You Do What You See: How © 2016 SAGE Publications ...people.oregonstate.edu/~flayb/MY PUBLICATIONS... · Alvin Thomas1 ,2, Cleopatra H. Caldwell3 4, Shervin Assari2,3, Robert J

Journal of Men’s Studies 1 –23

© 2016 SAGE PublicationsReprints and permissions:

sagepub.com/journalsPermissions.nav DOI: 10.1177/1060826516641104

men.sagepub.com

Article

You Do What You See: How Witnessing Physical Violence Is Linked to Violent Behavior Among Male African American Adolescents

Alvin Thomas1,2, Cleopatra H. Caldwell3,4, Shervin Assari2,3, Robert J. Jagers5, and Brian Flay6

AbstractAfrican American boys are more likely than same-aged counterparts to live in disadvantaged neighborhoods characterized by exposure to physical violence, lower socioeconomic status, poor parent education, and acts of violence. The current study used structural equation modeling to test the associations between witnessing violence, peer and parent expectations, peer behaviors, self-efficacy to avoid violence, and violent behavior as the outcome. Results suggest that African American boys who witnessed physical violence are more likely to engage in violence themselves. Peer and parent violence expectations, peer violence, and adolescent’s self-efficacy to avoid violence mediate this. These findings suggest potential for prevention of violent behaviors through modification of norms of male African American adolescents at risk for witnessing violence in their daily life.

Keywordsviolence, African American, adolescent, male, self-efficacy, positive psychology, norms

1Clinical Psychology Program, Palo Alto University, CA, USA2Department of Psychiatry, School of Medicine, University of Michigan, MI, USA3Center for Research on Ethnicity, Culture and Health, School of Public Health, University of Michigan, MI, USA4Department of Health Behavior and Health Education, School of Public Health, University of Michigan, MI, USA5Combined Program in Education and Psychology, School of Education, University of Michigan, MI, USA6College of Public Health and Human Sciences, Oregon State University, OR, USA

Corresponding Author:Alvin Thomas, Palo Alto University, 1791 Arastradero Road, Palo Alto, CA 94304, USA. Email: [email protected]

641104 MENXXX10.1177/1060826516641104Journal of Men’s StudiesThomas et al.research-article2016

2 Journal of Men’s Studies

Physical violence witnessed by adolescents includes physical assaults and assaults with injury, use of weapons, sexual victimization and harassment, property victimization (e.g., vandalism and robbery), and witnessing violent acts, including shootings (Finkelhor, Turner, Ormrod, & Hamby, 2009), and negative interactions with the police. Statistics from a 2010 nationally representative sample of 4,500 adolescents aged 17 years and younger (National Survey of Children’s Exposure to Violence) found that 61% were victims or witnesses to physical violence in the previous year (Sickmund & Puzzanchera, 2014). Boys were more likely to experience assaults whereas girls were more likely to be sexually victimized, and Blacks were more than twice as likely to experience multiple forms of violence compared with adolescents of any race (Sickmund & Puzzanchera, 2014). The Centers for Disease Control and Prevention’s (CDC) Youth Risk Behavior Survey (YRBS), a monitor of health risk behaviors among 15,425 ninth to 12th graders, reported that in 2011 about 33% of youth were in physical fights and one in 25 were seriously injured in fights (Centers for Disease Control and Prevention, 2011; Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, 2010). Recent incidents in Fergusson Missouri, and across the United States, suggest that violent and often deadly interactions between police and African Americans are also a significant risk faced by youth in disadvantaged neighborhoods (Gabrielson, Jones, & Sagara, 2014; Rosich, 2007).

Differences in youth exposure to violence may, in part, be explained by residence and level of exposure to concomitant social and economic status of the families of these youth (Hawkins et al., 2000). In fact, African American boys living in high-risk neighborhoods are more exposed to the kinds of social and economic factors that are associated with greater opportunity to witness physical violence and to repeat these behaviors whether as a response to victimization or as a way to avert victimization (Cassidy & Stevenson, 2005). This study addresses African American boys’ height-ened risk of experiencing physical violence as victim and or perpetrator. Our goal in this article is to better understand how individual characteristics and multiple social influences (i.e., peers, parents) contribute to the link between witnessing physical vio-lence and violent behavior for African American boys. This may inform efforts at supporting African American youth who have experienced, directly and indirectly, community violence including experiences similar to episodes like Ferguson.

Risk and Protective Factors for Youth Violence

Youth violence is influenced by multiple risk factors such as witnessing, being a victim of violence, and peers’ expectations of violence. There are also multiple protective fac-tors such as family influences, and individual strengths like self-efficacy to avoid vio-lence (Farrington & Ttofi, 2011; Loeber, Farrington, Stouthamer-Loeber, & White, 2008). The literature has implicated peers in the development of both youth deviant and prosocial behaviors (Loeber et al., 2008; Lösel & Farrington, 2012). However, while adolescence is marked by increased peer influence, parents still have an important role to play in adolescents’ prosocial decision making and in protecting them from negative outcomes. In addition, self-efficacy to avoid violence is an individual protective factor

Thomas et al. 3

known to reduce the likelihood of negative outcomes (Jagers, Morgan-Lopez, Howard, Browne, & Flay, 2007; Riner & Saywell, 2002) such as violent behaviors. Self-efficacy represents the individual’s beliefs about their ability to make positive choices in spe-cific situations (Bennett & Fraser, 2000). Eccles, Wigfield, and Schiefele (1998) pro-posed that the individuals’ own expectations for success and not just their outcome expectations are enshrined in the concept of self-efficacy. Violence self-efficacy there-fore includes confidence in one’s ability to avoid engaging in violence (McMahon et al., 2013), as well as the related expectation of success.

We use the protective factor model of the risk and resilience framework (Fergus & Zimmerman, 2005) to examine the influence of witnessing physical violence, peer and parent expectations about violence, peer violent behaviors, and self-efficacy on ado-lescents’ violent behaviors. This framework proposes that the presence of specific strengths or protective factors (e.g., self-efficacy, positive parent expectations) may lead to positive outcomes even in the face of risks (i.e., witnessing physical violence). The protective factor model holds that assets and resources (e.g., self-efficacy to avoid violence) reduce the influence of risk factors on negative outcomes. Based on the risk and resilience framework, individuals in high-risk contexts benefit from the availabil-ity of protective factors. The risk and resilience framework helps us study how protec-tive factors help youth overcome risk factors and resist deleterious outcomes (Fergus & Zimmerman, 2005; Zimmerman et al., 2013).

The literature holds that adolescents’ experiences with physical violence, whether as victim or a witness, are strongly associated with violent behavior (Calvete & Orue, 2011; Lindstrom Johnson, Finigan, Bradshaw, Haynie, & Cheng, 2011). Furthermore, in a study of 167 African American youth from sixth to eighth grade, Richards and colleagues (2004) found that adolescents who spent more unmonitored time with devi-ant peers were at greater risk of witnessing or becoming victims of violence, and that the amount of time spent in these risk contexts mediated the link between violence expo-sure and delinquent activity. We propose that witnessing physical violence influences adolescents’ beliefs about peer violence. Peers’ and parents’ behaviors and expectations are factors that affect adolescents’ self-efficacy to avoid physical violence, which in turn is related to violent behavior (Lindstrom Johnson et al., 2011; Robinson, Paxton, & Jonen, 2011). The current study proposes that for African American adolescent males, witnessing physical violence may not manifest as committing violence given the mediat-ing role of peer and parent expectations and adolescent self-efficacy to avoid violence.

Witnessed Physical Violence

Youth’s exposure to physical violence includes observing or hearing about actual vio-lence in their homes, schools, or neighborhoods (Kliewer et al., 2004). Among urban youth, 50% to 100% report witnessing violence in their community (Buka, Stichick, Birdthistle, & Earls, 2010; Stein, Jaycox, Kataoka, Rhodes, & Vestal, 2003), with African American boys at higher risk for witnessing violence than any other racial or gender group (Assari, Smith, Caldwell, & Zimmerman, 2015; Neumann, Barker, Koot, & Maughan, 2010). Youth from lower socioeconomic status (SES) families are


more likely to reside in poorer, urban neighborhoods characterized by higher crime levels (Neumann et al., 2010). This translates to increased exposure to violent experi-ences even as perpetrators. In fact, in a meta-analysis, Hawkins and colleagues (2000) found that family SES at 6 to 11 and 12 to 25 years of age was a strong predictor (though not the only predictor) of serious and violent delinquency, and that this was strongest for 6 to 11 year olds.

