Based on a populational survey conducted among 1400 adolescents aged between 12 and 17 years old, the aim of this study is to assess the relationships between their community violence experiences and their psychological health (anger, depressive symp- toms, and post-traumatic stress disorder symptoms). One MANOVA confirms that both boys and girls who report at least one incident of physical community violence present more psychological difficulties, especially anger. Subsequent MANOVAs show that anger intensity varies depending on whether the youth was a direct victim or a witness only, as well as on the diversity of the types of violent manifestations and on acquaintance with the perpetrator, whereas the presence of injuries has no significant effect. This study highlights the importance of considering the context of the community violence incident, to clearly understand its relationships with the youth’s psychological difficulties. . . . . . Keywords Context of violence Populational survey Trauma Anger Depression Physical violence Mental health problems affect 10–20% of children and ado- experienced physical community violence differ from those lescents worldwide (Kieling et al. 2011). To illustrate, a survey who have not on various indicators of psychological health, carried out in Quebec, Canada (Institut de la statistique du and which characteristics of the violent incident can account Québec - ISQ 2013) reveals that one out of five adolescents for various symptoms. presents a high level of psychological distress. Such distress is According to the World Health Organization’s concep- associated with the level of adversity that characterizes tual framework, community violence is a form of inter- youth’s environment, notably being in contact with violent personal violence perpetrated by strangers or acquain- or unsupportive others (ISQ 2013). In adolescence, young tances other than family members or intimate partners people gradually access a greater variety of environments, (Krug et al. 2002). Available data suggest that community which exposes them to new forms of adversity such as com- violence is a widespread social issue. According to a 2011 munity violence (Matjasko et al. 2013). Focusing on adoles- US victimization survey (Finkelhor et al. 2013), 21.1% of cents from the general population in a high-income country 10–13-year-old and 36.4% of 14–17-year-old adolescents (Canada), the present study evaluates whether youth who have witnessed an assault in their community during last year. Another study performed with a large sample of 9–18- year-old youths from seven European countries revealed that 76% of youth diagnosed with conduct problems re- * Marie-Hélène Gagné firstname.lastname@example.org ported at least one incident of community violence during last year, as compared to 34% of those without such di- agnosis (52% overall) (Kersten et al. 2016). In Québec, School of Psychology, Université Laval, 2325 rue des Bibliothèques, Pavillon Félix-Antoine-Savard 11ème étage, Québec, Québec G1V Canada, a survey representative of 12–17-year-old adoles- 0A6, Canada cents showed that 35.2% of participants reported Present address: Centre Hospitalier de l’Université Laval, Québec experiencing at least one form of community violence in City, Canada the past year, as a victim or a witness (Dubé et al. 2014). Psychology and Psychoeducation Department, Université du Québec A literature review suggested that being a victim or en Outaouais, Saint-Jérôme, Canada witness of community violence is a predictor of psycho- School of Social Work, Université de Montréal, Montréal, Canada logical distress in urban youth, especially depressive 412 Journ Child Adol Trauma (2018) 11:411–420 symptoms, anxiety, post-traumatic stress (PTS), and ag- than only a witness is associated with greater symptomatology gression (McDonald and Richmond 2008). Enlarging the (Lynch 2003), while others found no significant difference in focus to all youths aged 25 or less, Fowler et al. (2009) reported symptoms between victims and witnesses (Kennedy summarized the empirical literature available up to July and Ceballo 2014). According to Zona and Milan (2011), 2005 in a meta-analysis of 114 published and unpublished being the victim of a violent event predicts aggressive and studies linking community violence to three categories of depressive symptoms, whereas being a victim or a witness is psychological symptoms: externalized behaviors, internal- a predictor of PTS symptoms. This finding reflects the Fowler ized behaviors, and PTS. Overall, their findings confirm et al. (2009) meta-analysis: even though direct victimization strong associations between community violence and PTS most predicted symptomatology compared to witnessing symptoms (d = .786) as well as externalized behaviors – community violence, PTS symptoms were equally predicted especially aggressive behavior (d = .630), and moderate by victimization or witnessing. association with internalized behaviors (d = .454). These findings are supported by recent research in various coun- Perpetrator’sIdentity Another variable that may explain the tries (e.g.: Bacchini et al. 2011; Nöthling et al. 2016; level of psychological difficulties in young victims of com- Yearwood et al. 2017). munity violence is the relationship between the victim and the To understand more deeply the relationship between com- perpetrator; that is, whether the victim is acquainted or not munity violence and psychological problems in adolescents, it with the perpetrator. Interpersonal traumas are more strongly is important to consider moderating variables. McDonald and correlated with psychological distress and mental health prob- Richmond (2008) reviewed the following moderators: sex of lems than non-interpersonal traumas (Price et al. 2013). When the victim, family relationship characteristics, school connect- the perpetrator is an acquaintance, the effects are more delete- edness, parental mental health, ethnicity, and grade level. rious than when the perpetrator is a stranger, given the attach- Overall, findings are inconsistent regarding the moderating role ment issues and the relationship of trust initially established of these variables. In addition to adolescent’s sex, the present (Kennedy and Ceballo 2014; Lynch 2003). Therefore, being study focuses on the characteristics of community violence in- the victim of community violence by someone close has been cidents as potential