I.Research question: Are Aboriginal youth handed harsher sentences than Non-Aboriginal
youth, under the Young Offenders Act.
II. Summary of Research Article’s Argument
Latimer, J., & Laura, C. F. (2005). The sentencing of aboriginal and non-aboriginal youth under
the young offenders act: A multivariate Analysis1. Canadian Journal of Criminology and
Criminal Justice, 47(3), 481-500. Retrieved from
http://ra.ocls.ca/ra/login.aspx?inst=centennial&url=?url=https://search-proquest-com.ezce
ntennial.ocls.ca/docview/216098659?accountid=39331
In their article “The sentencing of aboriginal and non-aboriginal youth under the young
offender’s act: A multivariate Analysis 1” from the Criminal Justice Database, Latimer and
Laura (2005) attempt to pinpoint why Aboriginal youth receive higher sentencing than nonAboriginal youth. Using the data from five major cities within Canada, the Department of Justice
Canada noted that Aboriginal youth were incarcerated “at a rate 8 times that of non-Aboriginal
youth” (Latimer & Laura, 2005) and that statistic was higher in other provinces in Canada. They
also noted that there is a mass representation of Aboriginal youth within the justice system
because they outnumber non-Aboriginal youth in the age group that has the highest risk for
offending as well as being from a low socioeconomic background. There has always been a
strong correlation between poverty and the rate of crime. Therefore, it comes as no surprise that
marginalized Aboriginal youth are more likely to be caught in the justice system due to the lack
of education and employment rate and greater involvement in gangs, victimization and substance
abuse. Another factor that was worth noting was discrimination against the Aboriginal
community, especially during bail and sentencing hearings. Aboriginal youth were given
disproportionate sentences for their crimes, in comparison to their Caucasian counterparts.
This paper investigates the issue of discrimination against Aboriginal youth under the
Young Offenders Act during sentencing within five major cities in Canada. The study considers
variables such as previous findings of guilt, actual or threatened use of violence, reoffending
while on probation and multiple charges. Using these variables, the study set out to answer
discriminative questions surrounding the sentencing of Aboriginal youth. After conducting their
study, they concluded that Aboriginal youth are over represented in the correctional system due
to various discourses and intersections. For example, some contributing factors were historical
influences, socioeconomic status, substance abuse and so on. These contributing factors played a
role in determining whether or not a youth would be handed a lengthy sentence or not.
III. Critical Reflection
This study was very thorough in gathering information that proved that Aboriginal youth
do indeed receive higher sentencing rates. The different variables did cause the judges to issue
harsher penalties towards Aboriginal youth. However, discrimination was a secondary factor
when they considered punishment. Judges still focused on the youth’s record and due to the other
contributing factors, Aboriginal youth did reoffend at a higher rate than non-Aborignal youth,
under the YOA. One improvement, this study could have compared their results by collecting
data after the YOA switched to the YCJA to see if there was any conflicting data.
Article
The Sentencing of Aboriginal and Non-Aboriginal Youth under the Young Offenders Act:
A Multivariate Analysis1
Author: Latimer, Jeff; Laura Casey Foss
Publication info: Canadian Journal of Criminology and Criminal Justice ; Toronto Vol. 47, Iss. 3, (Jul
2005): 481-500.
ProQuest document link
Abstract:
This paper examines the sentencing of Aboriginal and non-Aboriginal youth under the Young Offenders
Act, using youth court data from five major cities in Canada. Three separate sentencing decisions are
examined using logistic and multiple regression analyses: (1) the imposition of a custodial sentence
versus a non-custodial sentence, (2) the use of secure custody versus open custody, and (3) the length of
the custodial sentence. First, there is no evidence that Aboriginal youth are more likely than nonAboriginal youth to receive a custodial sentence. Sentencing judges appear to base the decision to
impose custody on traditional factors such as criminal history and the seventy of the offence. Second,
there is no convincing evidence to suggest that Aboriginal youth are more likely than non-Aboriginal youth
to receive a secure custody sentence. Lastly, there is evidence to suggest, however, that Aboriginal youth
are likely to receive a longer custodial sentence than non-Aboriginal youth, regardless of standard
aggravating factors such as criminal history and offence severity. [PUBLICATION ABSTRACT]
Links:Search for full text
Full text:
Headnote
This paper examines the sentencing of Aboriginal and non-Aboriginal youth under the Young Offenders
Act, using youth court data from five major cities in Canada. Three separate sentencing decisions are
examined using logistic and multiple regression analyses: (1) the imposition of a custodial sentence
versus a non-custodial sentence, (2) the use of secure custody versus open custody, and (3) the length of
the custodial sentence. First, there is no evidence that Aboriginal youth are more likely than nonAboriginal youth to receive a custodial sentence. Sentencing judges appear to base the decision to
impose custody on traditional factors such as criminal history and the seventy of the offence. Second,
there is no convincing evidence to suggest that Aboriginal youth are more likely than non-Aboriginal youth
to receive a secure custody sentence. Lastly, there is evidence to suggest, however, that Aboriginal youth
are likely to receive a longer custodial sentence than non-Aboriginal youth, regardless of standard
aggravating factors such as criminal history and offence severity.