Peer Influences

Some literature has shown that peer influences may mediate the effect of protective factors on adolescent behavioral outcomes. In fact, an Aban Aya Youth Project (AAYP) study demonstrated that peer influence mediated the positive effect of the intervention on youth violence (Ngwe, Liu, Flay, & Segawa, 2004). Previous research has estab-lished a relationship between deviant peer norms and behaviors like juvenile delin-quency, substance abuse, and violence (e.g., Laird, Jordan, Dodge, Pettit, & Bates, 2001; Vitaro, Brendgen, & Tremblay, 2002). In particular, association with deviant peers is linked to violent behaviors for adolescent males (Copeland-Linder et al., 2007; Lambert, Ialongo, Boyd, & Cooley, 2005). In fact, while youth violence is generally associated with peer rejection, some youth are rewarded socially (e.g., popularity) for acting out peer-endorsed violent norms (Farrell et al., 2010; Prinstein & Cillessen, 2003). In a study of 503 adolescent boys, Pardini, Loeber, Farrington, and Stouthamer-Loeber (2012) found that high peer deviance was associated with youth violence at age 13 to 14 years, whereas low peer deviance was a protective factor at 15 to 18 years old. They reasoned that certain constructs like deviant peer expectations might func-tion as a risk for violence and a protection against future violence. Adolescents who supported non-violence engaged in fewer violent behaviors at ages 13 to 14 (Pardini et al., 2012). The current study addresses this issue specifically for African American boys. It probes paths through which peer violent behavior and peers’ violence expecta-tions are related to violent behavior for boys who have witnessed physical violence.

There is need to study whether peer violent behavior influences adolescents’ effi-cacy to avoid violent behaviors, and whether this link may be linked to peers’ violence expectations (Foney & Cunningham, 2002; Salzinger, Ng-Mak, Feldman, Kam, & Rosario, 2006). Farrell and colleagues (2010) conducted a qualitative study of 106 African American adolescents and found that peers exerted a strong influence on deci-sions to engage in or to avoid violence. These researchers also found that concern about status and reputation among their peers made adolescents more susceptible to the influence of peers’ expectations and behaviors. Adolescents whose social environ-ments have presented violence as an acceptable and successful means of gaining status among their peers (i.e., belief in peer expectation of violence) were less likely to avoid violence.

Adolescents often overestimate the extent of peers’ negative behaviors and expec-tations, and adolescents’ own behaviors are reflective of their skewed perception of the behaviors and expectations of their peers (Dishion & Owen, 2002; Prinstein & Wang, 2005). This misperception, pluralistic ignorance, results in a suppression of healthy

Thomas et al. 5

attitudes and behaviors in favor of unhealthy ones that are perceived to be normal (Berkowitz, 2004; Miller & McFarland, 1991). Even after controlling for adolescents’ own violent tendencies, Salzinger and colleagues (2006) found that peers’ violent norms or expectations were significant predictors of multiple negative outcomes, including witnessing, being a victim, and engaging in physical violence. This link has been found to be even more significant for adolescent males (Lambert et al., 2005).

Parents’ Non-Violent Expectations

Parental expectations may reduce adolescents’ engagement in violent behaviors (Brody, Kogan, Chen, & Murry, 2008) even in the face of peer influences. This protective role may function directly through parent messages about violence or indirectly through ado-lescents’ beliefs about parents’ expectations (Lindstrom Johnson et al., 2011). Parents in disadvantaged neighborhoods often compliment their protective parenting messages with messages reflective of the “code of the street”—an emphasis on maintaining the respect of others through the use of violence, toughness, and exacting retribution when perceived disrespect has occurred (Stewart & Simons, 2006; Wilkinson, 2003).

In a study of 168 African American parent–child dyads, Copeland-Linder et al. (2007) found that adolescents who believed that their parents support violence or who may have received messages reflective of the code of the street are more likely to engage in violence. For boys, this link was further solidified if they affiliated with deviant peers (Copeland-Linder et al., 2007). On the other hand, youth who believed their parents disapproved of violence as a problem-solving strategy may also hold non-supportive attitudes toward violence (Ohene, Ireland, McNeely, & Borowsky, 2006). Parents can protect their children by encouraging and communicating non-violence expectations. Adolescents’ belief in parent non-violent expectations and norms was more predictive of youth violence than family structure, the parent–child relationship, or parental monitoring (Orpinas, Murray, & Kelder, 1999). Several stud-ies have found that youth perceptions of their parents’ expectations and attitudes toward violence are more predictive of youth violence than explicitly stated parent norms (Ohene et al., 2006; Sieving, McNeely, & Blum, 2000). Perceived parental expectations, however, may protect adolescents against violent behaviors (Ohene et al., 2006). This highlights the critical role of parents’ norms and subsequent mes-sages in protecting youth.

Individual Factors

Self-efficacy refers to the individual’s beliefs about their ability to make positive choices in specific situations (Bennett & Fraser, 2000). While studies with European American youth have found that violence-avoidance self-efficacy beliefs influence pro-social and antisocial behaviors (Caprara, Regalia, & Bandura, 2002) fewer studies have explored this link for African American populations. Even fewer have explored this link in African American boys. The limited extant literature identifies self-efficacy to avoid violence as another protective factor associated with less violence among adolescents


(Jagers et al., 2007). Jagers and colleagues (2007) reported that for African American adolescents, self-efficacy beliefs were a negative predictor of violent behavior.

Adolescents who considered violence as an acceptable and successful means of gaining status among their peers were less likely to avoid violence. Farrell et al. (2010) showed that peer support, peer pressure, and concern about image were related to youth violence. Together, these peer influences reduced adolescents’ confidence in their ability to avoid violent engagements (self-efficacy to avoid violence). Youth who share the “code of the streets” may have higher expectations of peer support for vio-lence (Stewart & Simons, 2006). Peers may communicate support for violence through their promise to aid peers during a violent act, through taunting, and other verbaliza-tions that directly encouraged youth to engage in violence (Farrell et al., 2010). However, youth cited this as a main barrier to self-efficacy to avoid violence. Youth whose friends encourage or directly support violence, and those who perceive vio-lence as a means of saving face or establishing status, may be less confident in their ability to avoid violence. Conversely, they may think of this peer support as part of a mutual support system that provides an additional layer of protection in high-risk neighborhoods (Stewart & Simons, 2006).

Exposure to physical violence in the neighborhood and normative beliefs about violence predict use of violence in resolving social conflicts (Lindstrom Johnson et al., 2011; Lösel & Farrington, 2012; Robinson et al., 2011). Robinson and colleagues (2011) investigated how exposure to violence, normative beliefs about aggression, and depressive symptoms predicted adolescents’ use of an aggressive response style to conflict. This study of 80 African American adolescent males suggested that cogni-tions functioned as a pathway through which violent experiences may affect adoles-cents’ coping skills and behaviors. For African American boys in high-risk neighborhoods, these findings suggest that adjusting adolescents’ expectations and beliefs about violence may strengthen their self-efficacy to avoid violence, and reduce the likelihood of violent behaviors.

The Current Study

Incidents like the police killings of Mike Brown, Tamir Rice, and Eric Garner substan-tiate an increased risk for witnessing violence for African American males. This grow-ing trend may be even more potent for youth as the violence is perpetrated by an agency that is supposed to be inherently protective—the police, and also because youth exposure is continuous as each incident is carried across various forms of media. Against this backdrop, our goal in this study is to test whether multiple social influ-ences (i.e., peers, parents) and individual characteristics (self-efficacy) mediate the association between witnessing physical violence and acts of violent behavior among African American boys.