moderators of the relationship between vic- associated with more depressive and PTS symptoms as well as timization on one hand, and symptomatology on the other hand more aggressive behaviors than being victimized by a stranger (angry, depressive, and PTS symptoms). These characteristics (Lambert et al. 2012;Kennedy andCeballo 2014). are: victimization repetition, proximity (victim vs. witness), perpetrator’s identity, and severity (presence of injury). Injuries Violence involving injuries, especially those severe enough to damage the victim’s physical integrity, has been Sex Generally, boys report more community violence than associated with more PTS symptoms. A violent event involv- girls (Rose and Rudolph 2006), except for sexual violence ing bodily harm increases the risk of developing PTS by more which is more commonly reported by girls (Tolin and Foa than eight times (Kennedy and Ceballo 2014; Martin et al. 2006). However, the sex of the victim does not appear to alter 2006). However, it seems that no study has examined how the relationship between community violence and psycholog- the presence of injuries resulting from community violence ical symptoms (McDonald and Richmond 2008). This hy- is related to depressive or aggressive symptoms in the victims. pothesis will be tested in the present study. This will be explored in the present study. This body of empirical literature has some limitations. Multiple Victimization Polyvictimization research has shown Several studies use samples of high-risk children or adoles- that any experience of victimization is a predictor for further cents, often boys only, living in disadvantaged urban neigh- similar experiences (Saunders 2003;Turneretal. 2010). borhoods, of low socio-economic status or consisting solely of Youth who are victimized a first time are two to seven times African Americans, in the United States (Scarpa 2003). These more likely to be revictimized within a year in a similar or populations are typically affected by several other risk factors, different context, compared with those who are not victimized such as poverty, inadequate health care or overpopulation (Finkelhor et al. 2007). In certain contexts, young victims and (Lynch 2003). These confounding variables could thus ex- witnesses tend to experience violent events regularly, even plain part of the psychological problems linked to community daily (Foster et al. 2004), and being a frequent victim or wit- violence. This study overcomes these gaps by examining ness of community violence has been associated with more community violence among both male and female adoles- psychological problems (Scarpa et al. 2006). cents, from the general population. It uses a large adolescent sample recruited in the context of a population survey (Cyr et Victim Versus Witness As it relates to being a victim or a al. 2013). It focuses on physical assaults, since 31.0% of re- witness, research results prove contradictory. According to spondents reported being victim and/or witness of such as- some authors, being a direct victim of a violent event rather saults in their community. In comparison, 4.8% reported Journ Child Adol Trauma (2018) 11:411–420 413 living in a climate of violence (e.g., hate crimes, intimidation, postsecondary studies, compared with 60% in Québec’sgen- hearing about violence in the community…), and 5.1% report- eral population (Cyr et al. 2013). ed community sexual violence (from exhibitionism to rape) The response rate was 37.6%, which is acceptable accord- (Dubé et al. 2014). ing to current survey standards (Babbie 2007). Indeed, studies tend to show that low response rates in telephone surveys do Objectives and Hypotheses not seem to influence the representativeness of the data. Studies having attempted to maximize their response rates The purpose of this study is to document the relationships obtained only a small effect on data representativeness between community violence exposure and three indicators (Holbrook et al. 2007). of psychological health in youth. It is expected that (1) youth reporting a physical community violence experience in the Measures – Victimization Variables year preceding the survey will present more anger, PTS symp- toms, and depressive symptoms; and (2) there will be no in- Juvenile Victimization Questionnaire The Juvenile teraction effect between sex and violence exposure to explain Victimization Questionnaire (JVQ – Hamby et al. 2005)was psychological symptoms. the source for all the victimization variables, as it measures 34 Among youth reporting at least one experience of physical different manifestations of violence (sustained or witnessed), community violence during the year, the study aims to evalu- grouped into five categories: conventional crimes, child mal- ate whether multiple victimization experiences, as well as be- treatment, peer and sibling victimization, sexual violence, and ing a victim rather than only a witness, leads to differences in witnessing / indirect victimization. For each of these manifes- youth’s psychological health. On this subject, it is expected tations, the JVQ documents lifetime and last-year frequency, that (3) youth reporting various manifestations of physical presence and nature of physical injuries, as well as the number community violence will report more anger, PTS symptoms of perpetrators, their sex, and their identity. and depressive symptoms than youth who report only one manifestation; and (4) youth who were victims of physical Experience of Physical Community Violence The community assault in their community will report more anger than those violence measure derived from the JVC by Dubé et al. (2014) who were only witnesses of such assaults. is used in this study to measure community violence exposure Finally, among young victims, the study aims to document during the last year. These authors performed an exploratory whether being acquainted with the perpetrator and sustaining factorial analysis on 27 JVQ violence manifestations, perpe- injuries lead to differences in the psychological health of trated by peers or adults other than family members or inti- youth having suffered physical community violence. In this mate partners, that matched WHO’s definition of community regard, it is expected that (5) youth who were acquainted with violence. The resulting measure includes a scale of physical their perpetrator will present more anger, PTS symptoms, and community violence, comprising