13 Israeli Adolescents Exposed

8/2/2019 13 Israeli Adolescents Exposed

1/28

http://jea.sagepub.com/Adolescence

The Journal of Early

http://jea.sagepub.com/content/31/4/577The online version of this article can be found at:

DOI: 10.1177/0272431610366247

April 20102011 31: 577 originally published online 13The Journal of Early AdolescenceShahar

Kathryn A. Brookmeyer, Christopher C. Henrich, Guina Cohen and GolanProtective Role of Social Support

Israeli Adolescents Exposed to Community and Terror Violence: The

Published by:

http://www.sagepublications.com

can be found at:The Journal of Early AdolescenceAdditional services and information for

http://jea.sagepub.com/cgi/alertsEmail Alerts:

http://jea.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://jea.sagepub.com/content/31/4/577.refs.htmlCitations:

What is This?

- Apr 13, 2010Proof

- Jun 16, 2011Version of Record>>

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

2/28

Journal of Early Adolescence

31(4) 577603

The Author(s) 2011Reprints and permission:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/0272431610366247http://jea.sagepub.com

Israeli Adolescents

Exposed to

Community and

Terror Violence:

The Protective Role

of Social Support

Kathryn A. Brookmeyer1,

Christopher C. Henrich2, Guina Cohen3,

and Golan Shahar3

Abstract

This study investigates how social support may protect Israeli early adolescentswho have witnessed community violence from engaging in violent behavior whenthey have also witnessed terror violence. The study examines how support fromparents, school, and friends could serve as protective, despite the interactive riskeffects of witnessing community and terror violence. In general, results indicatethat support from parents operated as a protective factor, whereas supportfrom friends acted as a risk by increasing the likelihood of violent behavior.Support from school has both a protective and risk effect, depending on thetype of violence exposure witnessed. The extent to which these facets of socialsupport operated as risk or protective factors is conditional on whether youthhad also witnessed terror violence, and terror violence appear to moderateeach of the three facets of support differently. The implications of these findingsand intervention initiatives for Israeli youth are discussed.

1Centers for Disease Control and Prevention2Georgia State University3Ben-Gurion University of the Negev, Israel

Corresponding Author:

Kathryn A. Brookmeyer, Division of Violence Prevention, National Center for Injury Prevention

and Control Centers for Disease Control and Prevention

Email: [email protected]

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

3/28

578 Journal of Early Adolescence 31(4)

Keywords

early adolescence, community violence, terrorism, social support, social ecology

Youth violence is a public health problem in countries throughout the world

(Krug, Dahlberg, Mercy, Zwi, & Lozano, 2002). In this study, we focused on

violence in two communities in the Negev region of southern Israel, where

exposure to violent crime is a pressing risk in the lives of adolescents (Central

Bureau of Statistics, 2007; Israel Police, 2007). In southern Israel, juvenile

arrests accounted for 12% of all crimes in 2006 (Israel Police, 2007), and in

Jerusalem, 76% of adolescent males in school reported being in a physical

fight within the past year (Krug et al., 2002). There is also evidence that some

forms of youth violence may be on the rise. For example, more than 1,300 youth

carried a knife in Israel in 2006, representing nearly a 400% increase since

1996, and in the town of Sderot, one of the communities in this study, there

was a 6.8% increase in crime overall in 2006 (Israel Police, 2007).

Witnessing community violence exposure, such as seeing someone threat-

ened, attacked, wounded, or killed, has been shown to adversely affect youth

development (Osofsky, Wewers, Hann, & Fick, 1993). Witnessing violence

is a pervasive threat in Southern Israel (Central Bureau of Statistics, 2007),

where the risk of seeing rocket attacks, community violence, and the damageresulting is a pressing environmental stressor. Just as alarming as the high

prevalence of witnessing violence is the danger that adolescents who witness

violence will continue the cycle of violence (Widom, 1989) in their commu-

nities by perpetrating violence themselves. Bandura (1986) proposed that

such exposure to violence can teach new aggressive behaviors to children as

well as reduce inhibition to act in a violent manner. Indeed, witnessing vio-

lence is among the most consistent predictors of the use of aggressive behavior

and violence (Miller, Wasserman, Neugebauer, Gorman-Smith, & Kamboukos,1999; Schwab-Stone, Jones, Henrich, Leckman, & Rochkin, 2003), even

after controlling for previous levels of aggression (Gorman-Smith & Tolan,

1998). Gorman-Smith, Henry, and Tolan (2004) found that higher levels of

violence exposure in middle adolescence were related to an increased rate of

violence perpetration in late adolescence, though this relation changed with

family functioning.