Dans le présent article, on examine les peines imposées aux jeunes autochtones et non autochtones en
vertu de la Loi sur les jeunes contrevenants en exploitant les données des tribunaux de la jeunesse de
cinq grandes villes canadiennes. Plus précisément, on étudie trois prononcés de sentence distincts en
menant des analyses de régression logistique et multiple : (T) l'imposition d'une peine de garde par
rapport à l'imposition d'une peine non carcérale; (2) l'imposition d'une peine de garde en milieu fermé par
rapport à une peine de garde en milieu ouvert; et (3) la durée de la peine de garde. Or, les résultats
démontrent, tout d'abord, que les jeunes autochtones ne sont pas plus susceptibles que les non
autochtones de se faire infliger une peine de garde. En fait, les juges semblent fonder l'imposition de la
peine de garde sur des facteurs classiques tels que le dossier criminel et la gravité de l'infraction.
Ensuite, il n'existe aucune preuve convaincante qui indiquerait que les jeunes autochtones soient plus
susceptibles que les autres de se faire infliger une peine de garde en milieu fermé. Enfin, les données
semblent indiquer toutefois que les jeunes autochtones seraient susceptibles de se faire infliger une
peine de garde de plus longue durée que les autres jeunes, et cela, indépendamment des facteurs
aggravants classiques (antécédents criminels, gravité du crime, etc.).
Introduction
Recent research from the Department of Justice Canada has indicated that Aboriginal youth in Canada
were incarcerated at a rate 8 times that of non-Aboriginal youth (Latimer and Foss 2004). The problem
was particularly acute in northern and central Canada. In Saskatchewan, for example, Aboriginal youth
were 30 times more likely to be incarcerated than non-Aboriginal youth, and in the Yukon, Aboriginal
youth were 18 times more likely to be incarcerated than non-Aboriginal youth.
Roberts and Melchers (2003) have noted that the problem of Aboriginal over-representation in the
criminal justice system in Canada has been recognized in all principal correctional texts for years and is
widely acknowledged in the general population. During the last decade, several noteworthy governmental
initiatives have attempted to address the issue of Aboriginal over-representation, including the Royal
Commission on Aboriginal Peoples (1996) and the Standing Senate Committee on Aboriginal Peoples
(2003). Despite the recognition and attention, the problem has persisted for decades.
Why is this the case? A number of researchers have attempted to understand and explain this
phenomenon. Boe (2002), for example, has argued that over-representation is due, in part, to population
demographics. The Aboriginal population, in general, is much younger than the non-Aboriginal population
in Canada. Consequently, there is a higher proportion of Aboriginals than of non-Aboriginals in the age
group at highest risk for offending (i.e., less than 30 years of age).
La Prairie (1992, 2002) has maintained that, in addition to demographics and historical issues,
Aboriginals are over-represented largely as a result of their low socio-economic status. Mainstream
criminological theories have consistently linked poverty to a lack of education, employment, and criminal
behaviour. Even within Aboriginal communities, it remains true that the more socially and economically
marginalized the group, the higher the involvement in the criminal justice system.
Latimer and Foss (2004) found that a substantial proportion of Aboriginal youth in custody self-reported
high rates of victimization, substance abuse, and organized gang participation. These factors are widely
recognized as correlates of criminal behaviour (Latimer, Kleinknecht, Hung, and Gabor 2003; Andrews
and Bonta 1998). In other words, a history of child maltreatment, drug and alcohol abuse, and
involvement with antisocial peers places an individual at a significant risk for engaging in criminal
behaviour.
Ultimately, at the root of Aboriginal over-representation rests a complex set of inter-related factors. One
factor, however, that has not been well researched, particularly among youth, is discrimination at
sentencing. Are Aboriginals treated differently (e.g., more harshly) by sentencing judges in youth court
than non-Aboriginal youth? And is it possible that this may contribute to their over-representation?
Roberts and Melchers (2003) argued that discrimination at sentencing does not appear to be one of the
causes of over-representation. La Prairie and Stenning (2003) also raised serious questions about the
claim that over-representation is attributable to culturally insensitive and discriminatory practices. In
Australia, Weatherburn, Fitzgerald, and Hua (2004) have also argued that the leading current cause of
Aboriginal over-representation in prison is not discrimination but high rates of Aboriginal involvement in
serious crime.
Previous research, however, has suggested that discrimination against Aboriginals may exist across
several points in the Canadian criminal justice system, including at bail hearings (Aboriginal Justice
Inquiry of Manitoba 1991; Moyer 2004) and sentencing hearings (Schissel 1993). Latimer and Foss
(2004), using a snapshot method of counting all youth in custody on a given day, found that the median
custodial sentence length for Aboriginal youth in custody was 212 days, while the median sentence length
for non-Aboriginal youth in custody was only 182 days. Although it was not possible to control for factors
typically considered in sentencing (e.g., criminal history, severity of the offence), this finding suggests that
multivariate analysis needs to be conducted to determine why there was such a difference. Furthermore,
previous research into Aboriginal over-representation has focused primarily, and most times exclusively,
on adult offenders. Therefore, discrimination towards Aboriginal youth should not be ruled out as a related
factor, without sound empirical research.