the six following manifesta- depressive symptoms than those assaulted by a stranger; and tions: victim of simple assault, witness of simple assault, vic- (6) youth who were physically injured will present more PTS tim of armed assault, witness of armed assault, assault by a symptoms than those who were not. group, and attempted assault. For the purposes of this study, a youth reporting at least one occurrence on at least one of these items, involving a non-related perpetrator (peer or adult) was Method considered to have experienced physical community violence and was coded (1); others were coded (0). Sample In order to account for some characteristics of physical community violence experience, the following variables were This study analyses data collected during a survey on created from JVC answers: (1) diversity of manifestations of polyvictimization of Québec adolescents between 12 and physical community violence (one vs. more than one); (2) 17 years of age (Cyr et al. 2013), based on the methodology having been a victim or only a witness of physical community developed by Finkelhor et al. (2005). A telephone survey on violence; (3) having been the victim of an acquaintance vs the theme of polyvictimization was conducted in 2009 among strangers only; and (4) having sustained (or not) injuries relat- 1400 Québec adolescents (49.7% boys), in French and in ed to this violence. English. More than 85% of them were Caucasian. The base for this survey was a list of phone numbers created using a Diversity of the Manifestations of Physical Community random generation technique (Kish 1965). Despite the sample Violence This variable was calculated for youth reporting at being random, the respondents came from families that were least one experience of physical community violence. From slightly more educated than the general population (Nobert the six manifestations of physical violence measured, one di- chotomous variable was created based on whether the youth 2009). In this study, 72% of the youth’s parents had completed 414 Journ Child Adol Trauma (2018) 11:411–420 reported only one type of manifestation (0) or more than one (Jouvin 2001). Each item is measured according to a four- type of manifestation (1). point frequency scale (Bnever^ to Balmost always^). A score for each scale is calculated by adding up the answers to each Witness or Victim This variable was also calculated for youth item. In this sample, the internal consistency is α = .63 (de- having experienced physical community violence. Among the pressive symptoms), α = .73 (anger) and α = .78 (PTS). six manifestations measured, some refer to direct victimiza- tion, while others describe indirect victimization (being only a Procedure witness to an incident of violence). A dichotomous variable was created to distinguish between these two types of experi- Data were collected through computer-assisted telephone in- ences. Youth reporting at least one manifestation of direct terviews, conducted by a firm specializing in large-scale social victimization are coded (1), whether or not they were also a surveys. When used to address sensitive subjects, telephone witness of violence. Those who reported being only a witness surveys generate results comparable or superior to in-person of violence are coded (0). interviews. Because this type of survey preserves the anonym- ity of the respondents, they are more comfortable and tend to Victim of Stranger Vs. Acquaintance This variable was calcu- respond more conscientiously (Reddy et al. 2006). lated for youth who were direct victims of violence. It was When making the calls, the interviewer would first verify created from a JVQ sub-question that documents perpetrator’s whether the household included at least one 12- to 17-year-old identity. For all manifestations of physical community vio- youth who was inclined to participate in the study. Next, lence reported, if the perpetrator was an acquaintance in at sociodemographic information on the family was obtained. least one case, the youth is coded (1). Youths who report being Finally, the interviewer would make sure to receive verbal victims of strangers only are coded (0). consent from youth aged 14 years or over, or from the parents of children under 14 years old, as applicable. Ethical review Presence of Injuries This variable was also calculated for boards of the researchers’ universities approved the survey. youth who were direct victims. It was created based on a sub-question from the JVQ that asks whether any injuries, ranging from bruises to loss of consciousness, resulted from Analysis Strategy the violent event. As part of this study, youths who reported never being injured for any of the manifestations of physical First, descriptive analyses were performed on the victimiza- community violence were coded (0). Youths who reported tion variables (n and %) and on the psychological variables (M being injured at least once for at least one manifestation of and SD). Next, three bivariate correlation matrices (r)were physical violence were coded (1), regardless of the nature of used to document the relationships between the independent the injuries reported. variables (victimization variables and youth’s sex) and the dependent variables (psychological variables). Measures – Psychological Variables To meet the study objectives and to verify the hypotheses, a series of 2 × 2 factorial multivariate analyses of variance Trauma Symptom Checklist for Children The Trauma (MANOVAs) was performed. This type of analysis helps to Symptom Checklist for Children (TSCC) developed by maximize the probability of detecting significant differences Briere et al. (1995) is a questionnaire intended for 8- to 16- between groups by considering the variance shared by the year-old boys and girls who have been victims or witnesses of dependent variables and of identifying the interaction effects traumatic events. It measures various feelings, thoughts and between the independent variables. behaviors experienced or exhibited in the previous month. For each significant MANOVA, an a posteriori stepdown The French translation of the TSCC (Wright and Sabourin analysis was performed using the Roy-Bargmann method 1996) has been validated by Jouvin (2001) in normative and (Finch 2007). This procedure consists in adding the differ- clinical samples. Three of the six scales from the instrument ent psychological variables, one by one, in order of their are