Because of the persistent threat of witnessing violence (Central Bureau of

Statistics, 2007), the study focused on this unique form of violence exposure.However, we also measured victimization by violence, even though it is a rarer

event in this population (e.g., Comer & Kendell, 2007), because physical prox-

imity to violent events can be important to understanding the effects of vio-

lence exposure (e.g., Nader, Pynoos, Fairbanks, & Frederick, 1990). Although

infrequently measured in concert, these two forms of violence can have

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

4/28

Brookmeyer et al. 579

distinct behavioral outcomes (Fowler, Tompsett, Braciszewski, Jacques-Tiura,

& Baltes, 2009). For example, in past research, victimization by violence is

often linked to internalizing symptoms, such as depression and posttraumaticstress disorder (PTSD; e.g., Fitzpatrick, 1993; Nader et al., 1990), whereas

witnessing violence is among the most consistent predictors of aggression

(e.g., Schwab-Stone et al., 2003), especially for adolescents (Fowler et al.,

2009). Therefore, although this study took victimization into account, its pri-

mary focus was on witnessing violence due to the local context in Israel and

prior research linking witnessing to aggression and violence perpetration.

To understand the consequences of witnessing community violence on

adolescent well-being and behavior, one must take into account the larger

social environment in which violence exposure takes place (Bronfenbrenner,

1977). We framed this study using an ecological perspective. According to

this approach, understanding adolescent behavioral risk requires the consid-

eration of ecological effects, that is, the social contexts in which the risk

behavior occurs (Bronfenbrenner, 1977; Cicchetti & Lynch, 1993). Guided

by this approach, researchers examined the ways in which risk and protective

factors interact with one another across social contexts and their association

with adolescent violent behavior.

Using this ecological framework, the effect of witnessing community vio-lence on adolescent adjustment is influenced not only by the immediate con-

text of violence exposure but also by the interactions of multiple intersecting

social contexts of development (Bronfenbrenner & Morris, 1998). The ecol-

ogy of witnessing community violence in Israel is unique in that community

violence occurs in the context of an ongoing threat of another form of vio-

lence: terrorism. Terrorist attacks are defined by the U.S. Code, Section 2656f

as premeditated, politically motivated violence perpetrated against noncom-

batant targets by sub-national groups with the intent to threaten and intimi-date a population (Comer & Kendell, 2007). In 2006, more than 2,000 terror

violence attacks occurred in Israel, most of which consisted of rockets fired

into the western Negev region of the country (Israeli Ministry of Foreign

Affairs, 2008). Much of this terror violence occurred in the same communi-

ties where community violence also takes place. Within the western Negev

region, Sderot has endured has been subjected to thousands of rocket and

mortar attacks over the past 10 years, with increasing intensity from 2006 to

2008 (Israeli Ministry of Foreign Affairs, 2008). Comparatively, in Dimona,the other community under study, there has been less direct witnessing of

terror attacks, but, for many youth, exposure to terror violence has been

endured indirectly, from hearing about injuries and media coverage. The

presence of such exposure in both communities presents a pressing public

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

5/28

580 Journal of Early Adolescence 31(4)

health issue in terms of high rates of trauma and youth violence (Henrich &

Shahar, 2008; Siegel-Itzkovich, 2009).

