Exposure to Community Violence, Suicidality, and
Psychological Distress Among African American and Latino
Youths: Findings From the CDC Youth Violence Survey
M. Daniel Bennett, Jr.
Department of Social Work, Albany State University, Albany, Georgia, USA
School of Social Work and Department of Psychiatry, University of Michigan, Ann Arbor,
The present investigation explored risk and protective factors for suicidal ideation and behavior in a
sub-sample of African American and Latino adolescents (n = 2,626) who participated in the 2004
Centers for Disease Control Youth Violence Survey. Structural equation modeling was used to explore
exposure to violence at the community level as a contextual factor that could potentially influence
depressive symptomatology, substance abuse, parental support, social support, and suicidality among
study participants. Findings indicated that exposure to violence at the community level was not directly
related to suicidality among this population of urban adolescents. However, it was directly related with
several other variables under study in the model, which in turn were directly related with suicidality.
Tests of invariance revealed several across-group differences, particularly by race and gender, in how
the identified risk and protective factors in the model related to suicidality. Implications for research and
practice with urban, ethnic minority, adolescent populations are discussed.
Keywords: Community violence, suicidality, psychological distress, CDC, youth violence
Historically, suicide rates among ethnic minority youth (e.g., African American) have been relatively
low when compared to White youth (Oâ€™Donnell, Oâ€™Donnell, Wardlaw, & Stueve, 2004; Shaffer, 1988),
with the exception of Native Americans. However, racial disparities in youth suicide have narrowed.
For instance, rates of suicide among young African Americans have increased in recent years
(Borowsky, Ireland, & Resnick, 2001; Joe, 2008; Price, Drake, & Kucharewski, 2001). From 1980
to 1995, the suicide rate for African American youth between the ages of 15 and 19 increased 126 %
(Borowsky et al., 2001). Moreover, African American youth between the ages of 10 and 14 experienced
a 233% increase in completed suicide during this same time period (Borowsky et al., 2001).
Recent data show that the suicide rates among adolescent and young adult African Americans
leveled off, then declined during the early 2000â€™s but remain higher than they were in the early
1980s (Crosby and Molock, 2006). Although information on trends in suicide among U.S. Latinos
Address correspondence to M. Daniel Bennett, Jr., Department of Social Work, Albany State University, 504 College
Drive, Albany, GA 31705, USA. E-mail: [email protected]
Journal of Human Behavior in the Social Environment, 25:775â€“789, 2015
Copyright Â© Taylor & Francis Group, LLC
ISSN: 1091-1359 print/1540-3556 online
is limited, it has been shown that though the highest suicide rates occur among older adults, the
greatest proportion of suicides (approximately 50%) occur among those aged 10 to 34 years
(Centers for Disease Control [CDC], 2004).
Trends in non-fatal suicidal behavior have been documented among Latino youth. The Youth
Risk Behavior Surveillance Survey (YRBSS) consistently indicates that Latino youth are more
likely than both African American and White youth to attempt suicide (CDC, 2004, 2006, 2008).
Other studies have shown that the majority of lifetime suicide attempts described by Latinos were
reported as occurring when they were under 18 years of age (Fortuna, Perez, Canino, Sribney, &
There is a dearth of research on the contextual risk factors for suicide among African American
adolescents. Moreover, there is a limited amount of such research on Latino adolescents (Zayas,
Lester, Cabassa, & Fortuna, 2005; Goldstein et al, 2008). Further, even less research exists on those
factors that may serve to protect against the contextual risks for adolescent suicide.
African American and Latino adolescents tend to be concentrated in urban areas and as a result,
may be more vulnerable to stressors endemic to the urban environment (e.g., exposure to community level violence, social disorganization) than those living in other types of settings. This may in
turn contribute to elevated levels of risk for poor outcomes including, but not limited to, depression, substance abuse, and suicide. The current state of violence prevention research involving
young ethnic minority populationsâ€”particularly African Americans and Latinoâ€”has focused
primarily on aggressive peer-directed interpersonal violence. Less attention has been devoted to
the study of self-directed violence among these populations (Loeber, Farrington, StouthamerLoeber, Moffitt, & Caspi, 1998). Research examining the contextual factors associated with nonfatal suicidal behavior among African American and Latino adolescents is scarce at best. In an
effort to more thoroughly explore the risk and protective factors associated with self-directed
violence and to address certain limitations of the current state of research in this area, we report on
the prevalence, possible predictive risk factors, and potential mediators and moderators of suicidal
behaviors in a sample of urban African American and Latino youth who participated in the 2004
CDC Youth Violence Survey.
EXPOSURE TO COMMUNITY VIOLENCE: A CONTEXTUAL RISK FACTOR FOR
Violence exposure at the community level is all too commonplace in the lives of children and
adolescents, particularly for those growing up in urban settings (Asarnow et al., 2008; CDC, 1995;
Horowitz, McKay, & Marshall, 2005; Lambert, Ialongo, Boyd, & Cooley, 2005; Richters &
Martinez, 1993). Recent studies have suggested that over one-third of urban children and adolescents
have been exposed to community violence including violent victimization (Margolin & Gordis,
2000; Vermeiren, Ruchkin, Leckman, Deboutte, & Schwab-Stone, 2002). This widespread occurrence of community violence exposure among urban children and adolescents makes it necessary to
closely examine the potential negative outcomes that are invariably related to such exposure.
According to Fitzpatrick, Piko, Wright, and LaGory (2005), individuals who live in urban
settings have a greater exposure to interpersonal violence than do persons living in non-urban
environments. As such, urban residents are thought more likely to witness or be the victim of
some form of interpersonal violence than those living in non-urban environments. Moreover,
numerous studies have documented significant relationships between exposure to interpersonal
violence and a host of poor social and developmental outcomes including (1) anxiety, (2)
depression, (3) substance abuse, and (4) increased likelihood for violent victimization and
perpetration (Fitzpatrick & Boldizar, 1993; Fitzpatrick & LaGory, 2000; Fitzpatrick et al.,
2005; Lambert et al., 2005).