The following five hypotheses were tested:

Hypothesis 1: Witnessing physical violence is positively associated with violent behavior directly.

Thomas et al. 7

Hypothesis 2: The association between witnessing physical violence and violent behavior is mediated by self-efficacy to avoid violence.Hypothesis 3: Peer violence mediates the link between witnessing physical vio-lence and peers’ expectations.Hypothesis 4: Serial mediation exists such that witnessing physical violence func-tions through peer violent behavior, and peers’ expectations of violence and self-efficacy to avoid violence to influence violent behavior.Hypothesis 5: Self-efficacy mediates the relationship between parent expectations and youth’s violent behaviors.

Method

Procedure

The current study used data from the AAYP, a community study with a total of 1,153 participants. Participating schools had enrollments greater than 500 students with 80% African American and less than 10% Latino or Hispanic; Grades K through eighth; not on probation or slated for reorganization; and not a special designated school (e.g., magnet, academic center, and moderate mobility). A parent or guardian of each student was informed about the study and given the option to exclude their child from data collection. At the start of each survey administration, youth were reminded of the option to end their participation in the study at any time and to not answer any question(s) that made them feel uncomfortable. They were also informed of their right to withdraw their participation at anytime without consequence and all data have been kept confidential. Participants completed measures at four time points after the base-line measurement. Questions measured violent behaviors and experiences, substance use, sexual activity, social relationships, and family connections among other issues. At each wave, a three-member team administered the survey over a 2-hr period with a 5-min scheduled break (see Jagers, Morgan-Lopez, & Flay, 2009, for a more complete presentation of the AAYP).

Sample

We used data from the 553 African American adolescent male participants. The study received IRB approval and met ethical standards of research. Study participants were students from 12 schools recruited from poor metropolitan Chicago neighborhoods in 1993. Fifth grade was the first wave of data collection (1994-1995 school year). Less than 2% of parents requested exclusion of their children from the study (Jagers et al., 2009). Participating schools were randomly assigned to the Social Development Curriculum (n = 204), the School and Community (family and neighborhood Interventions) condition (n = 185), and the Health Enhancement Control condition (n = 182 participants). The groups did not differ significantly based on age, parent education level, length of time boys had lived in the neighborhood, or household income at baseline. Previous analyses showed no baseline differences on violence


measures after controlling for pre-intervention age and modeling school-level nesting (Jagers et al., 2009). The two intervention conditions had similar prevention effects compared with the Health Enhancement Control (Flay, Graumlich, Segawa, Burns, & Holliday, 2004). For this study, we combined these two intervention conditions (Ngwe et al., 2004). Intervention effects were controlled in the current study to account for potential influences.

Measures

Measures were based on multiple questionnaires (e.g., Youth Risk Behavior Surveillance Survey [YRBSS], National Health Interview Survey [NHIS], 1992). These Measures were adapted based on feedback from focus groups and pilot testing with adolescent and parents living in high-risk communities (Flay et al., 2004).

Covariates. Socio-demographic variables were standard measures of child’s age, aver-age household income, and parent education. Length of time lived in the neighborhood was operationalized as a continuous measure.

Witnessed physical violence. Adolescents’ witnessing physical violence was determined using a five-item measure. All items were dichotomous (0 = no and 1 = yes), indicating whether adolescents had ever witnessed certain acts of physical violence. The total score ranged from 0 to 5 with higher scores indicating witnessing more physical vio-lence. Item-scale correlations ranged from .33 to .52 (α = .68). Representative ques-tions included “Have you ever seen someone get shot at” and “Have you ever seen a friend or family member get cut.”

Peer violent behavior. Participants were asked how many of their classmates they thought were involved in violent behaviors. The scale consisted of two items “How many of the students in your grade get into a physical fight?” and “How many of the students in your grade carry a knife, a razor, or a gun?” Both items were measured on a Likert scale with 0 = none of them, 1 = some of them, 2 = about half of them, 3 = most of them, and 4 = all of them. These two items account for serious and more normative form of adolescent violence. Combined scores ranged from 0 to 8, where higher scores indicated that participants believed that more of their peers were involved in violent behaviors. The items were correlated at r = .409, p < .001, and a Spearman Brown coefficient of .281, p < .01. The literature has suggested additional reporting of the Spearman Brown coefficient when exploring the reliability of two-item scales (Eis-inga, Grotenhuis, & Pelzer, 2013).

Peers’ expectations of violence. Participants indicated how much their friends support violence or want them to engage in violent behaviors. The scale consisted of two items “Do your best friends want you to get into a physical fight?” and “Do your best friends want you to carry a knife, a razor, or a gun?” Both items were measured on a Likert scale with 0 = none of them, 1 = some of them, 2 = about half of them, 3 = most of them,

Thomas et al. 9

and 4 = all of them. Scores ranged from 0 to 8, where higher scores indicated that participants believed that peers wanted them to engage in violent behaviors. These two items accounted for serious and more normative forms of peer deviance. The items had a Pearson’s correlation of .436, p < .001, and a Spearman Brown coefficient of .566, p < .01.

Self-efficacy to avoid violence. Four questions were used to determine boys’ self-efficacy to avoid violence. The questions were as follows: How sure are you that you can (a) keep yourself from getting into physical fights, (b) keep yourself from carrying a knife, (c) stay away from situations in which you could get into fights, (d) can seek help instead of fighting. Responses were reported on a 0 to 4 scale where 0 = definitely cannot, 1 = maybe cannot, 2 = not sure, 3 = maybe can, and 4 = definitely can. Scores ranged from 0 to 16 and higher scores indicated higher levels of perceived ability to avoid violence. Item-scale correlations ranged from .43 to .65 (α = .83).

Parents’ non-violent expectations. Parents’ non-violent expectation was assessed using four items. Representative items included “Your parents want you to avoid carrying a knife or razor or gun?” and “Do your parents want you to stay away from situations where you could get into a fight?” All items were measured on a Likert scale where 0 = definitely no, 1 = probably no, 2 = not sure, 3 = probably yes, and 4 = definitely yes. The total score ranged from 0 to 16 with high scores indicating stronger belief that their parents expected them to avoid violence. Item-scale correlations ranged from .219 to .510 (α = .78).

Violent behavior. This outcome variable represents adolescents’ report of engagement in violent behavior which was assessed using seven questions adapted from the 1992 YRBS. The YRBSS was originally developed for high school students. To use these measures with this sample, questions were modified to reflect the earlier stages of vio-lence in which fifth- through eighth-grade students might engage. Participants indicated whether they had ever (a) threatened to beat up someone; (b) threatened to cut, stab, or shoot someone; (c) been in a physical fight; (d) carried a gun; (e) shot at someone; (f) carried knife or razor; and (g) cut or stabbed someone. Response choices were a simple dichotomy (0 = no; 1 = yes) for the lifetime involvement questions. A sum score was calculated for this measure. Scores ranged from 0 to 7 with high scores indicating more violent behaviors. Item-scale correlations ranged from .166 to .517 (α = .69).

Data Analysis

SPSS and AMOS were used for bivariate and multivariable analysis, respectively. In the first step, we described means, SDs, and frequencies when appropriate. For bivari-ate analysis, we used Pearson correlation test between study constructs. We used struc-tural equation modeling (SEM) for multivariable analysis. SEM allows for the simultaneous estimation of multiple meditational paths. This mode of analysis was also chosen because it allows for using multiple indicators to represent constructs, and


thus reduced measurement error. In addition, it facilitates the modeling of mediating relationships, error terms, and test coefficients.

A two-step modeling procedure was used, first fitting the measurement model and then testing the structural model (Kline, 2011). We used AMOS 19.0 (Arbuckle, 2009; Byrne, 2013). Missing data in the SEM procedure were addressed using full informa-tion maximum likelihood (FIML). We presented our three main variables (i.e., wit-nessed violence, self-efficacy to avoid violence, violent behaviors) as scales while the main mediators external to the individual (i.e., peer violent behavior, peers’ expecta-tions of violence, parents’ non-violent expectations) were presented as latent factors. Violent behaviors is the outcome variable. Missing values were replaced using EM if they were operationalized as a scale. FIML was used for the latent factors. We reported the following fit statistics: chi-square, the comparative fit index (CFI; >.90), the root mean squared error of approximation (RMSEA; <.06), and χ2 to degrees of freedom ratio less than 2 (Hu & Bentler, 1999; Lei & Lomax, 2005). We controlled for the effect of the intervention, and reported the standardized regression weights. To con-firm mediation, we calculated the Sobel test using coefficients for the relevant paths, in accordance with the recommendations of Baron and Kenny (1986).