Accordingly, we investigated the effects of two interacting environmentalcontextsexposure to community and exposure to terror violenceon

adolescent behavior. Although both community and terror violence undermine

the basic assumption that a community is safe, there are important distinctions

between these two forms of violence exposure, which may have differential

consequences for adolescent development and behavioral adjustment. Witness-

ing community and terror violence may affect adolescent outcomes through

distinct pathways; one is perpetrated by residents within communities,

whereas the other is motivated by political aims from outside populations

(Kazdin, 1997). As well, community violence is often experienced as an ongo-

ing environmental threat (Osofsky et al., 1993; Richters & Martinez, 1993),

whereas terror violence is by intention erratic and unpredictable (Comer &

Kendell, 2007). Although research has studied the effects of witnessing com-

munity violence on adolescent development, much less is known about the

developmental consequences of terror violence. Virtually no research has

investigated the interactive effect of witnessing community and terror violence

on adolescent adjustment and behavior (e.g., Krug et al., 2002; La Greca, 2007).

Understanding the Effect of Witnessing

Violence: Violent-Behavior Outcomes

As discussed, of particular concern in researching the effects of witnessing

violence on adolescent behavior is the idea that violence begets violence and

that exposure to violence can contribute to a cycle of violence (Widom, 1989).

Although evidence suggests a link between community violence exposure

and violent behavior among adolescents (e.g., Brookmeyer, Henrich, &Schwab-Stone, 2005; Gorman-Smith et al., 2004), minimal research has been

conducted on the relationship between terror violence exposure and violent

behavior outcomes (Comer & Kendell, 2007; La Greca, 2007). Recent work

on Israeli populations may suggest an association between perceived terror

violence danger and violent behavior (Even-Chen & Irzhaky, 2007), but

results have been mixed (Hobfoll, Canetti-Nisim, & Johnson, 2006; Ziv,

Kruglanski, & Shulman, 1974). In addition, no research to date has taken an

ecological approach to explore how witnessing terror violence may interactwith community violence exposure to increase adolescents risk of perpetrat-

ing violent behavior. This study makes a unique contribution by using an

ecological framework to address the potential interacting effects of witness-

ing community and terror violence and how these distinct environmental

contexts may be linked to perpetrating violent behavior.

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

6/28

Brookmeyer et al. 581

Role of Protection: Social Support

Using an ecological perspective, we investigated the proximal aspects ofadolescents social environments that may play a protective role by interact-

ing with the risk of witnessing violence and youth violent behavior outcomes.

Social support may buffer the link between the environmental risks of wit-

nessing terror and community violence and violent behavior outcomes. Much

work has been completed on how social support may be associated with com-

munity violence and violent behavior, but few studies have examined the

moderating effect of social support on terror violence. Social support may be

a critical asset in promoting well-being by buffering environmental stress

(Cohen & Willis, 1985; DuBois, Felner, Meares, & Krier, 1994). Proximal

social contexts of support, such as support from parents, school, and friends,

may be vital in attenuating the effects of community and terror violence

exposure. Examining how these important contexts may curb violent behavior,

despite risk exposure, is key to understanding how to bolster protection for

adolescents (Bronfenbrenner & Morris, 1998; Masten, 1999).

Support from parents has been cited widely as critical in protecting ado-

lescents from environmental risk and promoting positive development

(Garmezy, 1991; Resnick et al., 1997). Research indicates that parent supportis linked to decreased levels of aggression and violent behavior (Blum,

Ireland, & Blum, 2003) and that parent support may buffer the effects of

stress on adolescent aggression and violent behavior (Brookmeyer et al.,

2005; Dubow, Edwards, & Ippolito, 1997; Gorman-Smith et al., 2004).

Social support from school personnel can also play a protective role for

youth (Ozer & Weinsten, 2004). Schools serve as an important context for

youth development and often function as a place of relative safety within

violent communities (National Center for Educational Statistics, 1997). Feel-ings of connectedness to school have been found to positively influence ado-

lescent adjustment (Garmezy, 1991; Resnick et al., 1997) as well as to decrease

aggressive behavior (Haynes, Emmons, & Ben-Avie, 1997). However, the

degree to which a supportive school can operate as a buffer against violent

behavior when youth are exposed to community violence is unclear (Henrich,

Brookmeyer, Shahar, & Kidd, 2005; Ozer, 2005).

