplify their decisions…. [In the context of parole decisions] they will be more likely to accept the default, status quo outcome: deny a prisoner’s request” (p. 6892). Other research has demonstrated that executive function can be restored by breaks, short rests, and food intake.  

1,112 decisions of 8 Israeli judges (6 male, 2 female) presiding in parole hearings in 4 prisons in Israel were examined. In making their decisions, the judges were advised by two others (a criminologist and a social worker), but they alone made the decisions.  Most decisions involved requests for release, though about 22% involved requests to change conditions.  Judges typically took a mid-morning break (and were provided with a snack) as well as a lunch break.  Decisions were coded as either ‘accept’ the request or reject it. About half the ‘rejected’ group included a stipulation to review the case at a later date. 

At the start of each session (i.e., the beginning of the day or after the morning break), about 65% of the rulings were favourable.  By the time the judge took a break, success had decreased to close to zero.  After each break, the likelihood of a favourable ruling returned to about 65% and then decreased dramatically and consistently over time.  This pattern was evident for prisoners who had served different amounts of time (ranging from less than a year to more than 3 years).  The pattern (less success until the next break) was similar for those serving their first prison sentence as well as for those with a prior incarceration.  A more detailed multivariate analysis confirmed the effect.  This analysis demonstrated, however, that, overall, other factors were taken into account in the judge’s decision. Those with previous imprisonments were less likely to be released, those who had participated in rehabilitation programs were more likely to be released, and those who appeared before judges whose first decision of the day was favourable were more likely to receive a favourable ruling.  This last finding suggests that some judges were, overall, more likely than others to hand down favourable rulings, but that the ‘order’ effect held for them as well. 

Judges, when deciding when a break would occur, did not know the nature of the upcoming cases.  The order of the cases was determined by the arrival time of the prisoner’s lawyer, not by the judge, and the lawyers were kept separate from the proceedings. In fact, lawyers could not arrange for their cases to be heard at any special time in relation to the breaks.   A survey of lawyers showed that they had no knowledge that there was an advantage of having their case heard after a break. 

Conclusion. “When judges make repeated rulings, they show an increased tendency to rule in favour of the status quo.  This tendency can be overcome by taking a break to eat a meal, consistent with previous research demonstrating the effects of a short rest, positive mood, and glucose on mental resource replenishment” (p. 6892).  It cannot be determined from this study whether the effect is due to the break, the food, or both.   However, it is clear that extraneous factors can affect judges’ decisions concerning the liberty of offenders.  “The caricature that justice is what the judge ate for breakfast might be an appropriate caricature for human decision-making in general” (p. 6892).

Reference: Danziger, Shai, Jonathan Levav, and Liora Avnaim-Pesso (2011). Extraneous factors in judicial decisions. PNAS (Proceedings of the National Academy of Sciences of the USA), 108 (17), 6889-6892. 

Back to top

When criminal justice interventions appear to be ‘too good to be true’ it is probably because the results being described are too good to be true.

Previous research that has examined the quality of research designs used to evaluate crime prevention interventions has shown that weaker research designs tend to show stronger effects (see, for example, Criminological Highlights, 4(2)#8).  For example, simple “before vs. after” research designs that control for almost nothing (e.g., that do not control for other independent events occurring at the same time, trends that have nothing to do with the intervention, selection of different people or locations into the ‘treatment’ and ‘control’ groups, regression effects, maturation effects) tend to show stronger effects of a crime prevention intervention than do studies that have adequate controls.

There is a large body of research evaluating the effectiveness of various surveillance methods – e.g., closed circuit television (CCTV) and improved street lighting – on crime.  A review of only high quality studies in these areas suggests, for example,  that CCTV is effective in certain situations (e.g., in automobile parking lots) and that improved street lighting is more effective in reducing property than violent crimes. 

In recent years, there have been 5 systematic reviews of public area surveillance studies that looked at a total of 136 separate evaluations.  Each of these studies was coded as finding the intended positive effect (that the intervention reduced crime), no effect (or an effect that was reported as being not statistically significant) or a negative effect (supporting the conclusion that the intervention was harmful).  Overall 49% of the studies showed the intended positive effect, 43% showed no effect, and 7% showed a negative effect.   

A common problem with research in this area is in not having an adequate control or comparison group: 37% of the studies did not have a comparison group, and an additional 19% had a non-comparable control group. 

Of the 60 studies that had a control group that was comparable to the group that received the intervention, 37% showed crime reducing effects of the intervention.  Of those studies with no control or a non-comparable control, 59% presented results that suggested that there were crime reducing effects of the intervention. 

The difficulty, of course, is that often the interventions that are studied are implemented in a manner that makes it difficult or impossible to evaluate them adequately.  In addition, those responsible for criminal justice interventions whose careers may well be affected by the outcome of evaluations of what they are doing may not have the motivation to carry out studies with fully adequate controls. 

Conclusion:  This study demonstrates that good design does matter: favourable results in criminal justice research are more likely to be shown in inadequately designed studies than in studies with adequate designs.  The presence of a randomized or comparable control group is almost always a necessary condition  for  drawing any inferences about the efficacy of a criminal justice intervention.  Studies without adequate designs cannot be taken seriously.

Reference: Welsh, Brandon C., Meghan E. Peel, David P. Farrington, Henk Elffers, and Anthony A. Braga (2011).  Research Design Influence on Study Outcomes in Crime and Justice: A Partial Replication with Public Area Surveillance.  Journal of Experimental Criminology, 7, 183-198.

Back to top

Sentencing youths to life in prison without eligibility for parole is a popular public option for youths who murder for a very simple reason: Members of the public don’t know that there are plausible alternatives to this sanction.

In 2005, there were an estimated 2,225 people in the U.S. serving life sentences without eligibility for parole (LWOP) for murders that they committed when they were 17 years old or younger.  A 2010 U.S. Supreme Court held that LWOP for any offence other than murder was unconstitutional. LWOP is a rare sanction for people of any age outside of the U.S.  A survey of 154 countries found only 3 that allowed this sanction for youths and they, in total, only had 12 youths serving these sentences.   

Other studies show that the public’s preference for harsh penalties decreases dramatically when alternative sanctions are offered as possibilities.  For example, support for the death penalty for adult murderers in the US decreases substantially if other options – e.g., life with a long parole ineligibility period or LWOP – are offered as alternatives. 

In Michigan, children of any age can be sentenced to LWOP in an adult prison, though depending on the age of the child, other options are also possible.  In a survey carried out in Michigan in 2005-6, respondents were first asked “How strongly do you agree or disagree with Michigan’s current law that requires an adolescent to be sentenced to life without parole for certain offences?”  43% agreed or strongly agreed with the law.

In a follow-up question, respondents were asked which of six sentencing options they thought “an adolescent convicted of committing a homicide should receive as a punishment.”  When presented with a larger menu of choices, 5% favoured LWOP in an adult facility, and an additional 11% chose confinement in a juvenile facility until 18, and then LWOP. Hence a total of 16% favoured LWOP in one or the other of these forms. The most popular options involved confinement in a juvenile facility until 18, and then a prison sentence with the possibility of parole in 20 years (41%) or less than 20 years (25%).  An additional 5% preferred that the initial confinement be in an adult facility, but that parole would be possible after 20 years. 13% preferred confinement in a youth facility until age 21 and then release.  

In simple terms, however, the support for LWOP dropped from 43% of the respondents to 16%.    Support for LWOP was lower among females, African Americans, and those over 30 years old.

Conclusion:  The results are consistent with other findings showing that when respondents are provided with a range of plausible sanction options, support for highly punitive options often decreases dramatically.  Other studies have shown that the public may talk tough in sentencing matters, but in fact wants mitigating factors to be considered when sentences are being imposed (Criminological Highlights 12(1)#4).  Similarly, Canadian respondents were much more favourable toward conditional sentences of imprisonment after they understood the nature of the punitive conditions that could be imposed under a conditional sentence (Criminological Highlights 3(3)#4).  The public supports mandatory minimum sentences, but, simultaneously would prefer that there be some mechanism to ‘opt out’ of the mandatory requirement (Criminological Highlights, 6(2)#6,  8(6)#1).  The findings in this and other studies suggest that politicians who prefer to follow their constituents rather than lead them would do well to ensure that those whom they are following have complete knowledge of the options that are available.  