Present research
The present research addresses the issue of discrimination against Aboriginal youth during sentencing.
Specifically, this paper examines the sentencing under the Young Offenders Act (YOA) of Aboriginal and
non-Aboriginal youth in five cities in Canada.
There are three quantitative indicators in the youth criminal justice system that may provide evidence of
possible discrimination at sentencing. The first indicator is the overall use of custodial versus noncustodial sentencing options. The second indicator is the use of secure custody versus open custody. In
the youth criminal justice system, secure custody is typically more restrictive, akin to an actual prison,
while open custody is usually similar to a group home in a community setting. The third indicator is the
actual length of custodial sentences. Young and Brown (1993) have argued that sentence length is one of
the leading contributors to high incarceration rates.
In order to compare the sentencing of Aboriginal and non-Aboriginal youth across these three indicators,
it is important to control for various factors that judges consider in determining a sentence. Manson
(2001) has provided a general description of aggravating and mitigating factors typically recognized by
judges. These include the following:
1. previous findings of guilt
2. actual or threatened use of violence
3. offending while subject to release or probation conditions
4. multiple charges
In the area of sentencing, it is also generally accepted that there exists a phenomenon of "stepping up"
sentences in youth court, which Carrington and Moyer (1995) labelled the escalation of control measures.
In other words, with each subsequent finding of guilt, a judge will increase the severity of the sentence.
For example, a prior custodial sentence has been strongly correlated with a new current custodial
sentence (Carrington and Moyer 1995; Moyer 1996). Therefore, previous sentences may play a
significant role in current sentencing decisions.
In addition, previous research has also found that pre-trial detention is a factor in sentencing decisions in
youth court (Moyer 2004). Youth held in custody until sentencing were more likely to receive a custodial
sentence than youth who were released after arrest, even when controlling for other factors, such as
offence severity and criminal history.
Since it is generally recognized that the criminal justice system needs to respond to youth in a manner
distinct from the traditional adult response, age may also be a factor in sentencing. A 12-year-old youth
would very likely be viewed differently than a 17-year-old youth by a sentencing judge. This may also be
the case for gender. A male youth might be treated differently than a female youth.
Method
In order to examine possible discrimination against Aboriginal youth at sentencing, the research was
designed to answer the following three questions:
1. Are Aboriginal youth more likely than non-Aboriginal youth to receive custody?
2. Are Aboriginal youth more likely than non-Aboriginal youth to receive a secure custody sentence?
3. Are Aboriginal youth more likely than non-Aboriginal youth to receive a longer custodial sentence?
In order to answer question one, logistic regression was selected with custody (yes/no) as the dependent
variable and the following independent variables:
1. age
2. gender
3. Aboriginal status
4. prior findings of guilt
5. prior custodial sentence
6. seriousness of current offence3
7. violence used in current offence
8. number of current charges
9. subject to release/probation conditions at time of offence
10. current administration of justice offence
11. detained on current offence prior to sentencing
For question two, logistic regression was also selected, using the same 11 independent variables but with
custody level (open/secure) as the dependent variable. In order to answer question three, multiple
regression was selected, using the same 11 independent variables but with the length of the custodial
sentence as the dependent variable.
Sample
The sample consisted of 1,321 youth court cases with a finding of guilt, in five large cities in Canada, from
April 1, 1999, to March 31, 2000. The cases were randomly sampled from youth courts in Halifax (18.7%),
Toronto (21.7%), Winnipeg (20.5%), Edmonton (17.6%), and Vancouver/Surrey (21.6%).4
More than three quarters of the sample were male (79.7%), while 20.3% of the sample were female. The
age of the sample at sentencing ranged from 12 to 20 years, with a mean age of 15.85 years (SD = 1.48).
Aboriginal status was recorded for each youth, based upon information recorded within official court,
police, Crown, probation, or custody records. Approximately one quarter (23.8%) of the youth were
Aboriginal, whereas 76.2% of the youth were non-Aboriginal.
Results
Question 1: Are Aboriginal youth more likely than non-Aboriginal youth to receive custody?
Table 1 provides the number and proportion of cases that received custody, by gender, age, and
Aboriginal status. It appears that there was a significant relationship between Aboriginal status and
custody (χ^sup 2^ (1, n = 1,321) = 7.27, p < 0.01). There was also a significant relationship between
gender and custody (χ^sup 2^ (1, n - 1,321) = 9.21, p < 0.01), and a significant curvilinear relationship
between age groupings and custody (χ^sup 2^ (2, n = 1,321) = 16.03, p < 0.001).
Enlarge this image.