used in this study. The Depression scale (9 items, α =.84) theoretical importance, and thenremovingeachvariable measures feelings of sadness and loneliness, accompanied by previously entered in the model while converting it to a episodes of crying, feeling guilty, pessimism, or suicidal covariable. When the dependent variables are correlated, thoughts. The Anger scale (9 items, α = .81) measures angry this procedure can be used to detect those that most differ- or hateful thoughts, the desire to scream, to fight, and to insult entiate the groups under study. In the current case, and based others, as well as nervousness. The PTS scale (10 items, on the literature review, anger was entered first, followed by α = .84) measures the youth’s preoccupation with recent or PTS symptoms (anger withdrawn and then placed as a past traumatic events, accompanied by overwhelming covariable), and then depressive symptoms (withdrawal of thoughts that could lead to irritability, distraction or tension PTS symptoms, which are placed as a covariable). Journ Child Adol Trauma (2018) 11:411–420 415 Results A third 2 (diversity of the manifestations) × 2 (acquainted perpetrator vs. stranger) factorial MANOVA Of the 1400 youth in the sample, 435 (31.1%) experienced was performed for the 217 youth who reported being di- physical community violence in the year preceding the survey. rect victims of physical community violence, always with Among them, 217 (15.5%) reported at least one experience of the same three dependent variables. The results are report- direct victimization. Youth in the entire sample (N = 1400) ed in Table 3. They show a small effect of the diversity of present mean scores of 5.07 (SD = 3.43) for anger, 6.10 manifestations of physical community violence, F (3, (SD = 3.78) for PTS symptoms, and 4.29 (SD =2.64) for de- 207) = 2.68, η2 = .04, characterized by a significant diver- pressive symptoms. Among adolescents reporting at least one sity X perpetrator’s identity interaction effect, F (3, experience of physical community violence (n = 435), mean 207) = 3.46, η2 = .05. The stepdown analysis shows that scores were 5.68 for anger, 6.67 for PTS symptoms, and 4.51 this interaction effect, illustrated in Fig. 1, is significant for depressive symptoms. Among the direct victims (n =217), for anger only. Consequently, when the perpetrator is a mean scores were 6.45 for anger, 7.11 for PTS symptoms, and stranger, anger is significantly higher in youth having suf- 4.69 for depressive symptoms. fered diverse manifestations of community violence. This Correlations were calculated for the different variables trend is reversed when the perpetrator is an acquaintance; of interest based on the sub-samples identified above, however, in this case, the difference in the mean anger namely all the youth in the sample (N = 1400), victims score is not significant. and witnesses of physical community violence (n = 435), A last 2 (diversity of the manifestations) × 2 (presence or and direct victims (n =217). Overall, correlations are absence of injuries) factorial MANOVAwas performed for the weak, even though most of them are statistically signifi- 217 young victims, always with the same dependent variables. cant. Moderate to high correlations (.45 to .67) were The results are non-significant and are not presented in tables. found between the three psychological variables. Discussion Main Analyses This study aimed first at assessing the differences between A first 2 (experience of community violence) × 2 (sex) youth having experienced physical community violence, com- factorial MANOVA was performed for the entire sample pared with those not having had such an experience, on the (N = 1400) using the anger, PTS, and depression scores as following psychological variables: anger, PTS symptoms, and dependent variables. The results, which are reported in depressive symptoms. The results partially confirm the hy- Table 1, show a small and significant main effect of com- pothesis that youth having experienced violence would pres- munity violence experiences, F (3, 1396) = 7.36, η2=.02. ent more psychological problems, since this finding was ob- The stepdown analysis performed according to the Roy- served only for anger. It is important to emphasize that the Bargmann method shows that this effect is significant for variance shared between anger, PTS, and depression in our anger only. Therefore, youth having experienced commu- sample contributed to a significant result being observed for nity violence report significantly more anger than those anger only, given the analysis strategy selected. In this con- not reporting such an experience, regardless of their sex. text, the measure of anger could be interpreted as an indicator Given the absence of a sex effect and the weak or nil of a more general psychological ill-being in the adolescents correlations between sex and the other variables under from the general population. study, sex was not considered in subsequent analyses. This finding is consistent with those found in the liter- A second 2 (diversity of the manifestations) × 2 (witness ature, underscoring a strong association between commu- vs. victim) factorial MANOVA was performed for the 435 nity violence and aggressiveness (Fowler et al. 2009). For adolescents reporting an experience of physical community instance, being a victim or witness of community violence violence, with the same three dependent variables. The results, is strongly associated with the use of aggressive behaviors which are reported in Table 2, show a small and significant by youth (Scarpa 2001). An explanation is that violence main effect of being a direct victim of physical violence, F (3, exposure contributes to the development of hostile attri- 482) = 4.04, η2 = .03. The stepdown analysis shows that this butions towards the actions of others and to the normali- effect is significant for anger only. Direct victims thus report zation of aggressive behaviors (Janosz et al. 2008). significantly more anger than witnesses only, whether they Furthermore, youth exposed to violence have limited op- report only one or several types of manifestations of commu- portunities to interact appropriately with their peers and to nity violence. develop positive socializing experiences. Thus, they learn to adopt the same violent behaviors as those that they The correlation matrices are not presented here for the sake of brevity. However, they are available from the corresponding author. themselves have suffered or witnessed (Ostrov 2010). 