776 M. D. BENNETT, JR. AND S. JOE
Prior research has indicated that exposure to community-level violence may be associated with
violent behaviors among African American and Latino youth (Bell & Jenkins, 1993). National
surveys suggest that community violence is witnessed by approximately one-third of students in
junior high and high school (Schwab-Stone et al., 1999). For example, Mazza and Reynolds
(1999), in a study of adolescents (70% African American and 22% Latino) enrolled in the sixth
through eighth grades, found that 93% of participants had been exposed to violence in some way.
The impact of community violence exposure may also extend to certain health behaviors. For
example, experiencing or witnessing violence has been linked to substance abuse (Kilpatrick et al.,
2000; Sussman, Dent, & McCullar, 2000; Sussman, Dent, & Stacy, 1999), increased number of
sexual partners (Valois, Oeltmann, Walker, & Hussey, 1999), and psychiatric disorders such as
posttraumatic stress disorder, depression, and externalizing behavior disturbances (Buka, Stichick,
Birdthistle, & Earls, 2001; Fehon, Grilo, & Lipschitz 2001).
In a recent study of adolescent girls (50% African American), Berenson, Wiemann, and
McCombs (2001) found that compared with adolescents who were not exposed to violence,
those adolescents who had witnessed violence were two to three times more likely to report
using tobacco and marijuana, drinking alcohol or using drugs before sexual intercourse, and having
intercourse with a partner who had multiple partners. Fehon et al. (2001) found that psychiatrically
hospitalized adolescents who had a history of witnessing community violence reported significantly more suicidal ideations, posttraumatic stress disorder symptoms, substance use, and potential for violence than did those patients without such a history. This study, however, did not
examine race or ethnic differences.
Several studies have shown exposure to community violence to have a significant relationship
to suicidality among adolescents. Cohen (2000) found that the more violence an adolescent was
exposed to, the more he or she exhibited a vulnerability to suicide. In a study of students from
urban high schools in New York City (N = 630), researchers found that 87% of students had been
exposed to some form of peer-directed or community violence (e.g., knew someone who was
murdered, witnessed stabbing, shooting, beating, robbery, or involved in school fights). Further,
students who knew someone who had been murdered were twice as likely to report suicidal
ideations and four times as likely to report a suicide attempt. Last, witnessing a stabbing was
associated with twice the likelihood of suicidal ideations and three times the likelihood of a suicide
attempt (Pastore, Fisher, & Friedman, 1996).
Community violence exposure is also related to interpersonal and psychological distress (CooleyQuille, Boyd, Frantz, & Walsh, 2001). Yet, very little research has been conducted on the
relationship between exposure to community-level violence and suicidality, specifically among
African American and Latino adolescents (Mazza & Reynolds, 1999; Vermeiren et al., 2002).
African Americans and Latinos represent a sizeable portion of the urban population. As such, more
research is needed given that adolescents who live in urban areas tend to be exposed to particularly
large amounts of community violence.
Studies have indicated that suicide attempts among African American and Latino youth living in
urban areas occur at approximately twice the national rate (Price, Drake, & Kuharewski, 2001).
This suggests that the increased prevalence of suicidal behavior among these populations may in
part be related to certain social and environmental factors endemic to the urban environment
(Oâ€™Donnell et al., 2004; Willis, Coombs, Cockerham, & Frison, 2002). To date, studies on
community violence exposure and suicidality have used clinical or small community samples.
Currently, there are no studies using population-based samples of urban African American and
Latino adolescents. Therefore, the current investigation explores the independent relationship
between community violence exposure and suicidality among a community-representative sample
FINDINGS FROM THE CDC YOUTH VIOLENCE SURVEY 777
of African American and Latino adolescents using data from the Centers for Disease Control and
Prevention Youth Violence Survey (CDCYVS).
The CDC Youth Violence Survey: Linkages Among Different Forms of Violence
The CDCYVS was administered to a census of public school students enrolled in grades 7, 9, 11,
and 12 in a school district in an urban community in the northeastern region of the United States
(grades 11 and 12 were combined due to the low enrollment in each of those grades). The selected
school district was racially and ethnically diverse and located in a city with a population of fewer
than 250,000. This district operated 16 schools and served students at the targeted grades. All 16
schools were invited and agreed to participate in the study. These included middle schools, high
schools, and alternative schools.
Data collection occurred in April, 2004. Students voluntarily completed the anonymous, selfadministered 174-item questionnaire in classrooms during a 40-minute class period. Students
without parental permission or who did not want to participate in the study were assigned
individual desk work (by the classroom teacher), which they completed at their desks or at an
alternate location designated by the school during the survey administration. The teacher was not
present during the survey administration. The questionnaire, an optically scanable booklet in
multiple-choice format, was administered by field staff highly experienced in school-based survey
data collection. All English-speaking students in the targeted grades were invited to participate in
Students who could not complete the questionnaire independently (e.g., enrolled in a special
education class, required the assistance of a translator, had cognitive disabilities that would prevent
adequate understanding and responding to the survey; n = 151), or who were no longer attending
school (e.g., had dropped out of school, had been expelled, or were on long-term out-of-school
suspension; n = 202) were not eligible to participate in the study.