Table 2 provides the results of the logistic regression analysis for predicting custody. The Enter method
was used, which builds the equation by entering all of the variables at once. While a true R^sup 2^ cannot
be calculated with a logistic regression, a Pseudo R^sup 2^ provides a general estimate of the
performance of the model (i.e., the percentage of variance explained by the independent variables in the
regression). The Pseudo R^sup 2^ for the model (0.2910) indicated that the model explained
approximately 29% of the variance in the use of custody for youth with a finding of guilt.
Aboriginal status did not predict the use of custody as a sentence for youth. In fact, out of the 11
independent variables, only Aboriginal status and age did not enter the model as significant predictors.
The primary factors related to a custodial sentence were criminal history (i.e., prior custodial sentence
and prior findings of guilt), the use of pre-trial detention (i.e., detained on current offence prior to
sentencing), the seriousness of the current offence (i.e., number of current charges, violence used in
current offence, seriousness of current offence), and not complying with past sanctions (i.e., current
administration of justice offence and subject to release/probation conditions at time of offence). Male
youth were also 1.7 times more likely to receive custody than female youth.
Enlarge this image.
In order to examine the possibility of multi-collinearity, variance inflation factors (VIFs) were calculated for
each predictor. The VIF ratio coefficient is directly related to the predictability of a particular independent
variable by other independent variables present in the model. The highest VIF was estimated at 1.63 for
the variable measuring a prior finding of guilt. This value, however, was well below the standard threshold
and therefore remedial measures were not warranted.
Question 2: Are Aboriginal youth more likely than non-Aboriginal youth to receive a secure custody
sentence?
Table 3 provides the number and proportion of cases that received open and secure custody sentences,
by gender, age, and Aboriginal status. It appears that there was no relationship between Aboriginal status
and custody level or between gender and custody level. There was a linear relationship, however,
between age groupings and custody level (χ^sup 2^ (2, n = 405) = 5.96, p < 0.05).
Enlarge this image.
Table 4 provides the results of the logistic regression analysis for predicting secure custody among youth
who received a custodial sentence. If a youth received both open and secure custody, the case was
coded as a secure custody case. As with the first logistic regression, the Enter method was used. The
Pseudo R^sup 2^ was relatively low (0.1045) and explained approximately 10% of the variance in the use
of secure custody versus open custody for youth with custodial sentences.
The primary factors related to the use of secure custody were number of current charges, a prior custodial
sentence, being subject to conditions of release/probation at the time of the current offence, and being
detained prior to sentencing.
According to the odds ratio, an Aboriginal youth was 1.6 times more likely to receive a secure custody
sentence than a non-Aboriginal youth, regardless of his or her criminal history or the severity of the
offence. However, the probability (p = 0.0734) was slightly higher than what is considered statistically
significant under conventional standards (i.e., p < 0.05).
Enlarge this image.
Question 3: Are Aboriginal youth more likely than non-Aboriginal youth to receive a longer custodial
sentence?
Table 5 provides information on the relationship between demographic characteristics and the length of
custodial sentences. In terms of Aboriginal status, the median sentence length for Aboriginal youth was
90 days, while the median sentence length for non-Aboriginal youth was 35 days. Using sentence-length
groupings, this difference was statistically significant (χ^sup 2^ (2, n = 403) = 9.53, p < 0.01). The
relationship between gender and sentence-length groupings was also significant (χ^sup 2^ (2, n = 403) =
13.55, p < 0.01). This was not the case with age groupings and sentence-length groupings.
Table 6 provides the results of the multiple regression for predicting the length of a custodial sentence.
The distribution of the sentence length, in days, was negatively skewed, since most of the sentences
were relatively short. In order to reduce skewness and reduce the effect of very long sentence lengths,
the square roots of the sentence lengths in days were calculated and entered into the model as the
dependent variable. The Enter method was again used in this regression. The R^sup 2^ for the model
(0.2177) explained approximately 22% of the variance in sentence lengths for youth with custodial
sentences.
Enlarge this image.
Aboriginal status entered the model as a significant predictor of sentence length. The primary factors
related to sentence length were the seriousness of the offence (i.e., number of current charges,
seriousness of current offence, violence used in current offence), the use of pre-trial detention (i.e.,
detained prior to sentencing), gender (i.e., being male) and Aboriginal status (i.e., being an Aboriginal).
Discussion
Do the results of these regression analyses reveal discrimination against Aboriginal youth in the youth
criminal justice system? It is too early to declare the existence of a "smoking gun." However, the results
do reveal the possibility of discrimination against Aboriginal youth at sentencing.
When the use of general custody was examined, there was no difference between Aboriginal and nonAboriginal youth. In other words, when a sentencing judge in youth court was considering a custodial
sentence, the fact that the youth was Aboriginal was clearly not relevant.5 Typical factors such as prior
record and offence severity were the primary considerations. However, after the primary decision to
impose a custodial sentence has been made, two important secondary decisions need to be made in the
youth criminal justice system - custodial level and sentence length.
Enlarge this image.