We also examined the peer context. Social support from friends may oper-

ate quite differently than social support from adult sources (DuBois et al.,2002). There is evidence that social support from friends can be associated

with increases in behavior problems (DuBois et al., 2002; Dubow et al., 1997).

This potentially iatrogenic effect of peer social support is not surprising, given

that aggressive peers tend to clique together in friendship groups (Cairns,

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

7/28

582 Journal of Early Adolescence 31(4)

Cairns, Neckerman, Gest, & Gariepy, 1988) and can influence one another to

be more aggressive over time, particularly when they report high-quality

friendships with their aggressive peers (Espelage, Holt, & Henkel, 2003).Although there is a large body of research on the protective nature of

social support, little is known about the role of social support within the con-

text of terror violence exposure. Preliminary findings suggest that social sup-

port may have a protective effect in buffering the effect of terror violence

exposure on adolescent depression (Henrich & Shahar, 2008). To date, no

work has examined whether social support moderates adolescents response

to witnessing community violence when terror violence is also present. There-

fore, this study aimed to make a unique contribution by understanding the

interacting effects of witnessing community and terror violence and how dis-

tinct facets of social support can buffer violent behavior outcomes within an

ecological framework, representative of Israeli adolescents complex experi-

ence. In addition, understanding the role of social support in strengthening

adolescent resilience in the face of community and terror violence has impor-

tant implications for the design of interventions to promote adolescent adjust-

ment in the face of environmental risk.

Aims and Hypotheses

Little research has been completed on youths interacting exposure to com-

munity and terror violence and how these two forms of violence may be

linked to violent behavior outcomes, given the potential buffering role of

social support. In understanding how proximal social support from parents,

schools, and friends serve to protect Israeli adolescents witnessing commu-

nity violence from perpetrating violent behavior when terror violence was

also witnessed, we hypothesized that exposure to community violence is asso-ciated with violent behavior, and social support operates to buffer this

association. Nevertheless, when youth witness terror violence in addition to

community violence, we hypothesized that it might be more difficult for social

support to buffer dual exposure on violent behavior outcomes (Garbarino &

Kostelny, 1996), given that parents, friends, and school staff are also exposed

to community and terror violence (Joshi & ODonnell, 2003). By investigat-

ing support from parents, friends, and school staff, we examined whether

distinct aspects of adolescents social ecology may be able to offer protectiondespite the interactive risk effects of community and terror violence expo-

sure. Specifically, we investigated the potentially unique role of friend

support and hypothesized that this source of support may operate differently

from adult sources by potentially increasing the likelihood of aggression

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

8/28

Brookmeyer et al. 583

(e.g., DuBois et al., 2002; Espelage et al., 2003). We hypothesized that parent

and school personnel may be protective when youth experience one form of

violence but that it might be more difficult to protect against the interactingeffects of witnessing community and terror violence in preventing violent

behavior under such high-risk conditions.

Method

Participants and Procedure

This study was approved by the Israeli Ministry of Education and the univer-

sity review committees. Two of the largest secondary schools in Dimona and

Sderot, two economically disadvantaged towns in the Negev region of

Southern Israel, were contacted for recruitment and both schools agreed to

participate. The Sderot and Dimona schools were chosen because they are

relatively large, with a student population of approximately 1,000 and 1,300,

respectively, and because they both accommodate students from Grades 7 to 12,

ideal for future longitudinal designs. Consent was then obtained from parents

of students in Grades 7 to 9 in each school prior to data collection and a pre-

sentation was given to students about the study. Assent was also obtainedfrom the student participants, in accordance with the Israeli Ministry of Edu-

cation. Data were collected using a 175-item paper-and-pencil self-report

survey that assessed adolescents attitudes about school, family, friends, risk

behaviors, and community and terror violence exposure. Trained research

assistants administered the 45-min long assessment to students within each

classroom throughout each school. Students were given a small backpack in

appreciation of their participation.

