Expert answer:Reaction Paper

Solved by verified expert:Reaction Paper Guidelines A Reaction Paper is designed to develop and sharpen your critical thinking and writing skills. Your objective in writing this assignment is to define an issue clearly and to formulate and clarify your position on that issue by reacting to a controversial statement. Each of the five papers is worth 4% of the course grade. The paper must be at least 250-300 words, excluding the title page and reference page, and must adhere to the APA sixth edition writing format. Please reference the APA example given in the Resources section. If your paper does not comply with this format you will lose points. Please see the grading rubric that follows the list of Controversial Statements. Please cite at least one reference from a website provided in the Webliography and one from your text book. Please make sure to submit your paper to Turnitin.com. The instructions for Turnitin.com are located under the “Resources” section of your course navigation menu. Controversial Statement: Adolescent and adult offenders from dysfunctional homes should not be held accountable for predatory behaviors due to the environmental challenges and influences associated with their current or former homes.
chapter_9__early_risk_factors_for_young_homicide_offenders_and_victims.docx

Unformatted Attachment Preview

CHAPTER 9 Early Risk Factors for Young Homicide
Offenders and Victims
•
David P. Farrington
University of Cambridge
•
Rolf Loeber
University of Pittsburgh
•
Rebecca Stallings
University of Pittsburgh
•
Doni Lynn Homish
State University of New York at Buffalo
Homicide is the second leading cause of death for persons aged 15 to 24 in the United States,
after accidents and before suicide (Hoyert et al., 2001). According to 1997 data, 1 in 40 African
American males is murdered, compared with 1 in 280 Caucasian males. The peak ages at which
males died from this cause were between ages 18 and 25, with the rate of death in this group
being more than 100 per 100,000 African American males and more than 10 per 100,000
Caucasian males (Federal Bureau of Investigation, 2003). The rates of homicide offenders and
victims in ages 13–17 and 18–24 increased from 1985 to a peak in 1993, then declined. For ages
18–24, the rates at the peak in 1993 were almost double the rates in 1985; for ages 13–17, the
rates at the peak in 1993 were almost triple the rates in 1985. For individuals between the ages of
18 and 24, the peak homicide offending rate in 1993 was 280 per 100,000 for African Americans
and 30 per 100,000 for Caucasians; the peak homicide victimization rate in 1993 was 190 per
100,000 for African Americans and 20 per 100,000 for Caucasians (Cook & Laub, 1998).
In general, the ages of offenders and victims are positively correlated, with the peak risk for 19and 20-year-olds killing other 19- and 20-year-olds. Most Caucasian victims (84%) are killed by
Caucasian offenders; similarly, most African American victims (91%) are killed by African
American offenders. Most homicide offenders aged 18–24 killed with a gun; in the peak year of
1993, there were three times as many gun homicides as homicides by other means by this age
group (Fox & Zawitz, 1999; Snyder & Sickmund, 1999).
A great deal of research has focused on the characteristics of young homicide offenders.
Retrospective case-control studies show that they tend to come from broken homes and violent
families, have experienced parental alcoholism and child abuse, have low school achievement,
and have run away from home, truanted, and been suspended from school. Not surprisingly, they
also tend to have prior arrest histories. Much less is known about the backgrounds of young
homicide victims. However, it is known that assault victims disproportionately tend to have
committed offenses themselves (Lauritsen, Laub, & Sampson, 1992; Rivara et al., 1995).
As Kathleen Heide (2003) has pointed out, “The available literature on juvenile homicide
offenders is retrospective in nature. To say something definitive about etiological factors
associated with youth murder requires longitudinal studies of children” (p. 25). Only two
previous such studies of youth homicide have been performed, both using data from the
Pittsburgh Youth Study. Loeber and his colleagues studied the characteristics of 11 boys killed
and 29 wounded by guns (Loeber et al., 1999; Loeber et al., 2005). Compared with control
groups, these boys tended to have low academic achievement, depressed mood, poor parental
supervision, poor parent–boy communication, behavior problems of the father, the family on
welfare, and low socioeconomic status. The researchers also investigated factors that predicted
violent offenders out of all boys and factors that predicted homicide offenders out of all violent
boys. The factors that predicted homicide offenders were a high risk score, a positive attitude
toward substance use, conduct disorder, carrying a weapon, gang fighting, persistent drug use,
selling hard drugs, peer delinquency, peer substance use, and being held back in school.
FOCUS OF THIS CHAPTER
The key questions addressed in the research presented in this chapter are as follows:
1. Which risk factors, measured at the beginning of a prospective longitudinal survey, predict
homicide offenders and homicide victims out of a community sample 4 to 15 years later? This
question has never been addressed before.