For mediation, we used the Sobel (1982) first-order test, which is the most common product-of-coefficients test being used to test mediation. In the first step, we divided the indirect effect, α̂ , by the first-order delta-method standard error of the indirect effect. Then we compared the value against a standard normal distribution to test for significance. Mediation is considered to be present if significance is found (Sobel, 1982).

Results

Participants on average were 13.5 years old (SD = .62) and had lived an average of 3.6 years (SD = 1.36) in their neighborhood with a maximum of 7 years of neighborhood residence. Almost half of the boys (47%) lived in two-parent families. The average household income was $10,000 and only 14% had an income of more than $40,000. Eighty percent of parents had completed high school or a 2-year college degree. More than 20% of parents had some college-level course work, earned a 4-year college degree, or had earned some post college–level degree (see Table 1).

The majority of the African American boys (84%) reported having witnessed a phys-ical fight where someone was badly injured. Almost half of the boys had witnessed first-hand very serious violent acts including seeing someone get cut and stabbed or shot at. For almost 70% of these boys, their experiences were potentially more impact-ful as family members and friends were the victims of these violent acts. About 80% of the sample reported having threatened to beat up someone. This was the second most representative violent act. Physical fighting was the most common behavior with 96% reporting engagement in this behavior. Few had been involved in more serious behav-iors like physical fights that lead to injury (15%). Smaller portions of the sample (12%) had threatened to cut or stab someone, or engaged in very serious acts like carrying a gun (12%), cutting/stabbing someone (7%), or shooting at someone (5%).

βˆ

11

Tab

le 1

. D

escr

iptiv

e St

atis

tics

and

Cor

rela

tions

for

Dem

ogra

phic

s, N

orm

s, A

cts

of V

iole

nce,

and

Witn

esse

d V

iole

nce

in t

he T

este

d St

ruct

ural

Equ

atio

n M

odel

Am

ong

Blac

k Y

outh

in A

ban

Aya

You

th P

roje

ct (

N =

553

).

Var

iabl

esM

(SD

)1

23

45

67

89

10

1.

Witn

esse

d vi

olen

ce2.

64 (

1.40

)1

2

. V

iole

nt b

ehav

ior

2.82

(1.

34)

.384

**

3.

Peer

vio

lent

beh

avio

rs2.

98 (

1.06

).1

49**

.216

**

4.

Peer

s’ e

xpec

tatio

ns o

f vio

lenc

e2.

81 (

1.30

).1

41**

.293

**.3

28**

5

. Se

lf-ef

ficac

y to

avo

id v

iole

nce

11.1

7 (3

.07)

−.1

65**

−.3

29**

−.1

54**

−.2

46**

6

. Pa

rent

non

-vio

lent

exp

ecta

tions

12.0

5 (2

.60)

−.1

08*

−.1

47**

−.1

89**

−.1

99**

.277

7

. A

ge13

.5 (

.62)

.105

*.0

69.0

18.0

69−

.060

.073

—

8.

Inte

rven

tiona

—.0

08−

.023

−.0

57.0

50−

.011

.087

*−

.011

—

9.

Inco

mea

4.5

(2.3

1)−

.031

−.0

79−

.013

−.0

78.1

45**

.065

−.0

67.0

30—

10

. Pa

rent

edu

catio

na5.

61 (

2.27

)−

.115

**−

.010

.025

−.0

83.0

42.0

94**

−.1

07*

−.0

27.3

52**

—11

. Li

ved

in n

eigh

borh

ood

3.80

(1.

36)

−.0

08.0

26.0

47−

.007

.030

.028

.002

.066

.038

.015

a Ave

rage

inco

me

= $

10,0

00; a

vera

ge e

duca

tion

leve

l = v

ocat

iona

l edu

catio

n or

som

e co

llege

.*p

< .0

5. *

*p <

.001

.


Witnessing physical violence was positively correlated with other risk factors including peers’ expectations of violent behavior (r = .140, p < .01) and with adoles-cents’ violent behavior (r = .384, p < .01). Witnessing physical violence was nega-tively correlated with self-efficacy to avoid violence (r = −.165, p < .01) and parents’ non-violence expectations (r = −.106, p < .05). Self-efficacy to avoid violence was negatively related to witnessing physical violence and peers’ expectations of violence. Parents’ non-violence expectations were positively related to self-efficacy to avoid violence (see Table 1).

Fit statistics for the measurement model were good, χ2 = 89.706, df = 57, p = .004, χ2 / df = 1.574, CFI = .949, RMSEA = .032, 90% confidence interval [CI] = [.019, .045]. Standardized factor loadings for peer violent behavior and peers’ expectations of violence ranged from .358 to .793 and .528 to .833, respectively. The standardized loadings for parents’ non-violent expectations ranged from .516 to .678.

The full structural equation of the model also showed a good fit to the data, χ2 = 133.105, df = 87, p < .001, χ2 / df = 1.530, CFI = .928, RMSEA = .031, 90% CI = [.020, .041]. The path from witnessing physical violence to violent behavior was significant and positive (β = .301, p < .001), suggesting that those who had witnessed more violence engaged in more violent acts (see Figure 1). This confirmed the first hypothesis.

The second hypothesis was not confirmed, as the mediation path between self-efficacy to avoid violence and witnessing physical violence and engagement in violent behaviors did not reach statistical significance (ź = 1.806, p = .071). While witnessing physical violence had a negative effect on self-efficacy at both the bivariate and mul-tivariate levels, Sobel test results did not confirm a significant level of mediation through self-efficacy. Tests of the third hypothesis found that witnessing physical vio-lence also had an indirect effect on engagement in violence through adolescents’ expectation of peer violent behaviors (β = .317, p < .01). Peer violence influenced peers’ expectations of violence (β = .486, p < .01) and was associated with increased youth’s own violent behaviors (β = .296, p < .001).

Sobel tests were conducted to evaluate the indirect effect of witnessing violence through the two mediating variables. Peer violent behavior partially mediated the effect of witnessing physical violence on peers’ expectations of violence (ź = 1.988, p = .04). Peers’ expectations of violence partially mediated the relationship between peer violent behavior and adolescents’ violent behaviors (ź = 2.408, p = .02). These tests revealed a positive path from boys’ witnessing of physical violence to their report of peers’ physical violence. This indicated that boys who witnessed more physical violence also reported more violent acts among their peers. Boys who believed that their peers engaged in physical violence also believed that peers expect them to engage in physical violence. Boys who witnessed more physical violence had lower self- efficacy to avoid violence (β = −.088, p < .05), suggesting that their self-efficacy to avoid violence was negatively affected by witnessing physical violence.

For the fourth hypothesis, we found that the serial mediation continued from wit-nessing physical violence, peer violent behavior, and peers’ expectations of violence, and further functioned through self-efficacy to avoid violence. Peers’ expectations of violence were negatively associated with self-efficacy to avoid violence (β = −.205,

Thomas et al. 13

p < .001). The effect of peers’ expectations of violence on adolescents’ violent behav-ior was mediated by self-efficacy to avoid violence (ź = 2.641, p = .01).

The fifth hypothesis was confirmed with an indirect relationship between parents’ non-violent expectations and violent behavior, through self-efficacy to avoid violent behaviors. Parents’ non-violent expectations were linked to greater self-efficacy to avoid violence (β = .354, p < .001). Parents’ non-violent expectations were negatively associated with boys’ expectations of peer physical violence (β = −.345, p < .01). Boys who were more confident in their ability to avoid violence were less involved in vio-lent behaviors (β = −.177, p < .001). Self-efficacy to avoid violence completely medi-ated the effect of parents’ non-violent expectations (ź = − 2.729, p = .01) in association with violent behaviors.