416 Journ Child Adol Trauma (2018) 11:411–420 Table 1 Differences between youth reporting an experience of physical community violence and thosennot reporting any: MANOVA and stepdown analyses (N = 1400) Experience of physical Sex Experience of physical community violence * Sex community violence Absence Presence Male Female Absence * Presence * Absence * Presence * (n =965) (n =435) (n =696) (n =704) Male Male Female Female (n =430) (n =266) (n =535) (n =169) Variables M S.D. M S.D. M S.D. M S.D. M S.D. M S.D. M S.D. M S.D. Anger 4.77 3.11 5.62 3.57 5.19 3.40 4.95 3.46 4.79 3.17 5.79 3.62 4.77 3.43 5.39 3.49 PTSD symptoms 5.79 3.56 6.61 4.10 6.30 3.74 5.86 3.80 6.08 3.48 6.62 4.08 5.54 3.60 6.60 4.13 Depressive symptoms 4.18 2.57 4.50 2.77 4.39 2.70 4.20 2.58 4.31 2.63 4.49 2.81 4.07 2.51 4.51 2.71 F (3, 1396) (η ) 7.36**(.02) .66 (.00) 1.98 (.00) Stepdown analyses Anger 17.86** (.01) 1.19 (.00) 1.01 (.00) F (3, 1396) (η ) PTSD symptoms 2.99 (.00) .72 (.00) 3.67 (.00) F (4, 488) (η ) Depressive symptoms F (5, 487) (η )1.20(.00) .07 (.00) .25 (.00) M mean, S.D. standard deviation, η partial eta squared ** p < .01 * p <.05 Despite a small effect size, the current study shows that it Findings also support the hypothesis that sex does not in- is possible to detect an association between exposure to teract with exposure to community violence to explain differ- physical community violence and psychological outcomes ences in adolescents’ psychological symptoms. First, there in adolescents from the general population. Since various was no main effect of sex on the level of anger, PTS symp- forms of interpersonal violence remain relatively infre- toms, and depressive symptoms reported by adolescents. quent and mild in the general population, small effect Nonetheless, the current literature maintains that boys are sizes are expected and the capacity of detecting them is more likely than girls to report aggressive and antisocial be- a sound result in itself. haviors (Afifi et al. 2011). In this regard, it is important to keep Table 2 Differences according to the diversity of the manifestations of violence and being a witness or a victim: MANOVA and stepdown analyses (n =435) Diversity of the Witness or victim of Diversity * Witness or victim manifestations of physical violence physical violence Only one More than Witness Victim One * More than One * More than one type one type (n =218) (n =217) Witness one* Victim * Victim (n =313) (n =122) (n =199) Witness (n = 114) (n =103) (n =19) Variables M S.D. M S.D. M S.D. M S.D. M S.D. M S.D. M S.D. M S.D. Anger 5.62 3.60 5.84 3.77 4.92 3.12 6.45 3.97 4.94 3.12 4.74 3.16 6.81 4.06 6.05 3.86 PTSD symptoms 6.47 3.95 7.20 4.53 6.25 3.98 7.11 4.25 6.18 3.85 7.05 5.24 6.99 4.10 7.23 4.42 Depressive symptoms 4.47 2.71 4.61 3.09 4.32 2.56 4.69 3.05 4.32 2.48 4.36 3.37 4.73 3.06 4.65 3.05 F (3, 482) (η ) 1.74 (.01) 4.04** (.03) .17 (.00) Stepdown analyses Anger .95 (.00) 10.41** (.02) .32 (.00) F (3, 482) (η ) PTSD symptoms F (4, 481) (η ) 3.81 (.01) 1.69 (.00) .08 (.00) Depressive symptoms F (5, 480) (η ) .45 (.00) .02 (.00) .13 (.00) M mean, S.D. standard deviation, + plus, η partial eta squared. * p < .05 ** p <.01 Journ Child Adol Trauma (2018) 11:411–420 417 Table 3 Differences according to the diversity of the manifestations of violence and the perpetrator’s identity: MANOVA and stepdown analyses (n = 217) Diversity of the Perpetrator’s identity Diversity * Identity manifestations of physical violence Only one More than Stranger Acquaintance One * More than One * More than one type one type (n =24) (n =193) Stranger one* Acquaintance * Acquaintance (n = 114) (n =103) (n =15) Stranger (n =99) (n =94) (n =9) Variables M S.D. M S.D. M S.D. M S.D. M S.D. M S.D. M S.D. M S.D. Anger 6.81 4.06 6.05 3.86 7.25 4.81 6.35 3.86 6.33 3.85 8.78 6.04 6.88 4.11 5.79 3.52 PTSD symptoms 6.99 4.10 7.23 4.42 7.08 4.63 7.11 4.25 5.67 2.85 9.44 6.13 7.19 4.24 7.02 4.20 Depressive symptoms 4.73 3.06 4.65 3.05 5.21 2.90 4.63 3.07 5.27 2.28 5.11 3.89 4.65 3.16 4.61 2.98 F (3, 207) (η ) 2.68* (.04) .82 (.01) 3.46* (.05) Stepdown analyses Anger .60 (.00) 1.94 (.01) 4.06* (.02) F (3, 207) (η ) PTSD symptoms F (4, 206) (η ) 3.36 (.02) .27 (.00) 1.10 (.01) Depressive symptoms F (5, 205) (η ) 4.01* (.02) .27 (.00) 5.10* (.02) M mean, S.D. standard deviation, + plus, η partial eta squared. * p < .05; ** p <.01 The effects observed for depressive symptoms are not interpreted. Given the small number of subjects in certain groups and the absence of significant results for these symptoms in the preceding analyses, proceeding this way was deemed more conservative in mind that the measure used in the present study assesses the associated with physical community violence. Even though emotional and cognitive experience of anger (feeling furious, adolescent girls are generally less exposed to physical com- feeling cross, wanting to scream) rather than aggressive be- munity violence than their male counterparts (Dubé et al. havior per se. It is possible that boys and girls feel anger in a 2014;Rose and Rudolph 2006), those who are exposed report similar manner but that they express it differently. Second, no the same level of difficulties than boys. interaction effect between sex and physical community vio- Next, this study aimed to evaluate whether the diversity lence experience contributed to explain differences in the level of the reported manifestations of physical community vio- of psychological outcomes. This is consistent with the lence, as well as being a direct victim or a witness, resulted McDonald and Richmond’s review: being a boy or a girl in differences in the three psychological variables of interest makes no difference in the level of psychological problems among the 435 youth who reported experiencing at least one Fig. 1 Interaction effect between diversity of violence and perpetrator’s identity on anger (n =217) 418 Journ Child Adol Trauma (2018) 11:411–420 manifestation of physical community violence. The results Finally, the presence of injuries does not help explain the support the hypothesis that direct victims present slightly psychological difficulties of the young victims in this study. more anger than witnesses. One possible explanation is The literature indicates that being physically injured during a based on the theoretical conceptualization of interpersonal violent