• 2. How accurately can homicide offenders and victims be predicted at ages 7 to 13?
• 3. To what extent are risk factors similar for homicide offenders and victims?
• 4. Do early risk factors predict homicide offenders more accurately than homicide victims? This
relationship would be expected if homicide offenders were more antisocial and deviant than
homicide victims.
• 5. Do behavioral and attitudinal risk factors predict homicide victims more accurately than
explanatory risk factors? This relationship would be expected to the extent that behavioral and
attitudinal factors reflect the same underlying construct (e.g., an antisocial personality) as is
present in homicide offenders. By definition, explanatory factors measure a different underlying
construct. However, it is less clear that behavioral and attitudinal factors would predict homicide
victims more accurately, because it is less clear that being a homicide victim reflects an
underlying antisocial personality or any other key hypothetical construct (e.g., low self-control,
criminal propensity, deviance).
• 6. Are racial differences in the prevalence of homicide offenders and victims attributable to racial
differences in early risk factors?
•
METHODOLOGY
Samples
The Pittsburgh Youth Study is a prospective longitudinal survey of the development of offending
and antisocial behavior in three samples of approximately 500 Pittsburgh boys each, totaling
1517 boys. When they were first contacted to participate in the study in 1987–1988, random
samples of first-, fourth-, and seventh-grade boys enrolled in Pittsburgh public schools were
selected (Loeber, Farrington, Stouthamer-Loeber, & van Kammen, 1998; Loeber, Farrington,
Stouthamer-Loeber, & White, 2008). At that time, 72% of all children residing in Pittsburgh
attended public schools. The city of Pittsburgh covers the inner-city population of approximately
370,000 (in 1990), within of the Pittsburgh–Beaver Valley Metropolitan Statistical Area of
approximately 2,243,000.
Out of approximately 1000 boys in each grade selected at random for a screening assessment,
about 850 boys (85%) were actually assessed. The boys completed a self-report questionnaire
about antisocial behavior while their primary caretakers completed an extended Child Behavior
Checklist, and their teachers completed an extended Teacher Report Form (Achenbach &
Edelbrock, 1983; Edelbrock & Achenbach, 1984; Loeber, Stouthamer-Loeber, van Kammen, &
Farrington, 1989). We will refer to the primary caretaker as the mother, because this was true in
94% of cases. Participants did not differ significantly from the comparable male student
population in their scores on the California Achievement Test (CAT) or in their racial
composition (African American or Caucasian).
From the screening assessment, a risk score was calculated for each boy, indicating how many of
21 antisocial acts he had ever committed. Data from the boy, the mother, and the teacher were
then combined. The risk score was used to select the sample for follow-up, consisting of
approximately the 250 most antisocial boys in each grade and 250 boys randomly selected from
the remaining 600. Hence, the screening sample of about 850 per grade was reduced to a followup sample of about 500 per grade. In the first follow-up assessment conducted 6 months after the
screening assessment, the three samples of boys were aged (on average) 7, 10, and 13,
respectively. This chapter compares child risk factors measured in the screening and first followup assessments (in 1987–1988) with later information about homicide offenders and victims.
Risk Factors
Risk factors were classified as explanatory or behavioral/attitudinal. Explanatory variables were
those that clearly did not measure antisocial behavior, while behavioral/attitudinal variables
could have reflected the boy’s antisocial behavior. For example, a young mother and poor
parental supervision were classified as explanatory variables, while being truant and being
suspended from school were classified as behavioral variables.
The most contentious variable was peer delinquency. We classified it as a behavioral variable
because three-fourths of the delinquent acts committed by the boys were committed with their
peers. Therefore, boys who had committed delinquent acts usually had delinquent peers. Also,
Farrington and colleagues (2002) found that peer delinquency was the strongest correlate of a
boy’s delinquency in between-individual correlations (both measured at the same time) but did
not predict a boy’s later delinquency within individuals. These researchers concluded that peer
delinquency was not a cause of a boy’s delinquency, but rather measured the same underlying
construct as delinquency. In contrast, poor parental supervision was correlated with delinquency
between individuals and predicted delinquency within individuals.
The 21 explanatory risk factors measured in 1987–1988 and used in the present study were as
follows:
• • Child: Low guilt, hyperactivity–impulsivity–attention deficit (HIA), old for the grade (has been
held back), low achievement (both according to the California Achievement Test and according to
mother–teacher ratings), depressed mood, callous-unemotional (Frick, O’Brien, Wootton, &
McBurnett, 1994).
• • Parental: Young mother (a teenager when the boy was born), father seeking help for behavior
problems (rated by the mother, so available for all fathers whether present or absent), parental