The CDCYVS Sample
A total of 5,098 students met eligibility criteria, and 4,131 participated, yielding an 81% response
rate. Participants were enrolled from three grade levels: 1,491 in seventh grade, 1,117 in ninth
grade, and 1,523 in eleventh and twelfth grades combined. However, we limited our investigation
to the responses provided by African American and Latino study participants (n = 2,626). We
chose to focus on this sub-sample because prior research has established that these are particularly
vulnerable segments of the population about whom relatively little is known in regard to suicidality
and related risk and protective factors.
Instruments and Variables
Based on the existing literature and the range of items included in the data set, we identified both risk
and protective factors thought to influence suicidality among study participants. Exposure to community violence, depressive symptomatology, and substance abuse were identified as potential risk
factors, while parental support and social support were identified as potential protective factors.
Exposure to Community Violence
Frequency of exposure (through sight and sound) to violence in oneâ€™s home and neighborhood
was assessed by six items from the Childrenâ€™s Exposure to Community Violence Scale (Richters &
778 M. D. BENNETT, JR. AND S. JOE
Martinez, 1993). Study participants were asked to indicate how often they had seen or heard certain
violence-related incidents around their home and neighborhood. Scale items included statements
such as â€œI have heard guns being shotâ€ and â€œI have seen somebody get stabbed or shot.â€ Response
selections ranged from 1 (never) to 4 (many times). The Cronbachâ€™s alpha for the six-item measure
The frequency of depressive symptoms during the 30 days prior to administration of this survey
was assessed by the Modified Depression Scale (Orpinas, 1993). Scale items included questions
such as â€œDid you feel hopeless about the future?â€ and â€œWere you very sad?â€ Response selections
ranged from 0 (never) to 4 (always).
The frequency of self-reported drug and alcohol use was assessed by five items from the Youth
Risk Behavior Survey (YRBS). Study participants were asked to indicate the frequency of drug
and/or alcohol use as well as the age at which they began using substances. Items included
questions such as â€œDuring the past 12 months, on how many days did you have at least one
drink of alcohol?â€ and â€œDuring the past 12 months, on how many days did you use inhalants or
illegal drugs such as marijuana, cocaine, or heroin?â€ The response categories for these items ranged
from 0 (never) to 6 (every day or almost every day).
Parental support was assessed by nine items from the Rochester Youth Development Study (see
Thornberry, Krohn, & Bushway, 1998). Study participants were asked to indicate the frequency
with which their parents rewarded good behavior with some form of praise as well as the
importance of their parents knowing their whereabouts, friends, and activities. The response
categories for these items ranged from 1 (almost never) to 3 (almost always). Three additional
items were included to determine whether study participants were subjected to curfews and other
monitoring practices. These items included questions such as â€œHow important is it for your parents
to what you were doing when you were outside of the home?â€ The response categories for these
items ranged from 1 (not important) to 3 (very important).
The Vaux Social Support Record (VSSR) was used to measure study participantsâ€™ level of
satisfaction with perceived emotional advice and practical social support (Vaux, 1988). Participants
were asked to indicate how much they agreed or disagreed with a series of statements such as â€œAt
school there are adults I can talk to, who care about my feelings and what happens to meâ€ and
â€œThere are people in my family I can talk to, who give good suggestions and advice about my
problemsâ€ The response categories for the VSSR ranged from 0 (not at all) to 2 (a lot).
Standard descriptive analyses were conducted to classify and summarize basic information
about study participants. An overview of study participants by race and gender is presented in
Table 1. Reliability analyses using Cronbachâ€™s coefficient alpha were used to ascertain the internal
consistency of the multiple item scales and is presented in Table 2.
FINDINGS FROM THE CDC YOUTH VIOLENCE SURVEY 779
Exploratory Factor Analysis
An exploratory factor analysis (EFA) using a maximum likelihood extraction method with
Promax rotation was conducted for each set of variables in the current study. Absolute values
less than .40 were suppressed. EFA was utilized in order to verify the conceptualization of each set
of variables under consideration for the current study population. That is, EFA was used to
determine how and to what extent the observed variables were linked to their underlying
As shown in Table 2, reliability analyses using Cronbachâ€™s coefficient alpha were used to
ascertain the internal consistency of the multiple item scales used in the current study and to
confirm the internal validity of study items with the current study population.
Based on the results of the EFA, we generated a structural equation measurement model for each
construct in the current study. These models were used to define the relationship between the latent
and observed variables. They focused solely on the extent to which the observed variables were
theoretically and psychometrically linked to the latent construct (Byrne, 2001). Moreover, the
structural equation measurement models served to refine the results of the EFA.
Structural Equation Modeling
In the interest of parsimony, the factor score weights from each measurement model were used
to create weighted composites of the latent constructs. This allowed the latent constructs to be
represented as observed variables. Moreover, this strategy also served to reduce the number of
parameters to be estimated in the baseline structural model.
Demographic Sample Characteristics
African American Latino Total
n % n % n %
Female 576 41.2 820 58.7 1,396 53.1
Male 540 43.9 690 56.1 1,230 46.8
Total 1,116 42.5 1,510 57.5 2,626 99.9
Construct Number of Items Standardized Item Alpha
Exposure to violence 6 .87
Depressive symptoms 6 .84
Substance abuse 3 .79
Parental support 9 .71
Social support 4 .80
Suicidality 4 .79
780 M. D. BENNETT, JR. AND S. JOE
The baseline structural model posited exposure to (community) violence as a contextual factor
that influenced outcomes related to depressive symptomatology, substance abuse, parental support,
social support, and suicidality (Figure 1).