event influences the psychological symptoms report- violence among adolescents proposed by Crowther et al. ed, especially of PTS (Kennedy and Ceballo 2014;Ozer et al. (2013). These authors underscore the aspect of identity that 2008). Perceiving that one’s life is in danger and sustaining is predominant during this period in life and that could bring several physical injuries increase the risk of presenting such youth who are victims of violence to resort to violence symptoms (Martin et al. 2006). The absence of significant themselves as an adaptation strategy. This strategy would results in this study is probably due to the very small number promote their identity as a strong, brave and threatening of youth reporting this type of experience. Among the 77 person in the eyes of others, allowing them to counter the youth who reported being physically injured during an inci- violence that they have suffered and to prevent future vio- dent of community violence, only four reported serious inju- lence. However, it seems that this strategy does not help ries, such as broken bones or concussions. Others reported reduce the risks of sustaining violence, but rather increases mostly superficial injuries, such as bruises or cuts. it (Stewart et al. 2006). Furthermore, the use of such behav- Sustaining superficial injuries may not be sufficient to develop iors may lead to a vicious circle where the more youth resort PTS symptoms. to violence, the more they feel angry and the more this anger leads to the use of violence (Ostrov 2010;Scarpaand Ollendick 2003). Because witnesses do not have to defend Strengths and Limitations of the Study their own identity, since they are not directly involved in the violent event, they may be less inclined to use violent be- This study has strengths as well as limitations. The use of a haviors or to report feelings of anger associated with the large, randomly recruited populational sample favors the gen- events in question. eralization of the results and helps further understand all as- Meanwhile, significant results were found for diversity of pects of the adolescents’ community violence experience. the reported manifestations of physical community violence, Since the adolescents were interviewed directly, their opinion but only in the group of direct victims of violence, and in on this reality is presented in this study. Considering the dis- interaction with being acquainted with the perpetrator. tinction between being a direct victim of community violence Findings show that youth victimized by a stranger express and being only a witness helps bring out the subtleties of the significantly more anger than those victimized by an acquain- youth’s experience of violence. Additionally, taking several tance, but only in cases where they reported sustaining a di- variables into account to describe the youth’s experience of versity of manifestations of violence (interaction effect). physical community violence more specifically, such as being However, this result must be interpreted with caution, because acquainted or not with the perpetrator and the presence or only a small number of respondents (n = 24) reported being absence of injuries, helps to understand which parameters victims of strangers only, and only nine had suffered several make community violence harmful for adolescents in the gen- manifestations of violence on the part of strangers. eral population. In sum, the highest level of anger is reported by the victims Speaking to the limitations of the study, the use of only of several manifestations of physical violence perpetrated by three psychological health indicators precludes an in-depth strangers, whereas the lowest level of anger is observed in assessment of the psychological state of the youth in the sam- victims of several manifestations of physical violence by ac- ple as it relates to their community violence experience. Other quaintances. Considering that the perpetrator who is an ac- symptoms normally associated with community violence, quaintance is generally a youth under 18 years old, such as a such as anxiety, aggressive behaviors and low self-esteem, friend, a peer, or a neighbor (Dubé et al. 2014), this special would have helped provide a more complete picture. The relationship may play a role in the expression of anger. Friends use of a measure of community violence derived from an take up a lot of room in young people’s lives during adoles- instrument not designed for this purpose (the JVC) also con- cence, a period where creating numerous significant relation- stitutes a limitation to this study. Finally, the use of a cross- ships with peers and being accepted by them is a major con- sectional design precludes the conclusion of causal relation- cern (Waldrip et al. 2008). It is thus possible that youth who ship between community violence and psychological health. have been victimized several times by acquaintances will pre- fer to remain impassive and to not retaliate against the violent acts experienced, to avoid rejection by their peers. The victims Conclusion could come to deny the feelings experienced or to avoid think- ing about them, to preserve the relationship with their peers Given current concerns regarding youth’s mental health, the and their sense of belonging. main contribution of this study is to show that experiences of Journ Child Adol Trauma (2018) 11:411–420 419 Crowther, S., Goodson, C., McGuire, J., & Dickson, J. M. (2013). Having physical community violence could contribute to disturbing to fight. Journal of Interpersonal Violence, 28(1), 62–79. girls and boys alike, notably by feeding feelings of anger. Cyr, K., Chamberland, C., Clément, M.-È., Wemmers, J.-A., Lessard, G., Other studies are needed to establish the direction of this re- Collin-Vézina,D., Gagné,M.-H.,& Damant,D.(2013). lationship. Even if the effects observed are small, the fact that Polyvictimization and victimization of children and youth: Results from a populational survey. Child Abuse and Neglect, 37,814–820. they were detected in the general population is not benign and Dubé, C., Gagné, M.-H., Clément, M.