substance use.
• • Childrearing: Poor parental supervision, physical punishment by the mother, poor parent–boy
communication.
• • Socioeconomic: Low socioeconomic status (the Hollingshead measure, based on parental
education and occupational prestige), family on welfare, broken family (missing biological
parent), large family, small house, unemployed mother, bad neighborhood (both according to
1990 census data and according to mother ratings).
These factors were derived from the 40 distinct explanatory factors, excluding those not known
for all three samples, those with a large amount of missing data (e.g., variables referring to the
operative father or to the mother–father relationship), and those not strongly related to
delinquency in 1987–1988 (odds ratio [OR] > 2.0). Some variables were added because they
were related to violence (OR > 2.0) (Farrington, Loeber, & Stouthamer-Loeber, 2003). Most
variables were based on interviews with mothers, and in a few cases supplemented by data from
boys.
The 19 behavioral/attitudinal risk factors measured in 1987–1988 and used in the present chapter
were as follows:
• • Child behavior: Serious delinquency, covert behavior (concealing, manipulative, untrustworthy),
physical aggression, nonphysical aggression, cruel to people, runs away, disruptive behavior
disorder according to the Revised Diagnostic Interview Schedule for Children (Costello et
al., 1982), high risk score.
• • Child attitude: Favorable attitude toward delinquency, favorable attitude toward substance use.
• • Parental: Bad relationship with parents, counter-control (the bad behavior of the boy inhibits
parental attempts at socialization).
• • Peer: Peer delinquency, delinquent or poorly socialized friends, bad relationship with peers.
• • School: Suspended, truant, negative attitude toward school, low school motivation.
For the present analyses, the explanatory and behavioral risk factors were dichotomized within
each sample into, as far as possible, the “worst” quarter (e.g., the one-fourth of the sample with
the lowest attainment or poorest supervision) versus the remainder. This dichotomization
fostered a “risk factor” approach and made it easy to study the cumulative effects of several risk
factors. It also made all variables directly comparable by equating sensitivity of measurement
and permitted the use of the OR as a measure of strength of relationship, which has many
advantages (Fleiss, 1981). In the Pittsburgh Youth Study, conclusions about the most important
explanatory variables for delinquency were not greatly affected by the use of dichotomous as
opposed to continuous variables, or by different dichotomization splits (Farrington &
Loeber, 2000).
Dichotomizing the variables within each sample made all the samples comparable and made it
possible to combine them. Arguably, the quarter of the youngest sample with the poorest parental
supervision, for example, was comparable to the quarter of the oldest sample with the poorest
parental supervision, even though the absolute levels of parental supervision were different in all
three samples.
Homicide Offenders and Victims
Information about homicide offenders and victims was obtained from searches of local, state,
and federal criminal records; interviews with the participants; searches of local newspapers; and
the local coroner’s office. According to information collected up to February 2006, 33 of the
study participants had been convicted of homicide. Eight were from the youngest sample, 13
from the middle sample, and 12 from the oldest sample. The first homicide attributable to a study
participant was committed in 1992; the offenders committed their homicides between ages 15
and 26 (median age 19). Their victims’ ages ranged between 1 and 61 years (median age = 22).
The weapon used was a gun in 24 cases; in other cases, the weapons used were hands (2 cases),
knives (2), a brick (1), a metal rod (1), an automobile (1), a cord (1), and matches (in a case of
homicide by arson). The motives were retaliation (12), robbery (7), a drug deal gone wrong (5),
gang related (3), domestic (1), arson (1), and mental problems (1); in 3 cases, the motives were
not known by the police.
Prior to February 2006, 30 of the study participants had died because of homicide (and 12 had
died from other causes). Two were from the youngest sample, 12 from the middle sample, and
16 from the oldest sample. The first homicide victim was killed in 1993, and all victims died
between ages 16 and 27 (median age = 21). The weapon used was a gun in 27 cases. One
participant killed someone and then was killed himself. He was counted as an offender, leaving
33 homicide offenders and 29 homicide victims.
In addition to 33 participants convicted of homicide, 30 had been arrested for homicide but not
convicted. Some of these individuals may be convicted in the future, but many were not
convicted either because they terrorized potential witnesses or because the evidence was too
weak. Some, of course, may have been innocent. These 30 individuals were excluded from the
analyses. A further four individuals who died between ages 13 and 14 for reasons other than
homicide were also excluded, on the grounds that they were not at risk of becoming homicide
offenders or victims. Excluding these 34 cases left 33 homicide offenders, 29 homicide victims,
and 1421 controls.
RESULTS
Race Differences in Homicide