Data were collected in June 2007 as part of a larger study of Israeli ado-lescents risk and resilience. This cross-sectional data represents the first year

of data available and analyses of this data are ongoing, building from work

within the two communities. The sample from Dimona and Sderot totaled

(N= 179). The public school sampled in Sderot is a religious school and the

one selected in Dimona, a secular one. The relative small number of partici-

pants is due to difficulties in obtaining signed parental consent forms as stu-

dents often lost or forgot the forms, despite the high level of interest in the

study, and teachers occasionally misplaced them. Furthermore, data collec-tion had to be suspended in Sderot out of safety concerns, after an escalation

of rocket attacks. Schoolwide demographics for Sderot and Dimona were not

available at the local level. However, local community experts have indicated

that the students sampled are generally representative of the experiences of

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

9/28

584 Journal of Early Adolescence 31(4)

this age group in the larger towns of Sderot and Dimona, respectively. The

data, however, were not designed to yield a nationally representative sam-

pling frame, and the estimates presented here should not be interpreted as

representative of the Israeli population at large. In preliminary exploratory

analyses, regression results predicting violent behavior in the two communi-

ties were very similar; consequently, analyses are presented for the entiresample, with city included as a covariate. Please see Table 1 for a further

comparison of violence and social support data between the two communi-

ties; both communities are economically depressed and have high levels of

exposure to violence: In terms of terror violence, Sderot youth have been

exposed to terror attacks for almost half their lives. Both samples also expe-

rience a high level of media exposure and have witnessed terror violence by

nature of growing up in Southern Israel.

In our sample, 24.6% of students were from the town of Sderot (n= 44),located 1 km from the Gaza strip, and 75.4% of students were from the town

of Dimona (n = 135). Nearly 60% of students sampled were girls. Results

also indicated that 43.0% of students were in Grade 7, 34.1% were in Grade 8,

and 22.9% were in Grade 9 (M= 7.8, SD= 0.79). The majority of students

Table 1.Violence and Social Support Construct Comparison of the TwoCommunities Under Study

M SD Minimum Maximum

Sderot (n= 44)Committing violence 0.67 0.96 0.00 4.00Victim community violence 0.23 0.47 0.00 2.00Victim terror violence 0.25 0.43 0.00 1.00Witness terror violence 3.34 1.64 0.00 7.00Witness community violence 1.30 1.64 0.00 6.00Parent support 5.36 0.92 3.00 6.00Friend support 3.38 1.93 0.00 6.00

School support 1.31 1.23 0.00 4.00Dimona (n= 135)

Committing violence 0.67 0.93 0.00 5.00Victim community violence 0.37 1.00 0.00 6.00Victim terror violence 0.03 0.17 0.00 1.00Witness terror violence 0.32 0.68 0.00 4.00Witness community violence 1.34 1.61 0.00 6.00Parent support 5.34 1.09 1.00 6.00Friend support 4.45 1.64 0.00 6.00

School support 1.78 1.39 0.00 4.00

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

10/28

Brookmeyer et al. 585

recruited were in the age group of 13 to 15 years (96.0%). Students families

can generally be described as low income and working, as 73.2% of students

parents were both employed.

Measures

All three community-based violence measurescommunity violence expo-

sure, community violence victimization, and violent behaviorwere adapted

from Richters and Martinez (1993) and have repeatedly demonstrated reli-

ability in this age group using this shortened form (e.g., Brookmeyer et al.,

2005; ODonnell, Schwab-Stone, & Muyeed, 2002). All community violence

constructs assess violence experienced in neighborhoods and communities

only, and they are not designed to assess violence experienced in other con-

texts (e.g., the home environment).

Community violence witnessing. The community violence witnessing con-

struct assesses the degree to which adolescents witnessed different forms of

violent behavior in their neighborhoods in the past year. The community vio-

lence witnessing index consisted of six items, and each item was scored as

having never occurred (score of 0) or having occurred one or more times

(score of 1) and summed, such that scores ranged from 0 (no exposure) to6 (exposed to all events at least once). Youth were prompted to consider

whether they had seen the following things happen in your neighborhood in

the past year. Results indicated 21% of the participants reported seeing

someone else chased by gangs or individuals, 30.1% reported seeing some-

one else threatened with physical harm, 40.9% reported seeing someone else

getting beaten up or mugged, 15.6% reported seeing someone else being

attacked or stabbed, 22.6% reported seeing someone seriously wounded, and

5.9% reported seeing someone else shot at with a gun. Overall, 22.3% of theparticipants had been exposed to one of these forms of violence, 15.2% exposed

to 2 forms, and 19.5% had been exposed to 3 or more forms (M= 1.33,

SD= 1.61). Cronbachs alpha was .78.