In an effort to determine whether the baseline structural model operated equivalently across
different sub-populations of the sample, we estimated the model for entire sample (N = 2626)
and then for African Americans (n = 1,116), Latinos (n = 1,510), African American males (n =
540), African American females (n = 576), Latino males (n = 690), and Latino females (n =
820). The baseline model was estimated for each group separately with no between-group
constraints. We then analyzed the data for all groups simultaneously in order to obtain efficient
We present descriptive information about the population under study in Table 1. The mean age of
African American youth in the sample was 15.15 years. The mean age of Latino youth in the
sample was 14.8 years. The mean age for the entire sub-sample (African American and Latino
youth) was 14.95 years.
Next, we were interested in the prevalence of suicidality among study participants and how said
prevalence compared nationally. According to the 2003 YRBSS, approximately 8% of African
American youth and 11% of Latino youth surveyed reported attempting suicide at least once in the
past 12 months. Moreover, roughly 4% of African American youth and 5% of Latino youth
surveyed reported having made suicide attempts in the past 12 months that required medical
e9 1 Parental
FIGURE 1 Baseline model.
FINDINGS FROM THE CDC YOUTH VIOLENCE SURVEY 781
attention. This would seem to suggest that suicide ideation and suicide attempts are of growing
prevalence and concern among African American and Latino populations.
In the current sample, nearly 8% (n = 87) of African American study participants and 11% (n =
164) of Latino study participants reported having attempted suicide at least once in the past 12
months. Approximately 3% (n = 30) of African American study participants and 4% (n = 65) of
Latino study participants reported needing medical attention for a suicide attempt in the past 12
months. Finally, 13% (n = 146) of African American study participants and 17% (n = 255) of
Latino study participants reported having attempted suicide at least once in their life time. From
this brief comparison, it would seem that the prevalence of suicide related behaviors among the
current study population is almost identical to that found nationally in the 2003 YRBSS.
A reliability analysis was conducted on each set of items. This was done in an effort to examine the
internal consistency of the items used to measure the various constructs. The internal consistency
reliability of these items was estimated by Cronbachâ€™s coefficient alpha. The reliability analyses are
summarized and presented in Table 2.
We estimated the model shown in Figure 1 for the entire sample population (N = 2,626) and found
the model to be only marginally fitted to the data. The model yielded a chi-square (Ï‡2
) of 271 with
16 degrees of freedom (DF) and a significance (p) level of .000. The value (.978) of the goodnessof-fit index (GFI) was above the threshold for acceptable model fit (Byrne, 2001). However, the
Tucker-Lewis Index (TLI) produced a value (.834) well below the suggested .95 threshold for
acceptable model fit (Hu & Bentler, 1999). The comparative fit index (CFI) yielded a value (.926)
slightly below the recommended threshold of .95 for acceptable model fit (Hu & Bentler, 1999).
The .078 value of the root mean square error of approximation (RMSEA) suggested a mediocre
model fit (Byrne, 2001). The value (.0405) of the standardized root mean residual (SRMR) was
within the range for acceptable model fit (Hu & Bentler, 1995). Last, the Hoelterâ€™s critical N value
of 256 suggested an acceptable fit to the data (Hoelter, 1983). Thus, estimation of the model for the
entire sample population produced mixed results.
In an effort to identify a model better fitted to the data, the model-generating process (Byrne,
2001) was utilized (Figure 2).That is, we proceeded in an exploratory (rather than confirmatory)
manner to modify and re-estimate the model. Again, our goal at this point was to locate the source
of misfit in the model and identify a model that better described the sample data (Byrne, 2001;
Joreskog, 1993). Exposure to violence showed no relationship to suicidality. As such, we elected to
delete this path from the model. In addition, the errors for Social Supportâ€”Friends and Social
Supportâ€”School were correlated.
Despite these modifications, the value of the TLI remained well below the .95 threshold for
acceptable model fit. The value of the CFI (.947) approached the threshold for acceptable model fit.
The value of the RMSEA (.066) and the value of the SRMR (.0360) suggested a well-fitted model.
Thus, the modifications derived from the model generating process resulted in a model that was
marginally well fitted to the data.
We then estimated the baseline model for each group separately with no between-group
constraints (see Table 3). The model continued to be marginally well fitted to the data. One
exception occurred when the model was estimated for African American males. The model was
well fitted to this group. The values for the GFI, TLI, and CFI were all above the .95 threshold for
acceptable model fit. The values of the RMSEA, SRMR, and CN further suggested acceptable
782 M. D. BENNETT, JR. AND S. JOE
Tests of invariance revealed several across-group differences. As stated earlier, exposure to
violence showed no direct relationship to suicidality. However, it was significantly related to
depressive symptoms and substance abuse for all groups. Further, exposure to violence had a
statistically significant, negative relationship to parental support for all groups except African
American females. For this group, exposure to violence demonstrated no relationship to parental
support. Last, exposure to violence was not significantly related to social support for any of the
We did not test for mediating or moderating effects with respect to exposure to violence and
suicidality as there was no statistically significant relationship between these two factors. As such,
the model development approach was employed, and this path was subsequently deleted from the
model. Across-group differences were observed in how the remaining identified risk and protective
factors in the model related to suicidality.
Depressive symptoms and substance abuse were significantly related to suicidality for all study
participants. A statistically significant, negative relationship between parental support and suicidality was observed for both Latino male and female study participants. This relationship was
positive for African American study participants. For African American males, African American
females, and Latino males, parental support showed no relationship to suicidality. A statistically
significant, negative relationship was observed between social support and suicidality for African
Americans and African American males in the study. This relationship was non-significant for the
remaining study participants.
We also explored relationships among the identified risk and protective factors in the model and
how these relationships differed across groups. A statistically significant, negative relationship was
observed between depressive symptoms and parental support for all groups in the current study.
Similarly, a statistically significant, negative relationship was observed between substance abuse
and parental support for all groups in the study. Depressive symptoms were positively related to
social support for Latino males. This relationship was non-significant for the remaining study
e9 .18 Parental
FIGURE 2 Baseline model (estimated).
FINDINGS FROM THE CDC YOUTH VIOLENCE SURVEY 783
participants. A statistically significant, positive relationship was observed between substance abuse
and social support for African American females, Latinos, and Latino females. Once again, this
relationship was non-significant for remaining study groups.
Exposure to violence as victim, witness, or perpetrator or the experience of others occurs across all
areas of society. However, urban ethnic minority youth are exposed to violence far more frequently
than other segments of society (Farrell & Bruce, 1997; Fitzpatrick & Boldizar, 1993; Youngstrom,
Weist, & Albus, 2003). In the current investigation, exposure to violence had no direct relationship
to suicidality for study participants. This finding runs somewhat counter to prior research that
suggests that violence exposure is related to suicidality (Mazza & Reynolds, 1999; Oâ€™Donnell
et al., 2004; Vermeiren et al., 2002). However, it should be noted that while our sample was drawn
from a high-risk urban environment, it was not necessarily at increased risk for suicide. That is, the
sample was not drawn from a population of adolescents who were currently undergoing psychiatric
hospitalization or treatment for suicide-related behaviors. Therefore, violence exposure in and of
itself may not be sufficient to increase the risk of suicidality. Rather, it may exacerbate suicidality
in populations with preexisting risk for such behaviors.
Exposure to Violence and Parental Support
Although exposure to violence was not directly related to suicide, it was found to have significant
relationships to other factors in our model. We observed a statistically significant, negative relationship
between exposure to violence and parental support for all study participants. The one exception was
African American females; for this group, the relationship between exposure to violence and parental
support was non-significant. While it is not clear why exposure to violence failed to exert any
significant influence on parental support for African American females, our findings nonetheless
indicate that, on balance, living in an environment where individuals frequently encounter violence
may diminish the capacity of some parents to provide adequate levels of support to their children.
Exposure to Violence, Depressive Symptoms, and Substance Abuse
Exposure to violence was significantly related to both depressive symptoms and substance abuse
for all study participants. Thus, it appears that living in a setting where the occurrence of violence
is frequent may lead to stress-related emotional problems including depressive symptomatology
(Moses, 1999) and maladaptive coping responses including substance abuse (DuRant et al., 2000).
Last, we found no significant relationship between exposure to violence and social support for any
of our study participants. This finding is consistent prior research that suggests that social support
does not buffer against the effects of exposure to violence at the neighborhood or community level
(Muller, Goebel-Fabbri, Diamond, & Dinklage, 2000).
Depressive Symptoms, Substance Abuse, and Suicidality
Depressive symptoms had the strongest direct relationship to suicidality of any factor in the model.
This finding was consistent across all groups. Substance abuse was observed to have a strong direct
relationship to suicidality as well. Once again, this finding was consistent across all groups. There is a
substantive research literature on the link(s) between depressive symptomatology, substance abuse and
adolescent suicidality (Gould, Greenberg, Velting, & Shaffer, 2003). Although only recently have
researchers begun to explore these relationships among African American and Latino adolescent
784 M. D. BENNETT, JR. AND S. JOE
populations, it appears as though depressive symptoms and substance abuse may function consistently
across race and gender in how they impact suicidal thoughts and behaviors.
Parental Support and Suicidality
In the current investigation, parental support showed a statistically significant, negative relationship to suicidality for Latino male and female study participants. For African Americans, parental
support was positively associated with suicidality. This relationship was statistically non-significant for the remaining study groups. This suggests possible cultural differences in the role and
function of parental support for study participants.
The across-group differences in the relationship of parental support to suicidality may be
related in part to variation in parenting styles and practices. Prior research suggests four basic
parenting styles: authoritative, authoritarian, permissive, and disengaged (Pittman & ChaseLansdale, 2001; Smetana, 1995). Each of these parenting styles is said to vary in terms of
responsiveness to, and expectations of, child and adolescent behaviors. Recent studies have
indicated that, on balance, African American parenting practices tend to be more authoritarianâ€”
especially within the context of high-risk urban environmentsâ€”than parenting practices of other
ethnic groups (Smetana, 2000).
Some have argued that authoritarian parenting practices are linked to better outcomes for
African American adolescents due to the social environment in which many may live (Lamborn,
Dornbusch, & Steinberg, 1996). The current study, however, suggests that the benefits of authoritarian parenting might not extend to prudential issues that pertain to suicidality. In addition, the
research literature on parental support suggests that it consists primarily of two dimensionsâ€”
monitoring/supervision and emotional support/nurturing. These dimensions may be potential
sources of variation in child and adolescent outcomes as well. Past research has found that children
and adolescents who receive either extreme of the aforementioned parenting dimensions are at
greater risk for problem behavior (Stice, Barrera, & Chassin, 1993).
Social Support and Suicidality
Recent studies have indicated that emotional/social support, particularly from family members and
other adults, often functions as a resiliency factor that moderates suicidality for African American
and Latino youth (Oâ€™Donnell et al., 2004). However, in the current study, a statistically significant,
negative relationship was observed between social support and suicidality only for African
Americans and African American males in the study. This relationship was statistically nonsignificant for the remaining groups in the sample.
The variation in social support in the current study may be related to the composition of that
support. Prior research suggests that social support networks composed of adults tend to foster
resilience and promote better outcomes than do social support networks that are made up of
individuals from the adolescentâ€™s peer groups. In the current study, social support consisted of
three observed endogenous variablesâ€”family, friends, and schoolâ€”and one latent exogenous
factor. While measuring social support in this manner was in the best interest of parsimony, it
eliminated our ability to explore whether a particular aspect or dimension of social support is
related to suicidality. Therefore, we cannot be certain whether the variation in social support is
attributable to the composition of those networks or some other phenomenon.