-È., Chamberland, C., & Cyr, K. makes these results original, since most previous studies have (2014). La violence communautaire : portrait des jeunes Québécois. been conducted with high-risk samples. The study confirms [Community violence: A portrait of young Quebecers]. that being a direct victim is associated with more anger than Criminologie, 47(1), 127–148. Finch, W. H. (2007). Performance of the Roy-Bargmann stepdown pro- being exposed as a witness. It also suggests that being victim- cedure as a follow up to significant MANOVA. Multiple Linear ized by an acquaintance (most of the time a peer) may have a Regression Viewpoints, 33(1), 12–22. suppressor effect on this anger, since the youth may want to Finkelhor, D., Hamby, S. L., Ormrod, R., & Turner, H. (2005). The preserve the relationship. However, given the limitations of juvenile victimization questionnaire: Reliability, validity, and na- tional norms. Child Abuse & Neglect, 29(4), 383–412. https://doi. the sample, this last result must be interpreted with caution org/10.1016/j.chiabu.2004.11.001. andwillneedtobereplicated. Finkelhor, D., Ormrod, R. K., & Turner, H. A. (2007). Re-victimization patterns in a national longitudinal sample of children and youth. Acknowledgments Authors would like to thank Nadine Forget-Dubois Child Abuse & Neglect, 31,479–502. https://doi.org/10.1016/j. for statistical support. chiabu.2006.03.012. The study was supported by a grant from the Social Sciences and Finkelhor, D., Turner, H. A., Shattuck, A., & Hamby, S. L. (2013). Humanities Research Council (SSHRC) of Canada (#410-2008-2590) Violence, crime, and abuse exposure in a national sample of children to Marie-Ève Clément, Principal Investigator; and a scholarship from and youth. An update. JAMA Pediatrics, 167,614–621. https://doi. the Social Sciences and Humanities Research Council (SSHRC) of org/10.1001/jamapediatrics.2013.42. Canada to Christine Dubé. Foster, J. D., Kuperminc, G. P., & Price, A. W. (2004). Gender differences in posttraumatic stress and related symptoms among inner-city mi- nority youth exposed to community violence. Journal of Youth and Compliance with Ethical Standards Adolescence, 33(1), 59–69. Fowler, P. J., Tompsett, C. J., Braciszewski, J. M., Jacques-Tiura, A. J., & Conflict of Interest On behalf of all authors, the corresponding author Baltes, B. B. (2009). Community violence: A meta-analysis on the states that there is no conflict of interest. effect of exposure and mental health outcomes of children and ado- lescents. Development and Psychopathology, 21(1), 227–259. Ethical Standards and Informed Consent All procedures followed were https://doi.org/10.1017/S0954579409000145. in accordance with the ethical standards of the Canadian Tri-Council Hamby, S. L., Finkelhor, D., Ormrod, R., & Turner, H. (2005). The Policy Statement: Ethical Conduct for Research Involving Humans, and Juvenile Victimization Questionnaire. Retrieved from http://www. with the Helsinki Declaration of 1975, as revised in 2000. Informed unh.edu/ccrc/pdf/jvq/JVQModules.pdf consent was obtained from all participants for being included in the study. Holbrook, A., Krosnick, J. A., & Pfent, A. (2007). The causes and con- se quences of response rates in surveys by the news media and gov- Open Access This article is distributed under the terms of the Creative ernment contractor survey research firms. In J. M. Lepkowski, C. Commons Attribution 4.0 International License (http:// Tucker, J. M. Brick, E. de Leeuw, L. Japec, P. J. Lavrakas, et al. creativecommons.org/licenses/by/4.0/), which permits unrestricted use, (Eds.), Advances in telephone survey methodology (pp. 499–590). distribution, and reproduction in any medium, provided you give appro- Hoboken: Wiley. priate credit to the original author(s) and the source, provide a link to the Institut de la statistique du Québec, (2013). L’Enquête québécoise sur la Creative Commons license, and indicate if changes were made. santé des jeunes du secondaire 2010–2011. Le visage des jeunes d’aujourd’hui: leur santé mentale et leur adaptation sociale, Tome 2 [The 2010–2011 Quebec survey on the health of high school youth. The face of today’s youth: their Mental Health and their social adjustment, Book 2] (ISBN: 978-2-550-67805-2). Retrieved from References http://www.stat.gouv.qc.ca/statistiques/sante/enfants-ados/ adaptation-sociale/sante-jeunes-secondaire2.pdf Afifi, T. O., McMillan, K. A., Asmundson, G. J. G., Pietrzak, R. H., & Janosz, M., Archambault, I., Pagani, L. S., Pascal, S., Morin, A. J. S., & Sareen, J. (2011). An examination of the relation between conduct Bowen, F. (2008). Are there detrimental effects of witnessing school disorder, childhood and adulthood traumatic events, and posttrau- violence in early adolescence? The Journal of Adolescent Health: matic stress disorder in a nationally representative sample. Journal Official Publication of the Society for Adolescent Medicine, 43(6), of Psychiatric Research, 45,1564–1572. 600–608. https://doi.org/10.1016/j.jadohealth.2008.04.011. Babbie, E. (2007). The Practice of Social Research, 11th edition. Jouvin, E. (2001). Étude des qualités psychométriques de la traduction Belmont, California: Thomson Wadsworth. française du Trauma Symptom Checklist for Children (TSCC) Bacchini, D., Miranda, M. C., & Affuso, G. (2011). Effects of parental [Study of the psychometric qualities of the French translation of monitoring and exposure to community violence on antisocial the Trauma Symptom Checklist for Children (TSCC)] (Master’s the- hehavior and anxiety/depression among adolescents. Journal of sis). Université de Montréal. Interpersonal Violence, 26,269–292. Kennedy, T. M., & Ceballo, R. (2014). Who, what, when, and where? Briere, J., Elliott, D. M., Harris, K., & Cotman, A. (1995). Trauma symp- Toward a dimensional conceptualization of community violence exposure. Review of General Psychology, 18(2), 69–81. tom inventory: Psychometrics and association with childhood and Kersten, L., Fehlbaum, L., Praetzlich, M., Menks, W., Steppan, M., adult trauma in clinical samples. Journal of Interpersonal Violence, 10,387–401. Mannstadt, S., Dietrich, C., Raschle, N., & Stadler, C. (2016). 