In light of national statistics, it is not surprising that African American males in this sample were
more likely than Caucasian males to be both homicide offenders and victims: 28 of the 33
offenders (85%) were African American, as were 27 of the 29 victims (93%). Both figures were
disproportionate to the overall composition of the population (African Americans accounted for
780 of the 1421 controls, or 55%). Prospectively, 3.4% of African American males were
convicted of homicide, compared with 0.8% of Caucasian males, and 3.2% of African American
males were killed, compared with 0.3% of Caucasian males. Weighting back to the population of
Pittsburgh public schools, 2.7% of African American males were convicted of homicide,
compared with 0.5% of Caucasian males; and 2.6% of African American males were killed,
compared with 0.3% of Caucasian males. Unweighted data are used in this chapter, because the
main interest is in studying predictions for individuals rather than making population estimates.
For ease of exposition, this chapter presents retrospective rather than prospective percentages
(Table 9.1). The OR for race versus homicide offenders was 4.6, while the OR for race versus
homicide victims was 11.1. The confidence intervals are large because of the small numbers of
homicide offenders and victims.
Later Features
Table 9.1 also shows the relationship between classic homicide features (guns, gangs, and drugs)
and homicide offenders and victims. These features were measured in multiple data waves before
any male was a homicide offender or victim. They are not studied as early risk factors in this
chapter.
TABLE 9.1 Features of Homicide Offenders and Victims
Percentage of
Controls (1421)
Percentage of
Offenders (33)
Percentage of
Victims (29)
Odds
C-O
Ratio
C-V
55
85
93
4.6*
11.1*
Carried gun
6
20
33
4.0*
8.0*
Used weapon
8
25
29
3.8*
4.5*
Gang member
14
30
19
2.6*
1.4
Persistent
drug user
19
39
25
2.8*
1.5
Sold
marijuana
8
28
21
4.2*
2.9*
Sold hard
drugs
8
31
30
5.5*
5.1*
Feature
African
American
Abbreviations: C-O, contrasting controls and offenders; C-V, contrasting controls and victims.
Offenders and victims were not contrasted because of small numbers.
P < .05 Homicide offenders were not always more antisocial than homicide victims. In particular, homicide victims were more likely to carry guns (33%) than homicide offenders (20%). The OR for victims (8.0) was twice as great as the OR for offenders (4.0). Also, homicide victims were roughly as likely as homicide offenders to have used a weapon (e.g., when attacking someone or in a gang fight or robbery) or to have sold hard drugs. Homicide offenders, in contrast, were more likely to be gang members, to be persistent drug users, and to have sold marijuana. * Explanatory Predictors Eight of the 21 explanatory factors significantly predicted homicide offenders: broken family, bad neighborhood (according to the census), old for the grade (held back), family on welfare, low guilt, a young mother, an unemployed mother, and low socioeconomic status (Table 9.2). The strongest predictor was coming from a broken family: 88% of homicide offenders had a missing biological parent (usually the father) by the first year of the study compared with 62% of controls (OR = 4.4). Nine of the 21 explanatory factors significantly predicted homicide victims: a broken family, low achievement (according to the California Achievement Test), low guilt, old for the grade, behavior problems of the father, HIA, family on welfare, callous-unemotional, and large family size (Table 9.2). Again, the strongest predictor was a broken family; 90% of homicide victims had a missing biological parent compared with 62% of controls (OR = 5.2). Based on explanatory risk factors, homicide offenders were not more extreme than homicide victims; if anything, the odds ratios for predicting victims were greater than for predicting offenders. Notably, however, most of the significant predictors of homicide offenders were socioeconomic/demographic factors: a broken family, a bad neighborhood, the family on welfare, a young mother, an unemployed mother, and low socioeconomic status. In contrast, most of the significant predictors of homicide victims were individual factors: low achievement, low guilt, old for the grade (held back), HIA, and callous-unemotional. Explanatory Risk Scores A logistic regression analysis was carried out to investigate which of the eight significant predictors of homicide offenders were independent predictors. Three variables were found to be significant or nearly significant in a stepwise analysis: a bad neighborhood (likelihood ratio chisquared [LRCS] = 13.22, P = .0003), old for the grade (LRCS = 9.21, P = .002), and a young mother (LRCS = 2.85, P = .091). In the final model, the partial ORs were 3.2 for a bad neighbo ... Purchase answer to see full attachment

How it works

1. Paste your instructions in the instructions box. You can also attach an instructions file
2. Select the writer category, deadline, education level and review the instructions 
3. Make a payment for the order to be assignment to a writer
4.  Download the paper after the writer uploads it 

Will the writer plagiarize my essay?

You will get a plagiarism-free paper and you can get an originality report upon request.

Is this service safe?

All the personal information is confidential and we have 100% safe payment methods. We also guarantee good grades