Discussion

Previous studies have investigated the link between risk factors and violent behavior, but few have explored this association for African American boys who have witnessed physical violence. For instance, in a study based on the AAYP program, Ngwe and colleagues (2004) identified peer factors as mediators of the effect of the AAYP

Figure 1. Summary of the final structural equation model to test mediating effects of parents and peers norms and self-efficacy on the link between witnessed violence and violent behaviors among Black youth in Aban Aya Youth Project (N = 553).Note. All values are standardized regression coefficients. Hypotheses (Hyp.) tested in this model: Hyp. 1: d; Hyp. 2: f, i; Hyp. 3: a, b; Hyp. 4: a, b, e, i; Hyp. 5: h, i. χ2 = 133.105, df = 87, p = .001, χ2 / df = 1.530, N = 553, comparative fit index = .928, root mean squared error of approximation = .031, 90% confidence interval = [.020, .041].


intervention, which may protect adolescents from violent behaviors. This study used a risk and resilience framework to explore how peer, individual, and parent-based fac-tors help explain the link between witnessing physical violence and violent behavior for African American boys. The recent trend of violent and deadly police interactions with African American males in Fergusson and around the country suggests a need to explore how witnessing physical violence is related to similar behaviors and what fac-tors may help protect African American adolescent males.

The most crucial finding from this study is that African American boys’ perception of parents’ expectations or norms is related to higher self-efficacy in avoiding violence. Interestingly, their report of parents’ non-violent expectations was not directly related to violent behaviors among these adolescents; rather, parents’ protective role functioned through African American boys’ self-efficacy to avoid violence. Research has shown that this relationship applied whether parents’ non-violent expectations or norms were explicitly stated or perceived by adolescent (Farrell et al., 2010; Ohene et al., 2006). The current study contributes to the literature by identifying self-efficacy to avoid vio-lence as a critical path through which parents’ expectations may influence adolescent behaviors. As adolescents become more self-governing, it becomes more important to encourage their internal strengths as an effective channel for prevention efforts.

Farrell and colleagues (2008) found that when asked to identify factors that influ-enced violence-avoidance decisions, African American adolescents indicated that par-ents’ non-violence expectations or norms (“parents’ voice in their head”) were primary deterrents. We found that in the midst of the parent–peer struggle for influence over youth’s behavioral outcomes, parents remain relevant and effective in protecting African American adolescent boys. This study identifies parent non-violent expecta-tions as a way to effectively protect African American boys even when parents are ostensibly absent. This supports previous research (e.g., Farrell et al., 2010), which suggests that parents play a pivotal role during adolescence, representing a source of protection, especially for African American boys who have witnessed physical vio-lence. For beleaguered African American youth who have been exposed to violence, parental support remains a crucial form of protection (Thomas & Hope, 2015).

This study also adds to the literature which links witnessing physical violence and engagement in physical violence (Lindstrom Johnson et al., 2011; Neumann et al., 2010). The trauma of witnessing physical violence may make boys feel vulnerable (Cassidy & Stevenson, 2005). Boys may seek to understand their experiences through the lens of their friends’ violent behaviors as well as perception of their friends’ expec-tations related to violence. Negative peer influences are a significant risk for increas-ing violent behavior among African American boys (Copeland-Linder et al., 2007; Farrell et al., 2010; Lambert et al., 2005). We found that peer influence may be one mechanism through which witnessing physical violence may be linked to violent behaviors for African American boys. We found that witnessing more physical vio-lence is associated with a greater expectation of peer violence, and peers’ expectations of violence. Both are in turn related to violent behaviors for African American boys. Results of this study may have implications for the prevention of violence and delin-quency, as youth commit a significant portion of all violent crime (Snyder, 2000) and

Thomas et al. 15

experience among the highest rates of criminal victimization compared with any other age group (Bureau of Justice Statistics, 2001).

Adolescents tend to overestimate the negative behaviors of their peers while under-estimating their own negative behaviors (Prinstein & Wang, 2005). Adolescents’ belief that peers expect them to engage in violence, and an expectation that same-aged youth or peers are involved in violence, may normalize violent behaviors. This may lead adolescent to perceive violence as an acceptable pattern of behavior (Haynie, Silver, & Teasdale, 2006)—what Stewart and Simons (2006) refer to as the “code of the streets.” Youth in disadvantaged neighborhoods who have assimilated the “code of the streets” and have witnessed violence are at increased risk of having interactions with the justice system both as victims and perpetrators. They are also more likely to become cynical of governmental systems (Thomas & Hope, 2015). For youth in com-munities like Fergusson, this fear and cynicism is supported by negative experiences with the police and is related to witnessing violence. The effect of witnessing physical violence therefore may skew adolescents’ perception about acceptability and popular-ity of physical violence among peers, and influence African American males’ own behaviors. On the other hand, as youth continually witness violence, they come to see it as unavoidable, and therefore beyond their ability to efficaciously avoid. Youth involvement in violence could therefore also be seen as acceptance of reality.

The finding that adolescent males’ belief that peers expect them to be violent par-tially mediates the link between witnessing physical violence and adolescents’ violent behavior supports studies showing neighborhood effects on youth behaviors (Haynie, Silver, & Teasdale, 2006). The social disorganization model originally proposed by Shaw and McKay (1942) posits that specific social processes may mediate the associa-tion between living in unsafe neighborhoods and adolescent violence. In particular, social disorganization theory conceptualizes neighborhoods as complex systems of friends, same-aged peers, and family networks that operate through different socializa-tion processes (Bellair, 2000). Negative peer influence is conceptualized as a charac-teristic of social disorganization and is known to increase adolescent violence indirectly by increasing opportunities for adolescents to become involved in violent peer net-works (Haynie et al., 2006).

Adolescence has been identified as a developmental period characterized by the increased influence of peers competing with that of parents (Sim, 2000). Findings from the current study suggest that while peer influences may increase the likelihood of vio-lent behaviors for African American boys who witness physical violence, boys who believe that their parents do not want them to engage in violence may be less likely to engage in violence. Moreover, boys who believe that their peers engage in physical violence and have violent expectations of them are more likely to report less self- efficacy to avoid violence. While peers’ behaviors and expectations weaken boys’ self-efficacy to avoid violence in the face of having witnessed physical violence, parents have a protective role to play. The study identifies individual and parent factors that mitigate the relationship between witnessing physical violence and violent behaviors.

Previous studies have identified the deleterious role of witnessing physical violence, among other forms of violence (Lindstrom Johnson et al., 2011), and peer influences on


adolescents’ behavioral outcomes. The current study’s findings contribute to the litera-ture by identifying African American males’ self-efficacy to avoid violence as a protec-tive factor in light of peer influences and violent experiences. Peer influences (i.e., behaviors and expectations) appear to be a critical lens through which African American boys make meaning of their experiences of witnessing physical violence. Boys who wit-ness more violent acts perceive more peer violence and are likely to report that peers expect them to engage in violence. Boys’ self-efficacy to avoid violence is associated with fewer violent outcomes. For these boys, however, peers’ violent expectations could imperil them toward violent behaviors by reducing self-efficacy to avoid violence.

Limitations

There are several limitations to this study. The measures of peer violent behaviors and peers’ expectations of violence did not represent the variety of violent behaviors in which adolescents could engage; however, both measures had indicators of less and more vio-lent behaviors. Ngwe and colleagues (2004) have also demonstrated that estimates of friends’ behaviors were a significant mediator. While both of the current scales consisted of only two items, both measures were sound and made significant and interesting contri-butions to the overall model while addressing the key issue of physical violence in this study. Future research should determine whether this model works for different levels and kinds of violence when accounting for adolescents’ reports of peer violence by including a wider range of violent behaviors, and also examine gender differences.

This study featured reports from male adolescent African Americans only. Some research has demonstrated the additional benefit of having both parents and youth involvement in studies (Lewis et al., 2012). The exclusion of parent reports in this study, however, allowed us to explore more fully the adolescents’ perspectives and perceptions, especially with regard to the influence of cognitive risks and resources such as perceptions of peer and parent behaviors and expectations, and self-efficacy to avoid violence. An investigation of parent-based factors (e.g., monitoring, parenting styles, parent mental and physical health, and racial identity) that might influence adolescent behavior may also prove illuminating to the literature.

The current data suggest a sample of African American boys who engaged in high lev-els of physical violence evidenced by high reports of threatening to beat up someone and physical fighting. The data, however, provide no context for this. For instance, the reports of threatening and physical fighting may actually be age-appropriate sibling and peer dis-agreements which do not meet the higher descriptor connotation of violence. On the other hand, the data also contain more normally low levels of serious acts of violence (e.g., fights leading to injury, threats to cutting and stabbing, and carrying weapons). This allows the study to account for both normative – and high-level violence without limiting variance.

Attrition as well as inconsistencies or absence of some of the measures due to planned missingness across the intervention made the cross-sectional design most appropriate for the current study. Although our study is cross-sectional, like much of the research that investigates risks and protective factors for youth violence, we contribute to the literature by exploring structural models to determine potential pathways of influence among the

Thomas et al. 17

risks and protective factors. The findings do not address issues of causality; however, they identify multiple associations among risk and protective factors and violent behav-ior for boys who have witnessed physical violence, and highlight the significant role that boys’ own self-efficacy plays in deterring violent behavior. Additional research should examine the influence of risk and protective factors across time. Investigating these issues from a longitudinal perspective would determine how the protective value of key factors might change across developmental periods. In addition, this study focused on African American boys; thus, the influence of gender was not examined. An investiga-tion of gender differences might reveal how these risk and protective factors function differently for adolescent boys and girls in the same high-risk contexts.

African American boys are more likely than any other ethnic group, male or female, to develop the negative outcomes associated with living in high-risk neighborhoods (Neumann, Barker, Koot, & Maughan, 2010). Those African American boys who wit-ness physical violence can be resilient even in these high-risk contexts and the two main sources of strength were boys’ own self-efficacy beliefs and the conveyed non-violent expectations of parents. Witnessing violence remains a significant threat to positive development for African American boys, but parents’ non-violent expecta-tions strengthen boys’ self-efficacy to avoid violence. Peer violent behaviors and expectations increase the possibility of violent behaviors; however, boys’ self-efficacy to avoid violence reduces the likelihood of this outcome. These findings are insightful because they focus positive attention on African American boys, and identify possible resources for working with families. The study further accentuates the positive influ-ence of parents in protecting and supporting African American boys—an especially critical concern for adolescent and families living in poor, urban neighborhoods.

Implications

This study extends the current literature on how peers, parents, and youth’s individual strengths influence a major risk factor for the development of youth violence among African American boys who have witnessed physical violence. The findings of this study argue for the inclusion of adolescents’ cognitive skills/strengths in future efforts directed at attenuating and managing the effects of witnessing physical violence. Perception of peer violence and peer violent expectations are key mediators of witness-ing violence. Adolescents’ awareness of parents’ non-violent expectations proves criti-cal as these expectations boost African American boys’ self-efficacy to avoid violence (Ohene et al., 2006). Thus, skill building through parent reinforcement may prove effective in developing prevention efforts aimed at improving youth outcomes. In con-junction, community groups, health care providers, and public health specialists must work with parents to identify plausible and adaptive non-violent coping and problem-solving strategies that parents can then convey to their sons (Hall et al., 2012).

This study suggests that the influence of parents cannot be underestimated even during the peer-driven adolescence period. Social and mental health professionals should con-tinue to encourage positive parenting practices, especially those practices that facilitate parents’ transmission of critical messages, and positive behavior modeling for their sons


(i.e., communication, monitoring). Some research has suggested that parents who use physical punishment as their main parenting strategy may unwittingly convey a provi-olence message, especially for boys who may have witnessed violence in the neighbor-hoods (Ohene et al., 2006). Research should explore the specific parenting strategies and techniques that are used to convey these messages, as well as the role of the community context, including the influence of “parental figures” (e.g., grandparents, church elders), in subsidizing parental efforts. In addition, researchers should examine the effect of wit-nessing violence perpetrated by a parent or against a parent on youth violence behavior.

Schools and communities may find it beneficial to work with relevant agencies to help reduce community violence, and reduce adolescents’ exposure to these violent experiences. This kind of intervention may be even more essential as African American youth make meaning of the growing trend of deaths of African Americans males at the hands of the Police. Community efforts could support parent messages that encourage norms and beliefs that encourage adolescents to avoid violence. The school climate should encourage non-violence and challenge normalized youth violence. Schools must work diligently as part of the process of dispelling the often-misguided notion that “Everyone is doing it” when it comes to violence and other risky behaviors. School-based programs that promote non-violence and encourage norms against vio-lence can reduce this behavior (Flay et al., 2004; Jagers et al., 2009) by adjusting adolescents’ perception of peer behaviors and expectations. School-based programs may also reduce the influence of peer pressure and the perception of violence as a means for establishing social status or solving social conflict.

Parent expectations of non-violence and self-efficacy to avoid violence are both associated with negative peer behaviors and expectations. School-based violence pre-vention programs should include components that focus on addressing social- cognitive skills (Boxer & Dubow, 2001), and take special notice of boys who have witnessed physical violence. Although individual-level factors are an important target of prevention efforts, interventions may have a limited effect if they do not also address relevant environmental influences within the family, peer group, school, and commu-nity domains (Farrell et al., 2010). Our study suggests that for male African American adolescents who have witnessed physical violence, violent behaviors may best be understood as functioning through peer and parent expectations, peer behaviors, and adolescents’ self-efficacy to avoid violence.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Aban Aya Youth Project was supported by grants to Brian R. Flay (PI) from the Office of Minority Health administered by the National Institute for Child Health and Human Development (U01HD30078) and the National Institute on Drug Abuse (R01DA11019).

Thomas et al. 19

References

Arbuckle, J. L. (2009). Amos 18 user’s guide. Chicago, IL: Amos Development Corporation.Assari, S., Smith, J. R., Caldwell, C. H., & Zimmerman, M. A. (2015). Gender differences

in longitudinal links between neighborhood fear, parental support, and depression among African American emerging adults. Societies, 5, 151-170. doi:10.3390/soc5010151

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. doi:10.1037//0022-3514.51.6.1173

Bellair, P. (2000). Informal surveillance and street crime: A complex relationship. Criminology, 38, 137-167. doi:10.1111/j.1745-9125.2000.tb00886.x

Bennett, M. D., Jr., & Fraser, M. W. (2000). Urban violence among African American males: Integrating family, neighborhood, and peer perspectives. Journal of Sociology & Social Welfare, 27, 93-117.

Berkowitz, A. D. (2004). An overview of the social norms approach. In L. Lederman & L. Stewart (Eds.), Changing the culture of college drinking: A socially situated health com-munication campaign (pp. 187-208). Cresskill, NJ: Hampton Press.

Boxer, P., & Dubow, E. F. (2001). A social-cognitive information-processing model for school-based aggression reduction and prevention programs: Issues for research and practice. Applied & Preventive Psychology, 10, 77-192. doi:10.1016/S0962-1849(01)80013-5

Brody, G. H., Kogan, S. M., Chen, Y. F., & Murry, V. M. (2008). Long-term effects of the strong African American families program on youths’ conduct problems. Journal of Adolescent Health, 43, 474-481. doi:10.1016/j.jadohealth.2008.04.016

Buka, S. L., Stichick, T. L., Birdthistle, I., & Earls, F. J. (2010). Youth exposure to violence: Prevalence, risks, and consequences. American Journal of Orthopsychiatry, 71, 298-310. doi:10.1037//0002-9432.71.3.298

Bureau of Justice Statistics. (2001). Criminal victimization: 2000. Changes, 1993-2000 (NCJ 187007). Retrieved from http://www.bjs.gov/content/pub/pdf/cv00.pdf

Byrne, B. M. (2013). Structural equation modeling with AMOS: Basic concepts, applications, and programming. New York, NY: Routledge.