Reference: Kubiak, Sheryl Pimlott and Terrence Allen (2011).  Public Opinion Regarding Juvenile Life Without Parole in Consecutive Statewide Surveys. Crime & Delinquency, 57(4), 495-515.

Back to top

There are good reasons to believe that monitoring sex offenders with GPS devices will waste money and have no effect on reoffending.  

Most popular approaches to sex offender recidivism ignore what is known about sex offenders. Sex offenders are no more likely to reoffend than other offenders (Criminological Highlights 3(3)#3, 5(1)#4, 6(3)#3, 6(6)#8, 8(3)#8, 9(2)#5). Given that most sex offences are committed by people known to their victims, residence restrictions, registration and notification schemes are ineffective (Criminological Highlights, 4(1)#2, 5(6)#1, 7(4)#4, 8(6)#5, 9(2)#7), 10(3)#7, 11(4)#7, 11(6)#6). 

This paper looks at one of the newer technological approaches to sex offence recidivism: GPS (global positioning system) monitoring of sex offenders (GPS-MSO).  GPS-MSO is obviously based on the notion that there are ‘safe’ and ‘dangerous’ places for sex offenders to be.  These approaches fail to notice that these restrictions, themselves, are inconsistent with what is known about sex offending (e.g., that it typically occurs between people known to one another) and, empirically, do not contribute to public safety. 

GPS-MSO has been proposed in some locations as the next stage of control of sex offenders in the community. However, what evidence there is on its effectiveness does not support the view that GPS-MSO is effective. Nevertheless, because it is ‘new’ and ‘high tech’ it is seen as ‘good.’ One problem, of course, is by treating all sex offenders the same, jurisdictions fail to discover whether there are any highly specified circumstances in which GPS-MSO could be useful. 

Public surveys suggest that the vast majority of ordinary citizens (who are likely not to be informed about basic facts concerning sex offences or sex offenders) say that they would feel safer if GPS-MSO were to be broadly instituted. This belief ignores findings such as the fact that the vast majority of sex crimes targeting children (96% in one study) are committed by people known to the victim.    These findings also ignore the fact that GPS technology does not stop someone from offending.  GPS technology can be used to determine – after the fact – where an offender has been. It can be checked at regular intervals (e.g. daily or weekly) to monitor compliance with conditions or, in the much more expensive version, it can involve constant real-time monitoring.  Both of these approaches involve a substantial amount of error (e.g., if a train carrying a sex offender were to pass close to a school or park that was ‘off limits’ because of the presence of children). 

There are other concerns, however.  Rather than attempting to reintegrate sex offenders (see Criminological Highlights 9(3)#6, 11(2)#6) which can be effective at reducing recidivism, or provide effective treatment (Criminological Highlights 9(5)#7), GPS-MSO appears to be an effective way of funneling resources into unproven approaches.  Violation alerts – many being false alarms – occur frequently.  Furthermore, when they do occur, those given the task of monitoring must follow up, especially if the offender is subject to real-time monitoring. One agency with this task noted that “uncovering non-compliant behaviour patterns in GPS data is not always easy, but rather entails sifting through roughly 2 million data points per offender annually” (p. 184).  From the perspective of a community corrections office, if someone subject to GPS-MSO were about to commit an offence, it could be very difficult for those monitoring him to respond to an alert quickly enough to stop the offence from happening, especially given the number of false alarms.  

Conclusion: Almost certainly, “the incapacitative and public safety potential of this sanction has been overstated” (p. 185).  Proposals for its use are based on the notion that sex offending is caused by opportunity and location. “Most research, however, shows that sexual assaults of all types are rarely impulsive events; rather, offenders plan their assaults and use strategies to gain access to victims, acquire their trust, and commit the assault” (p. 185) – behaviours and crimes that would be unaffected by GPS monitoring. 

Reference: Payne, Brian K. and Matthew DeMichele (2011).  Sex offender policies: Considering unanticipated consequences of GPS sex offender monitoring.  Aggression and Violent Behaviour, 16, 177-187. 

Back to top

One problem with sex offender residency restrictions is that they can make it impossible for a sex offender to find a legal place to live.

Even though reoffending rates by sex offenders are not much different from reoffending rates of other offenders (See Criminological Highlights 3(3)#3, 5(1)#4, 6(3)#3, 6(6)#8) 8(3)#8, 9(2)#5), many jurisdictions have imposed special conditions on sex offenders who are released into the community.  Sex offence residence restrictions are now quite common even though they have been demonstrated to be ineffective (Criminological Highlights, 11(4)#7, 11(6)#6).  This paper demonstrates that ordinary residence restrictions placed routinely on sex offenders can also make it almost impossible for them to find a place to live.

Residence restrictions are based on the theory that sex offenders commit their offences close to where they live, and that residency restrictions will keep offenders away from potential victims.  The problem is that the evidence does not support the view that re-offending, when it does take place, involves victims who live near the offender.  On the other hand it has been pointed out that “forcing offenders to live away from family, friends, and community resources can result in social isolation and difficulty reintegrating into the community” (p. 237).  Another problem is that ordinary restrictions (e.g., not living within 1000 feet of a location where children are likely to be found) mean that most parts of a city may be off-limits for sex offenders who have returned to the community.

This study looks at two areas of New York State: Erie County and Schenectady County.  The restrictions on sex offenders vary across location and typically involve prohibitions against living  within 1000 to 2000 feet of schools, daycare centres, playgrounds, parks, and in the case of at least one municipality,  additional institutions such as libraries, skating rinks, and senior citizens’ residences.   

In this study, the targeted locations were identified, and the prohibited areas around them were identified. On the surface, these counties (which include rural and urban areas) would appear to have plenty of places for sex offenders to live:  Less than 20% of the total county area, and less than 25% of the space zoned for residences were off-limits.  However, when actual residences were examined, it was found that 89% of the residences in one county and 73% in the other were off-limits for sex offenders.  Only in rural areas (and, in one county, trailer parks) were the majority of residences legal for sex offenders.  In the two cities (Schenectady and Buffalo) 96% and 94%, respectively, of the actual residences were legally off limits to those who had been convicted of a sex offence. 

Registered sex offenders live in each of these two counties. Because they are required to register their residence with the police, it is straightforward to see if they live in restricted areas.  About 90% live in restricted areas, perhaps because they were resident there before the law was enacted in 2005.  But in addition, local police departments may not have the resources or motivation to enforce these restrictions (or even determine if they were resident in that location before 2005) in part because of the effort involved to amass the data needed to demonstrate proximity of the residence to a prohibited location.  It has also been suggested that these residence restriction laws were never meant to be enforced and were enacted purely for symbolic purposes. 

Conclusion:  In addition to not being effective in reducing reoffending, it would appear that residency restrictions that are imposed on sex offenders are largely unenforceable in part because they would leave sex offenders almost no place to live.  The problem, of course, is that residency restrictions – like public notification requirements – make it harder for sex offenders to reintegrate peacefully into the community. 

Reference: Berenson, Jacqueline A. and Paul S. Appelbaum (2011).  A Geospatial Analysis of the Impact of Sex Offender Residency Restrictions in Two New York Counties.  Law and Human Behavior, 35, 235-246.

Back to top

Most drug courts have been shown to reduce recidivism, but their effectiveness depends on the characteristics of the offenders and the way in which drug courts are operated.

There has been a large increase in the popularity of ‘drug courts.’  The evaluations of these courts tend to show that graduates have a somewhat lower likelihood of reoffending than those who are not given the opportunity to participate in drug courts. Part of the effectiveness of drug courts, of course, is that they connect offenders with treatment services.  At the same time, some drug courts have been criticized because drug addicted offenders are sometimes provided services sooner than more needy non-offenders.  It is clear that some drug courts work better than others. However, there has been little research on the factors that determine how effective a drug court is. 

In this paper, the characteristics of 76 distinct drug courts that had been evaluated were assessed using a combination of material in published reports and interviews with those knowledgeable about the operation of the courts. Overall it was estimated that without drug court intervention, the recidivism rate for offenders comparable to those who were sent to drug court would have been 54.5%.  These drug courts, on average, reduced this rate to 45.5%.  

Generally speaking, drug courts that exclude problem offenders (those with a history of non-compliance or violent offenders) are more successful than those that include these offenders, perhaps because such offenders are difficult to treat by any method that is available to a drug court.  Pre-adjudication drug courts were more effective than post-adjudication courts. “Drug courts are more effective when the benefit of graduation is avoiding conviction rather than avoiding a proscribed sanction” (p. 510).  Some drug courts with certain program requirements – restitution and education – were associated with better outcomes.  On the other hand, drug courts with other sorts of requirements – fines, community service, and employment – were associated with less favourable outcomes. 

Drug courts that require offenders to be in substance abuse treatment for longer periods of time seem to be associated with larger reductions in the rate of reoffending compared to drug courts that do not have such requirements.  Those drug courts requiring attendance at Alcoholics/Narcotics Anonymous were somewhat less effective at reducing recidivism than those that did not require this intervention.  Drug courts that bar offenders who fail a drug test from further participation in the drug court process were less effective than courts that had more flexible responses to failures on drug tests.   Drug courts that used many service providers, as well as drug courts that ran their own services, were more effective than those that used a small number of external service providers.

Conclusions:  Though the drug courts that have been properly evaluated are, on average, somewhat effective in reducing the rate of reoffending by the offenders assigned to them, drug courts themselves vary on a number of different dimensions.  These separate dimensions in combination can  have a non-trivial  impact on the overall effectiveness of a drug court. Though most (78%) of the drug courts in this study reduced recidivism, some drug courts appeared to increase the reoffending of those sent to them.  It would appear that those who are designing drug courts, as well as those requesting funds to continue running existing drug courts, need to be constantly monitoring and evaluating the effectiveness of what they are doing to ensure that their model of providing services is, in fact, the most effective one for the offenders who are assigned to them.

Reference: Shaffer, Deborah Koetzle (2011).  Looking Inside the Black Box of Drug Courts: A Meta-Analytic Review. Justice Quarterly, 38(3), 493-521.

Back to top

Visits by parents to incarcerated youths can reduce the youth’s depressive symptoms even in cases in which the relationship between the parents and the youth is not good.

Youths are especially at risk of experiencing depression when incarcerated because of stress that they experience in the absence of social support.  Youths report high rates of mental illness symptoms on arrival at custodial correctional facilities. Suicide and self-harm are most likely to occur during the early stages of incarceration. 

In general, social support of youths caught up in the youth justice system appears to act as a buffer against mental health problems.  From a policy perspective, however, there is a problem. It is known that family relationships for youths in the youth justice system tend to be especially problematic.  What is not known is whether in situations in which the parent-child relationships are not good, visits from parents are helpful to the youth.

This study examined the effects of parental visits on 276 males, age 14-17 in a secure California juvenile correctional facility.  Youths were interviewed within 48 hours of their arrival at the facility, and then weekly for another 3 weeks, and then after they had been incarcerated for two months.  Depressive symptoms were assessed with a 20-item scale.  The youth’s relationship with his parents was assessed using a ‘warmth and acceptance’ scale with 8 questions (e.g., “He/She lets me know through words and actions that he/she really cares about me.”)  Most visits the youth received were from parents (54%), though about 29% were from brothers or sisters.  In the first week, youths were not allowed visitors. About 12% were never visited by a parent during their first two months in custody. In an average week, about half of the youths had a parental visit. 

Initially, those youths who subsequently received 0, 1, 2, or 3 or more visits per week did not differ in their depressive symptoms.  Over time, however, those youths who received parental visits experienced more rapid decline in their depressive symptoms compared to youths who did not receive visits from parents.  Youths who had good relationships with their parents showed lower levels of depression.  However, visits from supportive parents as well as visits from less supportive parents had similar effects: they lowered levels of depressive symptoms. 

Conclusion: “It appears that any parental visits, regardless of parent-adolescent relationship quality, serve to reduce depressive symptoms during the first two months of incarceration” (p. 150).  Thus it would appear that encouraging parental visits for all youths would be good policy.  This may be especially important in facilities in which parental visits are suspended as punishment for bad behaviour of the youth.  Given that youths who get into trouble may be more likely to be depressed than other youths, it would appear that parental visits may be especially important for them.  Parental visits probably do not occur for a large number of structural reasons related to the situations parents find themselves in. “Given the benefits of prison visitation [by parents] on youth psychological adjustment, it is imperative that policy assists in overcoming barriers to prison visitation such as geographic distance and lack of resources (e.g., finances, time off from work, transportation child care, etc.” (p. 150). 

Reference: Monahan, Kathryn C., Asha Goldweber, and Elizabeth Cauffman (2011). The Effects of Visitation on Incarcerated Juvenile Offenders: How Contact with the Outside Impacts Adjustment on the Inside.  Law and Human Behavior, 35, 143-151.

Back to top

The decision of ordinary people to contact the police in response to crime and disorder depends on the nature of the community in which the problem arises.

It is well established from ethnographic research and from victimization surveys that many people do not call the police when they witness crime or experience victimization.  Calling the police is a strategic decision: it has costs (e.g., risks associated with turning a dispute over to the police, risks associated with contact with police, time associated with being a witness) and it may or may not have any personal or social benefit.  This is a 3-year long ethnographic study of two neighbourhoods in Los Angeles, California. Ethnographic studies are “useful in describing processes that underlie activities like calls for service [which are] the outcomes of processes that are ordinarily hidden from the instruments of conventional research approaches” (p. 139). 

In one neighbourhood that was studied – a high-density, primarily Latino neighbourhood – most residents would be described as the working poor.  They had migrated to the U.S. typically as ‘chains’ such that “each successive arrival is assisted in settling by predecessors…., creating relationships of sponsorship and reciprocity” (p. 140). People were accustomed to relying on others in various ways.  There were “wide-ranging and overlapping networks of communication” that were constantly developing.  Neighbourhood youths were not seen as dangerous, and although they were aware of crime, they seldom called the police, in part because of fear of retribution but also because they empathized with many of those involved in street crime (mainly drug-related crimes).  Although the police considered the neighbourhood to be high crime, residents reported having little fear of personal victimization. Another reason for not calling the police was that many were not yet legal immigrants.  Hence they sought other ways of dealing with problems within the neighbourhood rather than calling the police. Only the most serious matters lead to the police being called, and then, residents would often attempt to make the contact indirectly (e.g., through an apartment manager).  The police were generally seen as not being very useful, in that they were perceived as being unresponsive even when called. 

In the other location, residents came from a mixture of backgrounds and language groups, thus linguistically they were isolated from the majority of their neighbours.  The various groups identified their neighbourhood in different ways that were specific to their own ethnic backgrounds.  And the demographic character and lifestyles of the different groups were quite different.  Because the different groups had different lifestyles, the kinds of easy social relationships among all residents that existed in the first neighbourhood were absent in the second. Channels of communication operated along group lines and rarely crossed group lines.  Crime, then, especially for whites (23% of the population) was an important issue. Whites saw the police as public servants whose purpose was to deal with local issues. Calls to the police were often coordinated through their “Neighbourhood Watch” organization. Many calls related to ‘quality of life issues’ (e.g., noisy neighbours).  Other groups typically saw less reason to call the police. 

Conclusion: The idea behind ‘community policing’ is that the police should be responding to community problems and needs.  The risk, as demonstrated by the difference between these two neighbourhoods, is that the term ‘community’ may be invoked without knowledge of the range of variation in neighbourhoods. “In socially heterogeneous areas, where ‘community’ frequently means conflicting rather than common interests” (p. 159) responding to ‘the community’ can be problematic.  The differences between these two neighbourhoods in the manner in which they call in the police demonstrates that the ‘problems’ that the police hear about are filtered in quite different ways in the two neighbourhoods.

Reference: Ibarra, Peter R. (2003). Contacts with the Police: Patterns and Meanings in a Multicultural Realm.  Police & Society, Issue No. 7, 135-166.

Back to top

Publication Type

Volume Number


Issue Number