In the first instance (i.e., custodial level), are Aboriginal youth more likely than non-Aboriginal youth to
receive secure custody rather than open custody? The odds ratio indicated that Aboriginal youth were 1.6
times more likely than non-Aboriginal youth to be sentenced to secure custody, regardless of their
criminal history or the severity of their offence. Some might argue that this difference, although not
statistically significant, should nonetheless be viewed seriously. Hood (1995), for example, in defence of
the use of a probability level higher than .05, quotes Henkel's (1976) claim that "the choice of a
significance level ... is an arbitrary decision since neither substantive theory nor statistical theory dictate
that a particular level of significance be used" (77-78). The fact that the results would occur by chance
once in 14 times rather than the standard once in 20 times should not be a basis to completely dismiss
these findings.
With that being said, the Pseudo R^sup 2^ was relatively low and did not explain much of the variance in
the use of secure custody. This indicates that other factors, which were not entered into the model, would
likely explain more variance. One potential factor is a youth's level of risk measured by an actuarial
risk/need tool. Aboriginal youth may pose a greater statistical likelihood of reoffending compared to nonAboriginal youth, based upon certain factors, such as family dysfunction, antisocial peer involvement, and
substance abuse. A second factor that may influence the decision is prior custodial experiences, such as
escapes, punitive transfers between open/secure custody (i.e., s. 24(2)(9) of the YOA), and general
institutional misconduct. These variables were not collected and, consequently, were not entered into the
model.
The evidence in this case is therefore not compelling. As Roberts (2003) has suggested, when serious
concerns that have the potential to erode public confidence in the justice system are investigated, the
research must be impeccable. The probability, while reasonable, was not above the standard threshold
and the ability of the overall model to explain the variance in the use of secure custody was weak.
In the second instance (i.e., sentence length), the results of the regression analysis revealed evidence of
possible discrimination. Aboriginal youth in the sample received longer custodial sentences than did
nonAboriginal youth, regardless of their criminal history or the severity of their offence. The probability
was well above the standard threshold and the ability of the overall model to explain the variance in
sentence lengths was moderate. While it is possible that there are other factors that could be considered
in this model (e.g., risk level and previous custodial experiences), this evidence should not be dismissed
as quickly as the previous finding, particularly as sentence length is a crucial factor in understanding
incarceration rates.
In order to link sentence lengths and over-representation, it is useful to consider the concepts of stock
and flow. Stock refers to the existing population and flow refers to both admissions into custody and
releases out of custody. Sentence lengths have the ability to affect the overall stock, even if admission
rates are held constant. Consider the following equation, taken from Butts and Adams (2001), which is
used to forecast custody populations:
Enlarge this image.
If everything is held constant except sentence length, the stock population increases in direct relationship
with sentence length. If we assume, for example, that the original population was 5,000 youth and there
were 5,000 admissions in a year, with an average sentence length of 35 days, the stock population would
increase to 5,475 youth by the end of the year. If we increased the average sentence length to 90 days,
the stock population would increase to 6,209 youth. Over the course of years, it is, therefore, possible that
longer sentence lengths have played a role in the over-representation of Aboriginal youth in custody.
While this evidence cannot be easily dismissed, it raises a question: Why is it that there may be
discrimination at one stage of the youth criminal justice system (i.e., sentence length) and not at another
stage (i.e., imposing custody)? Could it be that other factors, such as risk level, explain the variance in
sentence length? And that Aboriginal status would fall out of the model if these variables were included?
It is possible. Bonta, La Prairie, and Wallace-Capretta (1997), using a sample of adult probationers,
reported that Aboriginal offenders demonstrated higher scores than non-Aboriginal offenders on an
actuarial risk/need instrument. Furthermore, Finn, Trevethan, Carrière, and Kowalski (1999) also reported
that incarcerated Aboriginal offenders were considered to be at a higher risk to reoffend than nonAboriginal incarcerated offenders. In other words, Aboriginal offenders present with a higher number of
factors that have been correlated with recidivism (e.g., employment status, alcohol/drug use, antisocial
peer involvement) compared to non-Aboriginal offenders.
Aboriginal youth in the current sample were more likely than nonAboriginal youth to present with serious
issues, such as alcohol and drug abuse, gang involvement, Fetal Alcohol Spectrum Disorder, and
depression/suicidal thoughts. Furthermore, the Aboriginal youth in the sample were more likely than nonAboriginal youth to be out of school, unemployed, and not living with their families. Unfortunately, this
information was missing for a high number of Aboriginal and non-Aboriginal youth and could not,
therefore, be entered into the regression models. Nonetheless, the available data lend support to the
notion that judges may be sentencing Aboriginal youth to longer sentences based upon risk/need
principles.