Community violence victimization. To assess our hypotheses and obtain a

measure of community violence exposure through witnessing violence only,

community violence victimization was statistically controlled for in the

regression analyses. The community violence victimization scale consisted

of 6 items and was dichotomized to assess whether or not adolescents hadbeen victimized by community violence in the past year. Community vio-

lence victimization was scored as having never been experienced(score of 0)

or having been victimized by at least one event(score of 1). The community

violence victimization scale was dichotomized due to low baseline scores

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

11/28

586 Journal of Early Adolescence 31(4)

because this construct was not the focus of the study. Youth were prompted

to consider whether they had the following things happen to you in your

neighborhood in the past year. Results indicated that 4.3% of youth reportedbeing chased by gangs or individuals, 13.4% reported being threatened with

physical harm, 10.8% reported being beaten up or mugged, 2.2% reported

being stabbed, 4.3% reported being seriously wounded in an incident of vio-

lence, and 1.6% reported being shot at with a gun. Overall, 20.1% of adoles-

cents reported that they had been victimized by at least one of these behaviors

(M= .20, SD= 0.40).

Violent behavior. The violent behavior index consisted of 5 items designed

to resemble the content of the community violence witnessing and victimiza-

tion scales, assessing the degree to which adolescents had committed differ-

ent forms of community violence in the past year. Each item was scored as

having never been committed(score of 0) or committed one or more times

(score of 1) and summed, such that scores ranged from 0 (no violence commis-

sion) to 5 (committed all events at least once). Results indicated that 45.2% of

adolescents reported starting a fistfight, 11.3% reported having hurt someone

badly in a physical fight so that they had to seek medical treatment, 7.0%

reported being involved in gang fights, 3.2% reported being arrested by the

police for violence, and 3.8% reported having carried a blade, knife, or gun.Overall, 34.1% of adolescents reported committing one of these behaviors,

and 12.7% reported committing 2 or more of these behaviors (M= .67, SD=

0.93). Cronbachs alpha was .72.

The following terror violence measures were adapted from Kirschenbaum

(2006), which have shown to be reliable within this population (e.g., Henrich

& Shahar, 2008). These measures were designed to assess terror violence

exposure and victimization resulting in physical injury.

Terror violence witnessing. The terror violence exposure scale consisted of7 items and was dichotomized to assess whether or not adolescents had ever

witnessed terror violence. This measured terror violence witnessing or whether

adolescents had been directly exposed through witnessing terror violence

themselves or exposed through the experience of others in the community. In

this scale, 74.3% of youth were exposed to either 0 or 1 instance of terror

violence, with 16.2% being exposed to one event. Therefore, because nearly

three quarters of adolescents were either exposed to no terror violence or to

one terror violence event, the scale was dichotomized. Terror violence wit-nessing was scored as having never been experienced(score of 0) or having

been exposed to terror violenceat least once (score of 1). Items included

were designed to assess witnessing of violence but not whether the adoles-

cent had been injured directly as a result of the terror attack (assessed through

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

12/28

Brookmeyer et al. 587

terror violence victimization measure). Results indicated that 16.1% of the

participants reported ever being present during a terror attack, 9.1% reported

a close family member having ever been injured, 9.7% reported a distantrelative having ever been injured, 10.2% reported a friend having ever been

injured, 17.7% reported ever having an acquaintance injured, 18.3% reported

houses near their homes had been hit by terror attacks in the last 5 years, and

21.5% reported homes in the vicinity of their home being hit by terror attacks

in the past 5 years (M= .42, SD= 0.49).