Depressive Symptoms and Parental Support
A statistically significant, negative relationship was observed between depressive symptoms and
parental support for all groups in the current study. This finding was consistent with prior research
FINDINGS FROM THE CDC YOUTH VIOLENCE SURVEY 785
and theory that posit that the attitudes and behaviors of individuals with symptoms of depression
may lead to the erosion of support systems and networks (Stice et al., 2004). Some have argued
that there may in fact be a reciprocal relationship between depressive symptoms and parental
support. The attitudes and behaviors of individuals with depressive symptoms may alter their
perceptions of the nature and quality of parental support. By the same token, parents and other
family members may feel alienated by the attitudes and behaviors of individuals with depressive
symptoms and as a result feel inept to provide necessary support (Lazarus & Folkman, 1984;
Slavin & Rainer, 1990).
Depressive Symptoms and Social Support
Depressive symptoms were positively related to social support for Latino males. This relationship
was non-significant for the remaining study participants. These results suggest that dysfunction or
disruption in support networks may be related to depressive symptomatology for Latino male study
participants. Moreover, prior research has shown that peer social support is positively linked to
depression among adolescents (Barrera & Garrison-Jones, 1992). Yet, it should once again be
noted that the manner in which social support was measured limited our ability to explore specific
aspects or dimension of this particular construct. That is, we not able to determine the composition
of social support networks of study participants.
Substance Abuse and Social Support
A statistically significant relationship was observed between substance abuse and social support for
Latinos, African American females, and Latino females. Once again, our findings regarding social
support were somewhat counter-intuitive. Social support is generally thought of as protective factor
that moderates the impact of certain risk factors. However, prior research suggests that the nature of
social support may be a potential source of variation in substance abuse outcomes for adolescents.
Peer social support has been positively linked to substance abuse while social support derived from
adults seems to buffer against substance abuse (Wills & Vaughan, 1989).
IMPLICATIONS FOR PROMOTING HEALTH OF YOUNG AFRICAN AMERICAN MALES
Implications for Research
Despite our findings, exposure to violence was not directly associated with suicidality among this
population of urban adolescents. Future research efforts should continue to examine this relationship using specific types of violence exposure variables rather than a generalized measure like the
one used in the current study. For instance, prior studies show consistently that child abuse is
significantly associated with higher risk for suicidal behavior. However, we were not able to
disaggregate different forms of violence exposure. In addition, further research is needed that
investigates gender differences in the effect of social support on suicide risk among African
Americans. The findings from such investigations will have several implications for prevention
and for working with potentially suicidal ethnic minority adolescents.
Implications for Practice
The findings of the current investigation have implications for practice, prevention, and intervention with African American and Latino adolescents. In our study, depressive symptoms and
substance abuse were found to have strong direct relationships to suicidality. Although causality
786 M. D. BENNETT, JR. AND S. JOE
could not be determined due to the cross sectional nature of the data, these findings do, however,
suggest that the presence of depressive symptoms and/or substance abuse may increase the likelihood for non-fatal suicidal behavior regardless of race or gender. As such, it is incumbent upon
public health practitioners and other human service professionals to be cognizant of the potential
ramifications associated with these risk factors and to effectively address them through evidencebased practice, prevention, and intervention methods.
Our study also found that exposure to violence had no direct relationship to suicidality. Yet, it
was found to have direct relationships to both depressive symptoms and substance abuse. In turn,
these risk factors were related to suicidality. This perhaps suggests that certain community-level
dynamics, although distal in nature, have the potential to influence particular social and developmental outcomes through the emergence and function of proximal risk factors. Thus, contextual
antecedents (e.g., exposure to violence) warrant ample consideration as an overarching risk factor
that may contribute to mediating risk processes that are potentially linked to poor social and
developmental outcomes including, but not limited to, suicidality.
Last, we observed several within-group and across-group differences with respect to the
variables and constructs under study. This would seem to indicate that the nature and extent of
risk processes and protective mechanisms may be a function of race/ethnicity and gender.
Therefore, it is once again incumbent upon practitioners of prevention and intervention efforts to
understand that not all interventions are suitable for all groups. Rather, programs should be tailored
to the specific needs of the target population. This includes acknowledgement of the role of culture,
be it age-related, gender-related, or race/ethnicityâ€“related.
The authors acknowledge Dr. Alex Crosby of the Division of Violence Prevention at the Centers
for Disease Control and Prevention for his insightful comments and suggestions that helped in the
completion of this submission.
Asarnow, J. R., Baraff, L. J., Berk, M., Grob, C., Devich-Navarro, M., Suddath, R., . . .Tang, L. (2008). Pediatric emergency
department suicidal patients: Two-site evaluation of suicide ideators, single attempters, and repeat attempters. Journal of
the American Academy of Child and Adolescent Psychiatry, 47(8), 958â€“966.
Barrera, M. Jr., & Garrison-Jones, C. (1992). Family and peer social support as specific correlates of adolescent depressive
symptoms. Journal of Abnormal Child Psychology, 20(1), 1â€“16.
Bell, C., & Jenkins, E. J. (1993). Community violence and children on Chicagoâ€™s Southside. Psychiatry: Interpersonal and
Biological Processes, 56, 46â€“54.
Berenson, A. B., Wiemann, C. M., & McCombs, S. (2001). Exposure to violence and associated health-risk behaviors
among adolescent girls. Archives of Pediatrics and Adolescent Medicine, 155, 1238â€“1242.
Borowsky, I. W., Ireland, M., & Resnick, M. D. (2001). Adolescent suicide attempts: Risks and protectors. Pediatrics, 107
Buka, S. L., Stichick, T. L. Birdthistle, I., & Earls, F. J. (2001). Youth exposure to violence: Prevalence, risks, and
consequences. American Journal of Orthopsychiatry, 71, 298â€“310.