Calvete, E., & Orue, I. (2011). The impact of violence exposure on aggressive behavior through social information processing in adolescents. American Journal of Orthopsychiatry, 81, 38-50. doi:10.1111/j.1939-0025.2010.01070.x

Caprara, G. V., Regalia, C., & Bandura, A. (2002). Longitudinal impact of perceived self-regulatory efficacy on violent conduct. European Psychologist, 7, 63-69. doi:10.1027//1016-9040.7.1.63

Cassidy, E. F., & Stevenson, H. C., Jr. (2005). They wear the mask: Hypervulnerability and hyper-masculine aggression among African American males in an urban remedial disciplinary school. Journal of Aggression, Maltreatment & Trauma, 11, 53-74. doi:10.1300/J146v11n04_03

Centers for Disease Control and Prevention. (2011). Youth risk behavior surveillance—United States. MMWR, Surveillance Summaries 2012, 61 (no. SS-4). Retrieved from www.cdc.gov/mmwr/pdf/ss/ss6104.pdf

Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. (2010). Web-based injury statistics query and reporting system (WISQARS). Retrieved from www.cdc.gov/injury

Copeland-Linder, N., Jones, V. C., Haynie, D. L., Simons-Morton, B. G., Wright, J. L., & Cheng, T. L. (2007). Factors associated with retaliatory attitudes among African American adolescents who have been assaulted. Journal of Pediatric Psychology, 32, 760-770. doi:10.1093/jpepsy/jsm007


Dishion, T. J., & Owen, L. D. (2002). A longitudinal analysis of friendships and substance use: Bidirectional influence from adolescence to adulthood. Developmental Psychology, 38, 480-491. doi:10.1037//0012-1649.38.4.480

Eccles, J. S., Wigfield, A., & Schiefele, U. (1998). Motivation to succeed. In W. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology (5th ed., Vol. III, pp. 1017-1095). New York, NY: Wiley.

Eisinga, R., Grotenhuis, M., & Pelzer, B. (2013). The reliability of a two-item scale: Pearson, Cronbach or Spearman–Brown? International Journal of Public Health, 58, 637-642. doi:10.1007/s00038-012-0416-3

Farrell, A. D., Erwin, E. H., Bettencourt, A., Mays, S., Vulin-Reynolds, M., Sullivan, T., & Meyer, A. (2008). Individual factors influencing effective non-violent behavior and fighting in peer situations: A qualitative study with urban African American adolescents. Journal of Clinical Child & Adolescent Psychology, 37, 397-411. doi:10.1080/15374410801955821

Farrell, A. D., Mays, S., Bettencourt, A., Erwin, E. H., Vulin-Reynolds, M., & Allison, K. W. (2010). Environmental influences on fighting versus non-violent behavior in peer situa-tions: A qualitative study with urban African American adolescents. American Journal of Community Psychology, 46, 19-35. doi:10.1007/s10464-010-9331-z

Farrington, D. P., & Ttofi, M. M. (2011). Bullying as a predictor of offending, violence and later life outcomes. Criminal Behaviour and Mental Health, 21, 90-98. doi:10.1002/cbm.801

Fergus, S., & Zimmerman, M. A. (2005). Adolescent resilience: A framework for understand-ing healthy development in the face of risk. Annual Review of Public Health, 26, 399-419. doi:10.1146/annurev.publhealth.26.021304.144357

Finkelhor, D., Turner, H., Ormrod, R., & Hamby, S. (2009). Violence, abuse, and crime expo-sure in a National Sample of Children and Youth. Pediatrics, 124, 1411-1423.

Flay, B. R., Graumlich, S., Segawa, E., Burns, J. L., & Holliday, M. Y. (2004). Effects of 2 prevention programs on high-risk behaviors among African American youth: A random-ized trial. Archives of Pediatrics & Adolescent Medicine, 158, 377-384. doi:10.1001/arch-pedi.158.4.377

Foney, D. M., & Cunningham, M. (2002). Why do good kids do bad things? Considering mul-tiple contexts in the study of antisocial fighting behaviors in African American urban youth. Journal of Negro Education, 71, 143-157. doi:10.2307/3211233

Gabrielson, R., Jones, R. G., & Sagara, E. (2014, October 10). Deadly force, in Black and White: A ProPublica analysis of killings by police shows outsize risk for young Black males. Retrieved from http://www.propublica.org/article/deadly-force-in-black-and-white

Hall, J. E., Simon, T. R., Mercy, J. A., Loeber, R., Farrington, D. P., & Lee, R. D. (2012). Centers for Disease Control and Prevention’s expert panel on protective factors for youth violence perpetration: Background and overview. American Journal of Preventive Medicine, 43, S1-S7. doi:10.1016/j.amepre.2012.04.026

Hawkins, J. D., Herrenkohl, T. I., Farrington, D. P., Brewer, D., Catalano, R. F., Harachi, T. W., & Cothern, L. (2000). Predictors of youth violence. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

Haynie, D. L., Silver, E., & Teasdale, B. (2006). Neighborhood characteristics, peer networks, and adolescent violence. Journal of Quantitative Criminology, 22, 147-169. doi:10.1007/s10940-006-9006-y

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure anal-ysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1-55. doi:10.1080/10705519909540118

Thomas et al. 21

Jagers, R. J., Morgan-Lopez, A. A., & Flay, B. R. (2009). The impact of age and type of inter-vention on youth violent behaviors. The Journal of Primary Prevention, 30, 642-658. doi:10.1007/s10935-009-0200-1

Jagers, R. J., Morgan-Lopez, A. A., Howard, T. L., Browne, D. C., & Flay, B. R. (2007). Mediators of the development and prevention of violent behavior. Prevention Science, 8, 171-179. doi:10.1007/s11121-007-0067-4

Kliewer, W., Cunningham, J. N., Diehl, R., Parrish, K. A., Walker, J. M., Atiyeh, C., & Mejia, R. (2004). Violence exposure and adjustment in inner-city youth: Child and caregiver emotion regulation skill, caregiver-child relationship quality, and neighborhood cohesion as protec-tive factor. Journal of Clinical Child & Adolescent Psychology, 33, 477-487. doi:10.1207/s15374424jccp3303_5

Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). New York, NY: Guilford Press.

Laird, R. D., Jordan, K. Y., Dodge, K. A., Pettit, G. S., & Bates, J. E. (2001). Peer rejection in childhood, involvement with antisocial peers in early adolescence, and the development of externalizing behavior problems. Development and Psychopathology, 13, 337-354. doi:10.1017/S0954579401002085

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, 29-48. doi:10.1007/s10464-005-6231-8

Lei, M., & Lomax, R. G. (2005). The effect of varying degrees of nonnormality in structural equa-tion modeling. Structural Equation Modeling, 12, 1-27. doi:10.1207/s15328007sem1201_1

Lewis, T., Thompson, R., Kotch, J. B., Proctor, L. J., Litrownik, A. J., English, D. J., . . . Dubowitz, H. (2012). Parent-youth discordance about youth-witnessed violence: Associations with trauma symptoms and service use in an at-risk sample. Child Abuse & Neglect, 36, 790-797. doi:10.1016/j.chiabu.2012.09.009

Lindstrom Johnson, S. R., Finigan, N. M., Bradshaw, C. P., Haynie, D. L., & Cheng, T. L. (2011). Examining the link between neighborhood context and parental messages to their adolescent children about violence. Journal of Adolescent Health, 49, 58-63. doi:10.1016/j.jadohealth.2010.10.014

Loeber, R., Farrington, D. P., Stouthamer-Loeber, M., & White, H. R. (2008). Violence and serious theft: Development and prediction from childhood to adulthood. New York, NY: Routledge.