Assuming that an Aboriginal youth does present before the courts as a higher risk to reoffend, based
upon any number of criteria (e.g., actuarial tools, clinical judgement, pre-sentence report), does this
provide an adequate rationale for imposing a longer sentence? There are two ways of looking at this
question. First, there is the possibility that longer sentences are imposed in order to address the need
aspect of risk/need scores (i.e., for rehabilitation). While the YOA did not specifically address this issue,
the Youth Criminal Justice Act (YCJA) provides clear direction. Under the new legislation, the courts can
and should address the rehabilitative needs of youth, but only within the limits of proportionality
(Barnhorst 2004). Therefore, longer sentence lengths based upon high needs are not justifiable. The
second possibility is that longer sentences are imposed in order to address the risk aspect of risk/need
scores (i.e., a high likelihood of committing a new offence). This would assume, however, that both the
circumstances of the current offence and the likelihood of future criminal behaviour should be considered
a component of proportionality. Unfortunately, this is a legal question that is well beyond the scope of this
current research. Assuming that risk should not be a component of proportionality, there does not appear
to be a rationale for imposing longer sentences based on risk/need principles.
For two central reasons, this research should be replicated using data collected since the implementation
of the YCJA. First, assuming that the risk/need hypothesis is correct, the emphasis on proportionality
under the YCJA will likely affect the sentencing practices of youth court judges, particularly with Aboriginal
youth. Second, in 1996, the Criminal Code was amended to include section 718.2(e), which states that all
available sanctions other than imprisonment should be considered for all offenders, with particular
attention to the circumstances of Aboriginal offenders. In 1999, the Supreme Court of Canada (R. v.
Gladue) provided an interpretation of section 718.2(e) and concluded that custodial sentences for
Aboriginal offenders may be shorter than sentences of non-Aboriginals for the same offence. The YCJA
includes a provision similar to section 718.2(e) of the Criminal Code that was not contained in the YOA.
Section 38(2)(d) of the YCJA essentially states that all available sanctions other than custody should be
considered for all young persons, with particular attention to the circumstances of Aboriginal young
persons. Therefore, this section may also affect sentencing deliberations for Aboriginal youth.
There are several issues that this research did not consider. First, there is the possibility of differences
across regions. In other words, longer sentences for Aboriginal youth may be localized to a few provinces
or even to a few courthouses. While the sample size was too small to run separate regressions for
questions two and three, this was possible for question one. The results for the individual regressions by
courthouse for question one were not different than for the overall model. In other words, Aboriginal
status was not a predictor of custody in any of the five sites. While it is possible that regional variations in
the use of secure custody and sentence lengths might affect the results for questions two and three, one
could also assume that, since the individual regressions for question one were the same as for the overall
regression, this would be the case with questions two and three.
Second, this research did not consider the possibility of discrimination at other stages in the youth
criminal justice system. For example, police decision making (i.e., the decision to lay or recommend
formal charges) was not examined. Clearly, this has the ability to determine which cases are even
brought before the courts for trial and sentencing. Furthermore, decisions, such as sentence length, may
be partially based on sources of information beyond the control of sentencing judges, such as pre-
sentence reports or testimony in court. It is possible that these information sources may, at times, portray
Aboriginal youth in a manner distinct from that for non-Aboriginal youth.
Conclusion
Aboriginal youth are clearly over-represented in the youth correctional system. The causes of this
phenomenon are quite complex and interrelated and include historical influences, demographics, poverty,
victimization, and substance abuse. The present research provides conflicting evidence when
discrimination at sentencing is examined as a possible factor. In simple terms, sentencing judges appear
to base the decision to impose custody on traditional factors such as the criminal history of the youth and
the severity of her or his offence.
Ruling out discrimination at the secondary decision stage is not as easy. In terms of the level of custody,
the results are not convincing and it is better to err on the side of caution. In terms of custodial sentence
lengths, however, there is evidence to suggest that Aboriginal youth are more likely than non-Aboriginal
youth to receive a longer sentence, regardless of standard aggravating factors such as criminal history
and offence severity. This may be in response to the fact that Aboriginal youth present with higher
risk/need levels and not due to their Aboriginal status per se. Under the YCJA, however, this rationale
would not be justifiable.
Future research should replicate the present study using data collected since the implementation of the
YCJA and compare the results across the two time periods. In addition, increasing the number of
independent variables in the multivariate analysis to include factors such as needs, risk level, and
previous custodial experiences would be valuable. Finally, additional research that examines the
possibility of discrimination at other points in the criminal justice system would further our understanding
of this issue.
Footnote
Notes
1. The authors would like to gratefully acknowledge several individuals who provided assistance on this
project. First, we would like to thank Sharon Moyer for supervising the data collection and data entry
process across the country and Paul Verbrugge for creating the research database used in this paper.
Second, we would like to acknowledge Dr Fernando Mata, who provided statistical advice on this project.
Finally, we are grateful to three anonymous reviewers who provided valuable comments on an earlier
draft of this paper.
2. The opinions expressed herein are solely those of the author and do not necessarily represent the
views of the Department of Justice Canada.
3. The seriousness of the current offence was determined using the Seriousness Index developed by the
Canadian Centre for Justice Statistics, Statistics Canada, which ranks offences from most to least serious
based upon sentence lengths and potential harm to victims.
4. Percentages do not total 100% due to rounding.
5. We acknowledge, however, that some might also consider this to be problematic. By ignoring
Aboriginal status, sentencing judges are not considering the special circumstances of Aboriginal
offenders according to parliamentary (e.g., s. 38(2)(d) of the YCJA) and judicial direction (e.g., R. v.