Terror violence victimization. The terror violence victimization scale assessed

whether or not adolescents had been victimized by being physically injured

or by sustaining damage to their homes as a result of terror violence. Again,

terror violence victimization was controlled for in the analyses to achieve a

measure of terror violence witnessing, parallel to measuring community vio-

lence victimization. Two items were used in constructing this score: whether

or not adolescents were ever physically injured as a result of a terror/missile

attack, and whether or not adolescents home had been hit in the past 5 years

by terror attacks. These 2 items were scored yes (score of 1) or no (score

of 0), summed, and had a maximum score of 1. There were no adolescents

who were victimized by both events. Results indicated that 3.8% of adoles-

cents reported ever having been physically injured as a result of a terrorattack, and 4.3% reported that their own homes were hit by terror attacks in

the past 5 years (M= .08, SD= 0.28).

Social support from parents, friends, and school personnel was measured

through an abbreviated form of the Perceived Social Support Scale (DuBois et al.,

1994), which was designed to assess perceived support and connectedness.

Parent support. Perceived parent support was assessed through 6 items

scored yes (score of 1), no (score of 0), or dont know (score of 0). The

yes responses were summed across items. Cronbachs alpha was .70. Sam-ple items include My family provides me with the support I need, and I trust

the emotional support my family provides. Scores ranged from 0 (no per-

ceived support) to 6 (all support items endorsed;M= 5.35, SD= 1.05).

Friend support. Perceived friend support was assessed through 6 items

scored yes (score of 1), no (score of 0), or dont know (score of 0). The

yes responses were summed across items, with a Cronbachs alpha of .69.

Sample items include I have a friend I can turn to when I am feeling down,

and My friends provide me with the support I need. Scores ranged from 0 (noperceived support) to 6 (all support items endorsed;M= 4.19, SD= 1.78).

School personnel support. Perceived support from school personnel was

assessed through 4 items scored yes (score of 1), no (score of 0), or

dont know (score of 0). The yes responses were summed across items,

by muhammad ardi on October 13, 2011jea.sagepub.comDownloaded from

8/2/2019 13 Israeli Adolescents Exposed

13/28

588 Journal of Early Adolescence 31(4)

with a Cronbachs alpha of .78. Sample items include The staff at my school

helps me solve my problems, and The staff at my school provides me with

the support and encouragement I need. Scores ranged from 0 (no perceivedsupport) to 4 (all support items endorsed;M= 1.66, SD= 1.36).

Results

Of the 179 participants with demographic data, a few (

8/2/2019 13 Israeli Adolescents Exposed

14/28

589

Table2.

CorrelationsAmongtheMeasuredVariables(N=1

79)

1

2

3

4

5

6

7

8

9

10

11

1.

Gender

2.G

rade

.05

3.P

arentemployment

-.0

4

-.03

4.C

ity

-.0

5

.25**

-.0

1

5.V

ictimcom

munity

-.3

8**

.03

.08

.07

violence

6.

Victimterrorviolence

.13

-.13

.07

-.3

4**

.05

7.W

itnessco

mmunity

-.3

0**

-.08

.10

-.0

1

.49**

.01

violence

8.

Witnessterror

.07

-.21**

-.0

2

-.6

5**

.03

.32**

.07

violence

9.

Violentbeh

avior

-.4

0**

-.03

.09

.01

.32**

-.0

4

.47**

.09

10.

Parentsupport

.24**

-.04

.21**

-.0

1

-.3

9**

-.0

4

-.2

2**

-.1

1

-.2

5**

11.

Friendsupport

.30**

.21**

.17*

.26**

-.1

5*

-.1

7*

-.1

0

-.2

2**

-.0

1

.36**

12.

Schoolsup

port

.17*

-.12

.06

.15*

-.1

4

-.1

1

-.1

0

-.1

2

-.1

4

.28**

.36**

Note:Forgender,m

ale=1,female=2.Forpar

entemployment,b

othparentsemployed=2,oneparentemployed=1,noparentsemployed=0.F

orcity

location,S

derot=

1,D

imona=2.Correlationsb

etweendichotomousvariables

representphicoefficients.

*p