Byrne, B. (2001). Structural equation modeling with AMOS: Basic concepts, applications and programming. Mahwah, NJ:
Lawrence Erlbaum Associates.
Centers for Disease Control and Prevention (CDC). (2004). Suicide among Latinosâ€”United States, 1997â€“2001. Morbidity
and Mortality Weekly Report, 53(22).
Centers for Disease Control and Prevention (CDC). (2004). Youth risk behavior surveillanceâ€”United States, 2003.
Surveillance Summaries, May 21, 2004. Morbidity and Mortality Weekly, 53(SS-2).
Centers for Disease Control and Prevention (CDC). (2006). Youth risk behavior surveillanceâ€”United States, 2005
Surveillance Summaries. June 9, 2006. Morbidity and Mortality Weekly, 55(SS-5).
FINDINGS FROM THE CDC YOUTH VIOLENCE SURVEY 787
Centers for Disease Control and Prevention (CDC). (2008). Youth risk behavior surveillanceâ€”United States, 2007.
Surveillance Summaries, June 6, 2008. Morbidity and Mortality Weekly, 57(SS-4).
Cohen, E. M. (2000). Suicidal ideation among adolescents in relation to recalled exposure to violence. Current Psychology,
Cooley-Quille, M., Boyd, R. C., Frantz, E., & Walsh, J. (2001). Emotional and behavioral impact of exposure to community
violence in inner-city adolescents. Journal of Clinical Child Psychology, 30(1), 199â€“206.
Crosby, A. E., & Molock, S. D. (2006). Introduction: Suicidal behaviors in the African-American community. Journal of
Black Psychology, 32, 263â€“261.
DuRant, R. H., Altman, D., Wolfson, M., Barkin, S., Kreiter, S., & Krowchuk, D. (2000). Exposure to violence and
victimization, depression, substance use, and the use of violence by young adolescents. Journal of Pediatrics, 137(5),
Farrell, A. D., & Bruce, S. E. (1997). Impact of exposure to community violence on violet behavior and emotional distress
among urban adolescents. Journal of Clinical Child Psychology, 26(1), 2â€“14.
Fehon, D. C., Grilo, C. M., & Lipschitz, D. S. (2001). Correlates of community violence exposure in hospitalized
adolescents. Comprehensive Psychiatry, 42(4), 283â€“290.
Fitzpatrick, K. M., & Boldizar, J. P. (1993) The prevalence and consequences of exposure to violence among AfricanAmerican youth. Journal of the American Academy of Child & Adolescent Psychiatry, 32(2), 424â€“430.
Fitzpatrick, K. M., & LaGory, M. (2000). Unhealthy places: The ecology of risk in the urban landscape. New York, NY:
Fitzpatrick, K. M., Piko, B. F., Wright, D. R., & LaGory, M. (2005). Depressive symptomatology, exposure to violence, and
the role of social capital among African American adolescents. American Journal of Orthopsychiatry, 75(2), 262â€“274.
Fortuna, L., Perez, D., Canino, G., Sribney, W., & Alegria, M. (2007). Prevalence and correlates of lifetime suicidal ideation
and attempts among Latino subgroups in the United States. Journal of Clinical Psychiatry, 68, 572â€“581.
Goldston, D., Molock, S. D., Whitbeck, L., Murakami, J. L., Zayas, L. H., & Nakayama-Hall, G. (2008). Cultural
considerations in adolescent suicide prevention and psychosocial treatment. American Psychologist, 63, 14â€“31.
Gould, M. S., Greenberg, T., Velting, D. M., & Shaffer, D. (2003). Youth suicide risk and preventive interventions: A review
of the past 10 years. Journal of the American Academy of Child & Adolescent Psychiatry, 42(4), 386â€“405.
Hoelter, J. W. (1983). The analysis of covariant structures: Goodness of fit indices. Sociological Methods and Research, 11,
Horowitz, K., McKay, M., & Marshall, R. (2005). Community violence and urban families: Experiences, effects, and
directions for intervention. American Journal of Orthopsychiatry. 75(3), 356â€“368.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fix indexes in covariance structure analysis: Conventional critera versus
new alternatives. Structural Equation Modeling, 6, 1â€“55.
Joreskog, K. G. (1993). Testing structural equation models. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation
models (pp. 294â€“316). Newbury Park, CA: Sage.
Kilpatrick, D. G., Acierno, R., Saunders, B., Resnick, H. S., Best, C. L., & Schnurr, P. P. (2000). Risk factors for adolescent
substance abuse and dependence: Data from a national sample. Journal of Consulting and Clinical Psychology, 68, 19â€“
Lambert, S. F., Ialongo, N. S., Boyd, R. C., & Cooley, M. R. (2005). Risk factors for community violence exposure in
adolescence. American Journal of Community Psychology, 36(1â€“2) 29â€“48.
Lamborn, S. D., Dornbusch, S. M., & Steinberg, L. (1996). Ethnicity and community context as moderators of the relations
between family decision making and adolescent adjustment. Child Development, 67, 283â€“301.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal and coping. New York, NY: Springer.
Loeber, R., Farrington, D. P., Stouthamer-Loeber, M., Moffitt, T., & Caspi, A. (1998). The development of male
offending: Key findings from the first decade of the Pittsburgh Youth Study. Studies in Crime and Crime Prevention
Margolin, G., & Gordis, E. B. (2000). The effects of family and community violence on children. Annual Review of
Psychology, 51(1), 445â€“479.