Gladue).
References
References
Aboriginal Justice Inquiry of Manitoba
1991 Report of the Aboriginal Justice Inquiry of Manitoba. Winnipeg, MN: Government of Manitoba.
Andrews, Don A. and James Bonta
1998 Psychology of Criminal Conduct, 2nd ed. Cincinnati, OH: Anderson Publishing.
Barnhorst, Richard
2004 The Youth Criminal Justice Act: New directions and implementation issues. Canadian Journal of
Criminology and Criminal Justice 46: 232-250.
Boe, Roger
2002 Future demographic trends may help Canada's Aboriginal youth. Forum on Corrections Research
14:13-16.
Bonta, James, Carol La Prairie, and Suzanne Wallace-Capretta
1997 Risk prediction and re-offending: Aboriginal and non-Aboriginal offenders. Canadian Journal of
Criminology 39:127-144.
Butts, Jeffrey A. and William Adams
2001 Juvenile Justice Bulletin: Anticipating Space Needs in Juvenile Detention and Correctional Facilities.
Washington, DC: Office of Juvenile Justice and Delinquency Prevention.
Carrington, Peter J. and Sharon Moyer
1995 Factors affecting custodial dispositions under the Young Offenders Act. Canadian Journal of
Criminology 37:127-162.
Finn, Anne, Shelly Trevethan, Gisèle Carrière, and Melanie Kowalski
1999 Female Inmates, Aboriginal Inmates, and Inmates Serving Life Sentences: A One-Day Snapshot.
Ottawa: Canadian Centre for Justice Statistics, Statistics Canada.
Henkel, Ramon E.
1976 Tests of Significance. Thousand Oaks, CA: Sage Publications.
Hood, Roger
1995 Race and sentencing: A reply. Criminal Law Review (April): 271-279.
La Prairie, Carol
1992 Aboriginal Over-Representation in Correctional Institutions and Implications for Crime Prevention.
Ottawa: Solicitor General Canada.
La Prairie, Carol
2002 Aboriginal over-representation in the criminal justice system: A tale of nine cities. Canadian Journal
of Criminology 44:181-208.
La Prairie, Carol and Philip Stenning
2003 Exile on Main Street: Some thoughts on Aboriginal over-representation in the criminal justice
system. In D. Newhouse and E. Peters (eds.), Not Strangers in These Parts: Urban Aboriginal Peoples.
Ottawa: Policy Research Initiative.
Latimer, Jeff and Laura Casey Foss
2004 A One-Day Snapshot of Aboriginal Youth in Custody across Canada: Phase II. Ottawa: Research
and Statistics Division, Department of Justice Canada.
Latimer, Jeff, Steven Kleinknecht, Kwing Hung, and Thomas Gabor
2003 The Correlates of Self-Reported Delinquency: Analysis of the National Longitudinal Survey of
Children and Youth. Ottawa: Research and Statistics Division, Department of Justice Canada.
Manson, Allan
2001 The Law of Sentencing. Toronto: Irwin Law.
Moyer, Sharon
1996 A Profile of the Juvenile Justice System in Canada. Ottawa: Federal Provincial-Territorial Task
Force on Youth Justice.
Moyer, Sharon
2004 Pre-Trial Detention under the Young Offenders Act: A Multi-Site Study of Urban Courts. Ottawa:
Research and Statistics Division, Department of Justice Canada.
Roberts, Julian
2003 Introduction: Commentaries on policing in Toronto. Canadian Journal of Criminology and Criminal
Justice 45, 343-346.
Roberts, Julian and Ronald Melchers
2003 The incarceration of Aboriginal offenders: Trends from 1978 to 2001. Canadian Journal of
Criminology and Criminal Justice 45: 211-243.
Schissel, Bernard
1993 A sociolegal analysis of Canadian youth justice: The impact of offender's race on judicial decisions.
Journal of Criminal Justice 21: 533-552.
Weatherburn, Don, Jackie Fitzgerald, and Jiuzhao Hua
2004 Reducing Aboriginal over-representation in prison. Australian Journal of Public Administration 62:
65-73.
Young, Warren and Mark Brown
1993 Cross-national comparisons of imprisonment. In M. Tonry (ed.), Crime and Justice: A Review of
Research, vol. 17. Chicago, IL: University of Chicago Press.