Lösel, F., & Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. American Journal of Preventive Medicine, 43, S8-S23. doi:10.1016/j.amepre.2012.04.029

McMahon, S. D., Todd, N. R., Martinez, A., Coker, C., Sheu, C. F., Washburn, J., & Shah, S. (2013). Aggressive and prosocial behavior: Community violence, cognitive, and behav-ioral predictors among urban African American youth. American Journal of Community Psychology, 51, 407-421. doi:10.1007/s10464-012-9560-4

Miller, D. T., & McFarland, C. (1991). When social comparison goes awry: The case of plural-istic ignorance. In J. Suls & T. Wills (Eds.), Social comparison: Contemporary theory and research (pp. 287-313). Hillsdale, NJ: Lawrence Erlbaum.

Neumann, A., Barker, E. D., Koot, H. M., & Maughan, B. (2010). The role of contextual risk, impulsivity, and parental knowledge in the development of adolescent antisocial behavior. Journal of Abnormal Psychology, 119, 534-545. doi:10.1037/a0019860

Ngwe, J. E., Liu, L. C., Flay, B. R., & Segawa, E. (2004). Violence prevention among African American adolescent males. American Journal of Health Behavior, 28(Suppl. 1), S24-S37. doi:10.5993/AJHB.28.s1.4


Ohene, S. A., Ireland, M., McNeely, C., & Borowsky, I. W. (2006). Parental expectations, physical punishment, and violence among adolescents who score positive on a psychoso-cial screening test in primary care. Pediatrics, 117, 441-447. doi:10.1542/peds.2005-0421

Orpinas, P., Murray, N., & Kelder, S. (1999). Parental influences on students’ aggres-sive behaviors and weapon carrying. Health Education & Behavior, 26, 774-787. doi:10.1177/109019819902600 603

Pardini, D. A., Loeber, R., Farrington, D. P., & Stouthamer-Loeber, M. (2012). Identifying direct protective factors for nonviolence. American Journal of Preventive Medicine, 43, S28-S40. doi:10.1016/j.amepre.2012.04.024

Prinstein, M. J., & Cillessen, A. H. (2003). Forms and functions of adolescent peer aggression associated with high levels of peer status. Merrill-Palmer Quarterly, 49, 310-342.

Prinstein, M. J., & Wang, S. S. (2005). False consensus and adolescent peer contagion: Examining discrepancies between perceptions and actual reported levels of friends’ deviant and health risk behaviors. Journal of Abnormal Child Psychology, 33, 293-306. doi:10.1007/s10802-005-3566-4

Richards, M. H., Larson, R., Miller, B. V., Luo, Z., Sims, B., Parrella, D. P., & McCauley, C. (2004). Risky and protective contexts and exposure to violence in urban African American young adolescents. Journal of Clinical Child & Adolescent Psychology, 33, 138-148. doi:10.1207/S15374424JCCP3301_13

Riner, M. E., & Saywell, R. M. (2002). Development of the social ecology model of adolescent interpersonal violence prevention (SEMAIVP). Journal of School Health, 72(2), 65-70. doi:10.1111/j.1746-1561.2002.tb06517.x

Robinson, W. L. V., Paxton, K. C., & Jonen, L. P. (2011). Pathways to aggression and violence among African American adolescent males: The influence of normative beliefs, neigh-borhood, and depressive symptomatology. Journal of Prevention & Intervention in the Community, 39, 132-148. doi:10.1080/10852352.2011.556572

Rosich, K. J. (2007). Race, ethnicity, and the criminal justice system. Washington, DC: American Sociological Association.

Salzinger, S., Ng-Mak, D. S., Feldman, R. S., Kam, C. M., & Rosario, M. (2006). Exposure to community violence processes that increase the risk for inner-city middle school children. The Journal of Early Adolescence, 26, 232-266. doi:10.1177/0272431605285712

Shaw, C., & McKay, H. (1942). Juvenile delinquency and urban areas. Chicago: University of Chicago Press.

Sickmund, M., & Puzzanchera, C. (Eds.). (2014). Juvenile offenders and victims: 2014 national report. Pittsburgh, PA: National Center for Juvenile Justice.

Sieving, R. E., McNeely, C. S., & Blum, R. W. (2000). Maternal expectations, mother-child connectedness, and adolescent sexual debut. Archives of Pediatrics & Adolescent Medicine, 154, 809-816. doi:10.1001/archpedi.154.8.809

Sim, T. N. (2000). Adolescent psychosocial competence: The importance and role of regard for parents. Journal of Research on Adolescence, 10, 49-64. doi:10.1207/SJRA1001_3

Snyder, H. N. (2000). Juvenile arrests, 1999. Washington, DC: Office of Juvenile Justice and Delinquency Prevention.

Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in structural equation models. In S. Leinhardt (Ed.), Sociological methodology, 1985 (pp. 290-312). Washington, DC: American Sociological Association.

Stein, B. D., Jaycox, L. H., Kataoka, S., Rhodes, H. J., & Vestal, K. D. (2003). Prevalence of child and adolescent exposure to community violence. Clinical Child and Family Psychology Review, 6, 247-264. doi:10.1023/B:CCFP.0000006292.6107 2.d2

Thomas et al. 23

Stewart, E. A., & Simons, R. L. (2006). Structure and culture in African American adoles-cent violence: A partial test of the “code of the street” thesis. Justice Quarterly, 23, 1-33. doi:10.1080/07418820600552378

Thomas, A., & Hope, E. C. (2015). Walking away hurt, walking around scared: A cluster analy-sis of violence exposure among young Black males. Journal of Black Psychology. Advance online publication. doi:10.1177/0095798415603539

Vitaro, F., Brendgen, M., & Tremblay, R. E. (2002). Reactively and proactively aggressive children: Antecedent and subsequent characteristics. Journal of Child Psychology and Psychiatry, 43, 495-505. doi:10.1111/1469-7610.00040

Wilkinson, D. L. (2003). Guns, violence, and identity among African American and Latino youth. New York, NY: LFB Scholarly Publishing.

Zimmerman, M. A., Stoddard, S. A., Eisman, A. B., Caldwell, C. H., Aiyer, S. M., & Miller, A. (2013). Adolescent resilience: Promotive factors that inform prevention. Child Development Perspectives, 7, 215-220. doi:10.1111/cdep.1204

Author Biographies

Alvin Thomas, PhD, is an Assistant Professor and Associate Director for the Center for excel-lence in Diversity at Palo Alto University. He was trainied as a Clinical Psychologist, and com-pleted a Post Doctoral Fellowship in the Department of Psychiatry at the University of Michigan. Dr. Thomas’ work focuses on positive youth development, and risk and protective factors related to the health of African-American men and boys, including nonresident father involve-ment in parenting.

Cleopatra H. Caldwell, PhD, is a Professor of Health Behavior and Health Education and Director of the Center for Research on Ethnicity, Culture, and Health (CRECH) at the School of Public Health at the University of Michigan. She is also a Faculty Associate with the Program for Research on Black Americans (PRBA) at the Institute for Social Research. Her research focuses on psychosocial and environmental factors influencing the health and health behaviors of ethnically diverse adolescents, particularly in the area of health disparities.

Shervin Assari, MD, MPH, is a research faculty at the Department of Psychiatry, with a joint appointment at the Center for Research on Ethnicity, Culture, and Health, University of Michigan, Ann Arbor. He studies mental health disparities due to ethnicity, gender, and place. Assari is the author of 130 peer review papers in PubMed.

Robert J. Jagers is an Associate Professor of Education and Psychology in the School of Education at the University of Michigan. He is the director of Wolverine Pathways, the univer-sity’s new diversity pipeline initiative. His basic and applied research focuses on early adoles-cents’ academic, social and emotional learning and youth civic activism, especially among boys and young men.

Brian Flay, DPhil, is a Professor of Health Promotion & Health Behavior in the School of Social and Behavioral Health Sciences in the College of Public Health and Human Sciences at Oregon State University. His training was in social psychology at Waikato University in New Zealand. Most of Dr. Flay’s work has concerned the development and evaluation of programs for the prevention of substance abuse, violence, and AIDS. Recent studies focus on positive youth development, including social-emotional and character development.

Documents

You Do What You See: How © 2016 SAGE Publications ...people.oregonstate.edu/~flayb/MY PUBLICATIONS... · Alvin Thomas1 ,2, Cleopatra H. Caldwell3 4, Shervin Assari2,3, Robert J