Mazza, J. J., & Reynolds, W. M. (1999). Exposure to violence in young inner-city adolescents: Relationships with suicidal
ideation, depression, and PTSD symptomatology. Journal of Abnormal Child Psychology, 27(3), 203â€“213.
Moses, A. (1999). Exposure to violence, depression, and hostility in a sample of inner city high school youth. Journal of
Adolescence, 22, 21â€“32.
Muller, R. T., Goebel-Fabbri, A. E., Diamond, T., & Dinklage, D. (2000). Social support and the relationship between family
and community violence exposure and psychopathology among high risk adolescents. Child Abuse & Neglect, 24(4),
Oâ€™Donnell, L., Oâ€™Donnell, C., Wardlaw, D., & Stueve, A. (2004). Risk and resiliency factors influencing suicidality among
urban African American and Latino youth. American Journal of Community Psychology, 33(1/2), 37â€“49.
788 M. D. BENNETT, JR. AND S. JOE
Orpinas, P. (1993). Skills training and social influences for violence prevention in middle school: A curriculum evaluation.
(Unpublished doctoral dissertation). University of Texas Health Science Center at Houston, School of Public Health,
Pastore, D. R., Fisher, M., & Friedman, S. B. (1996). Violence and mental health problems among urban high school
students. Journal of Adolescent Health, 18, 320â€“324.
Pittman, L. D., & Chase-Landsdale, P. L. (2001). African American adolescent girls in impoverished communities: Parenting
style and adolescent outcomes. Journal of Research on Adolescence, 11(2), 199â€“224.
Price, J. H., Drake, J. A., & Kucharewski, R. (2001). Assets as predictors of suicide attempts in African American inner-city
youths. American Journal of Health Behavior, 25(4), 367â€“375.
Range, L. M., Leach, M. M., McIntyre, D., Posy-Deters, P. B., Marion, M. S., Kovac, S. H., â€¦ Vigil, J. (1999).
Multicultural perspectives on suicide. Aggression and Violent Behavior, 4(4), 413â€“430.
Richters, J., & Martinez, P. (1993). The NIMH community violence project I: Children as victims of and witnesses to
violence. In D. Reiss, J. Richters, M. Radke-Yarrow, & D. Scharff (Eds.), Children and violence (pp. 7â€“21). New York,
NY: Guilford Press.
Schwab-Stone, M., Chen, C., Greenberger, E., Silver, D., Lichtman, J., & Voyce, C. (1999). No safe haven II: The effects of
violence exposure on urban youth. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 359â€“367.
Shaffer, D. (1988). The epidemiology of teen suicide: An examination of risk factors. Journal of Clinical Psychiatry, 49, 36â€“41.
Slavin, L. A., & Rainer, K. L. (1990). Gender differences in emotional support and depressive symptoms among
adolescents: A prospective analysis. American Journal of Community Psychology, 18(3), 407â€“421.
Smetana, J. G. (1995). Parenting styles and conceptions of parental authority during adolescence. Child Development, 66(2),
Smetana, J. G. (2000). Middle-class African American adolescentsâ€™ and parentsâ€™ conceptions of parental authority and
parenting practices: A longitudinal investigation. Child Development, 71(6), 1672â€“1686.
Stice, E., Barrera, M. Jr., & Chassin, L. (1993). Relation of parental support and control to adolescentsâ€™ externalizing
symptomatology and substance abuse: A longitudinal examination of curvilinear effects. Journal of Abnormal Child
Psychology, 21(6), 609â€“629.
Sussman, S., Dent, C. W., & McCullar, W. J. (2000). Group self-identification as prospective predictor of drug use and
violence in high risk youth. Psychology of Addictive Behavior, 14, 192â€“196.
Sussman, S., Dent, C. W., & Stacy, A. W. (1999). The association of current stimulant with demographic substance use,
violence related, social and intrapersonal variables among high risk youth. Addictive Behavior, 24, 741â€“748.
Thornberry, T. P., Krohn, M. D., & Bushway, S. D. (1998, October). Predicting resistance using measures of onset.
Washington, DC: Office of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice.
Valois, R. F., Oeltmann, J. E., Walker, J., & Hussey, J. R. (1999). Relationship between number of sexual intercourse partners
and selected health risk behaviors among public high school adolescents. Journal of Adolescent Health, 25, 328â€“335.
Vaux, A. (1988). Social support: Theory, research, and intervention. New York, NY: Praeger.
Vermeiren, R., Ruchkin, V., Leckman, P., Deboutte, D., & Schwab-Stone, M. (2002). Exposure to violence and suicide risk
in adolescents: A community study. Journal of Abnormal Child Psychology, 30(5), 529â€“537.
Willis, L., Coombs, D. W., Cockerham, W. C., & Frison, S. L. (2002). Ready to die: A postmodern interpretation of the
increase of African American adolescent male suicide. Social Science & Medicine, 55, 907â€“920.
Wills, T. A., & Vaughan, R. (1989). Social support and substance use in early adolescence. Journal of Behavioral Medicine,
Youngstrom, E., Weist, M. D., & Albus, K. E. (2003). Exploring violence exposure, stress, protective factors and behavioral
problems among inner city youth. American Journal of Community Psychology, 32(1/2), 115â€“129.
Zayas, L. H., Lester, R. J., Cabassa, L. J., & Fortuna L. R. (2005). Why do so many Latina teens attempt suicide? A
conceptual model for research. American Journal of Orthopsychiatry, 75, 275â€“287.
FINDINGS FROM THE CDC YOUTH VIOLENCE SURVEY 789
Copyright of Journal of Human Behavior in the Social Environment is the property of Taylor
& Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a
listserv without the copyright holder’s express written permission. However, users may print,
download, or email articles for individual use.
Exposure to Community Violence, Suicidality, and