Cases cited
R. v. Gladue, [1999] 1 S.C.R. 688.
AuthorAffiliation
Jeff Latimer2
Senior Research Officer Research & Statistics Division Department of Justice Canada
Laura Casey Foss
Research Assistant Research & Statistics Division Department of Justice Canada
Subject: Juvenile offenders; Court records; Regression analysis; Imprisonment; Criminal justice; Native
North Americans
Location: Canada
Publication title: Canadian Journal of Criminology and Criminal Justice; Toronto
Volume: 47
Issue: 3
Pages: 481-500
Number of pages: 20
Publication year: 2005
Publication date: Jul 2005
Publisher: University of Toronto Press
Place of publication: Toronto
Country of publication: Canada, Toronto
Publication subject: Criminology And Law Enforcement, Education
ISSN: 17077753
CODEN: CJCCD4
Source type: Scholarly Journals
Language of publication: English
Document type: Feature
Document feature: Tables Equations References
ProQuest document ID: 216098659
Document URL: http://ra.ocls.ca/ra/login.aspx?inst=centennial &
url=?url=https://search.proquest.com/docview/216098659?accountid=39331
Copyright: Copyright Canadian Criminal Justice Association Jul 2005
Last updated: 2016-02-20
Database: ProQuest Central
Bibliography
Citation style: APA 6th - American Psychological Association, 6th Edition
Latimer, J., & Laura, C. F. (2005). The sentencing of aboriginal and non-aboriginal
youth under the young offenders act: A multivariate Analysis1. Canadian Journal of
Criminology and Criminal Justice, 47(3), 481-500. Retrieved
from http://ra.ocls.ca/ra/login.aspx?inst=centennial&url=?url=https://search.proquest.co
m/docview/216098659?accountid=39331
Contact ProQuest
Database copyright © 2019 ProQuest LLC. All rights reserved. - Terms and Conditions
Research Article Analysis (20%)
F2019
I. Select a contemporary Canadian research article from a refereed journal
In this assignment, you work independently. Your task is to critique an argument made by a researcher
or a team of researchers who have written a research article on a contemporary Canadian field-related
issue. The article must be one other than the ones that you and your group worked on in your Literature
Review. The issue addressed by the researcher(s) needs to fall within the original research question that
you posed for your Literature Review. As a result, you will continue the process of building focused
knowledge and thinking critically about arguments on an issue of interest in the field. You may even find
it useful to cite evidence from one or more articles used in your Literature Review to support your
critique of the research article for this assignment, though this is not required.
II. Plan your analysis
Your analysis of the argument advanced in the research article that you have selected needs to be
critical. In planning your analysis, make use of as many relevant skills as you like from our lessons (and
from Browne and Keeley’s Asking the Right Questions, if you like). Consider, for instance, any significant
ambiguity in the argument’s language (its key terms), its assumptions, any fallacies it contains, its use of
various sorts of evidence, any faults in its treatment of causes, its use of statistics, any place(s) where
significant information is omitted, anywhere that alternative conclusions are overlooked, and other
possible tactics for questioning, challenging, and critically appraising the argument. Ultimately, ask
yourself this: “On what basis could this argument be rejected?” or “Even if I were to accept this
argument, how could it be refined and improved?”
Organize your main points into a thesis statement that addresses the article’s argument. Your thesis
needs to express your acceptance or rejection of the article’s argument on the basis of a set of critical
points that you intend to make about it. Summarizing the article’s essential argument briefly within your
introductory paragraph will allow you to set up the analysis of it that follows. Your summary should
identify the argument’s issue, conclusion and reasons, but not the details of its evidence and style since
these take us beyond summarizing. The body of your analysis should be devoted to making critical
observations and posing searching questions that open up the argument’s flaws, any misgivings about
it that you may have, and relevant perspectives that the researcher(s) may have overlooked. However,
do not use the article’s argument as a springboard to discuss the issue in detail or to make your own
argument on the issue. Your aim is to critique the argument, not to write an argument of your own. In
a brief concluding paragraph, suggest new directions that you believe future research on this issue
would benefit from pursuing.
III. Compose your analysis
Your analysis should be a minimum of 500 words, though it may well be longer. Those that have scored
well in the past have been significantly longer, often 750 words at least. Quality matters more than
quantity, but for an assignment like this to be successful, it needs to be thoroughly critical, and that
cannot easily be done in a short span. As the assessment criteria for this assignment show, when grading
it, emphasis will fall on both its analytical strength and its effectiveness as a piece of writing. The style
you use when analyzing an argument is actually a part of the analysis itself: in critiques of this kind,
Research Article Analysis (20%)
F2019
effective style tends to be efficient, direct, precise, and evidence-oriented. Wherever appropriate, use
the terms that researchers use when discussing their work, and that we have reinforced in class.
Proofreading, editing, and documentation skills are also critical parts of the quality of an analysis, so give
them attention too. Making good use of the models provided will also support your efforts to perform a
rigorous analysis and to learn from the experience.
IV. Submit your analysis
Submit both a hard copy of your Research Article Analysis to me on the due date and also a soft copy to
the designated drop box on e-centennial. To assess your analysis, I need first to read the research article
that you have selected, so be sure to do one of the following:
i.
ii.
iii.
submit it to me in hard copy, or
upload/post a soft copy of it (many exist in pdf format) with your assignment to the
dropbox, or
integrate into your assignment a working link to it—but make sure the link works.
Unless I receive a hard copy of the research article that you have analyzed, or at least convenient
access to it through a working link or a soft copy of the article in the designated dropbox, I will not
read and grade your Research Article Analysis.
Purchase answer to see full
attachment