420 Journ Child Adol Trauma (2018) 11:411–420 Prevalence of community violence exposure and its effects on men- behavioral development of girls and boys. Psychological Bulletin, 132(1), 98–131. tal health problems in European adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 55- Saunders, B. E. (2003). Understanding children exposed to violence: (supplement), S123. https://doi.org/10.1016/j.jaac.2016.09.074. Toward an integration of overlapping fields. Journal of Kieling, C., Baker-Henningham, H., Belfer, M., Conti, G., Ertem, I., Interpersonal Violence, 18,356–376. Omigbodun, O., Rohde, L. A., Srinat, S., Ulkuer, N., & Rahman, Scarpa, A. (2001). Community violence exposure in a young adult sam- A. (2011). Child and adolescent mental health worldwide: Evidence ple: Lifetime prevalence and socioemotional effects. Journal of for action. The Lancet, 378, 1515–1525. https://doi.org/10.1016/ Interpersonal Violence, 16(1), 36–53. https://doi.org/10.1177/ S0140-6736(11)60827-1. Kish, L. (1965). Survey sampling. New York: Wiley. Scarpa, A. (2003). Community violence exposure in young adults. Krug, E. G., Mercy, J. A., Dahlberg, L. L., & Zwi, A. B. (2002). The Trauma, Violence & Abuse, 4(3), 210–227. world report on violence and health. The Lancet, 360,1083–1088. Scarpa, A., Haden, S. C., & Hurley, J. (2006). Community violence Lambert, S. H., Boyd, R. C., Cammack, N. L., & Ialongo, N. S. (2012). victimization and symptoms of posttraumatic stress disorder: The Relationship proximity to victims of witnessed community violence: moderating effects of coping and social support. Journal of Associations with adolescent internalizing and externalizing behav- Interpersonal Violence, 21,446–469. iors. American Journal of Orthopsychiatry, 82(1), 1–9. Scarpa, A., & Ollendick, T. H. (2003). Community violence exposure in a Lynch, M. (2003). Consequences of children’s exposure to commu- young adult sample: III. Psychophysiology and victimization inter- nity violence. Clinical Child and Family Psychology Review, act to affect risk for aggression. Journal of Community Psychology, 6(4), 265–274. 31,321–338. https://doi.org/10.1002/jcop.10058. Martin, M., Germain, V., & Marchand, A. (2006). Facteurs de risques et Stewart, E. A., Schreck, C. J., & Simons, R. L. (2006). I ain’t gonna de protection dans la modulation de l’état de stress post-traumatique. let no one disrespect me: Does the code of the street reduce or [Risk and protective factors in the modulation of post-traumatic increase violent victimization among African American adoles- stress disorder]. In S. Guay & A. Marchand (Eds.), Les troubles cents? Journal of Research in Crime and Delinquency, 43, liés aux événements traumatiques - Dépistage, évaluation et 427–458. traitements [Disorders related to traumatic events—Detection, eval- Tolin, D. F., & Foa, E. B. (2006). Sex differences in trauma and posttrau- uation and treatments] (pp. 51–86). Montréal: Les presses de matic stress disorder: A quantitative review of 25 years of research. l’Université de Montréal. Psychological Bulletin, 132(6), 959–992. Matjasko, J. L., Vivolo-Kantor, A. M., Henry, D. B., Gorman-Smith, D., Turner, H. A., Finkelhor, D., & Ormrod, R. (2010). The effects of ado- & Schoeny, M. E. (2013). The relationship between a family- lescent victimization on self-concept and depressive symptoms. focused preventive intervention, parenting practices, and exposure Child Maltreatment, 15,76),76–76), 90. to violence during the transition to adolescence: Testing a media- Waldrip, A. M., Malcolm, K. T., & Jensen-Campbell, L. A. (2008). With tional model. Journal of Aggression, Maltreatment & Trauma, 22, a little help from your friends: The importance of high-quality 45–66. https://doi.org/10.1080/10926771.2013.743947. friendships on early adolescent adjustment. Social Development, McDonald, C. C., & Richmond, T. R. (2008). The relationship between 17(4), 832–852. community violence exposure and mental health symptoms in urban adolescents. Journal of Psychiatric and Mental Health Nursing, 15, Wright, J., & Sabourin, S. (1996). Traduction française du trauma symp- 833–849. https://doi.org/10.1111/j.1365-2850.2008.01321.x. tom checklist for children. [French translation of the trauma symp- Nobert, Y. (2009). Éducation, insertion en emploi et formation continue. tom checklist for children]. Unpublished manuscript. Montréal: [Education, integration in employment and continuing education]. Université de Montréal. Donnéessocialesdu Québec. Retrieved from http://www.stat.gouv. Yearwood, K., Vliegen, N., Chau, C., Corveleyn, J., & Luyten, P. (2017). qc.ca/publications/conditions/pdf/extr_donnees_sociales09.pdf Prevalence of exposure to complex trauma and community violence Nöthling, J., Suliman, S., Martin, L., Simmons, C., & Seedat, S. (2016). and their associations with internalizing and externalizing symp- Differences in abuse, neglect, and exposure to community violence toms. Journal of Interpersonal Violence. Published online. https:// in adolescents with and without PTSD and depression. Journal of doi.org/10.1177/0886260517731788. Interpersonal Violence. Published online: https://doi.org/10.1177/ Zona, K., & Milan, S. (2011). Gender differences in the longitudinal impact of exposure to violence on mental health in urban youth. Ostrov, J. M. (2010). Prospective associations between peer victimization Journal of Youth and Adolescence, 40,1674–1690. and aggression. Child Development, 81(6), 1670–1677. https://doi. org/10.1111/j.1467-8624.2010.01501.x. Ozer, E. J., Best, S. R., Lipsey, T. L., & Weiss, D. S. (2008). Predictors of Christine Dubé , Ph.D., is a Clinical Psychologist at Centre Hospitalier posttraumatic stress disorder and symptoms in adults: A meta-anal- de l’Université Laval in Quebec City, Canada. ysis. Psychological Trauma: Theory, Research, Practice, and Policy, 129(1), 3–36. Marie-Hélène Gagné , Ph.D., is a Community Psychology Professor at Price, M., Higa-McMillan, C., Kim, S., & Frueh, C. (2013). Trauma Université Laval in Québec City, Canada, and holds the Partnership Chair experience in children and adolescents: An assessment of the effects for the Prevention of Child Maltreatment. of trauma type and role of interpersonal proximity. Journal of Anxiety Disorders, 27,652–660. Marie-Ève Clément , Ph.D., is a Psychology Professor at Université du Reddy, M. K., Fleming, M. T., Howells, N. L., Rabenhorst, M. M., Québec en Outaouais in Saint-Jérôme, Canada, and holds the Canada Casselman, R., & Rosenbaum, A. (2006). Effects of method on Research Chair on Violence towards Children. participants and disclosure rates in research on sensitive topics. Violence and Victims, 21(4), 499–506. Claire Chamberland , Ph.D., is an Emeritus Social Work Professor at Rose, A. J., & Rudolph, K. D. (2006). A review of sex differences in peer Université de Montréal in Montréal, Canada. relationship processes: Potential trade-offs for the emotional and
Journal of Child & Adolescent Trauma – Springer Journals
Published: Jun 2, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera