Profiles of Community Violence Exposure Among African American Youth

Abstract

The current study employed latent class analysis (LCA) to identify distinct profiles of community violence exposure and their associations to desensitization outcomes in 241 African American early adolescents (M age = 12.86, SD = 1.28) in the sixth through eighth grade from under-resourced urban communities. Participants self-reported on their exposure to community violence, as well as on depressive and anxiety symptoms. The LCA revealed three distinct classes: a class exposed to low levels of violence (low exposure class), a class exposed to moderately high levels of victimization (victimization class), and a class exposed to high levels of all types of violence (high exposure class). Consistent with predictions, the high exposure class showed the lowest levels of depressive symptoms, suggesting a desensitization outcome. Gender and age were also examined in relation to the classes, and age was significantly associated with an increased risk of being a member of the high exposure class relative to the low exposure class. Using person-based analyses to examine desensitization outcomes provides useful information for prevention and intervention efforts, as it helps to identify a specific subgroup of youth that may be more likely to show desensitization outcomes in the context of community violence.

Keywords

community violence mental health and violence violence exposure

Although there have been some successful efforts to curtail violence and crime in urban communities, African American adolescents who reside in these communities remain vulnerable to high levels of exposure to community violence (Busby, Lambert, & Ialongo, 2013; Gibson, Morris, & Beaver, 2009). Furthermore, the pathologic adaptation model (PAM) suggests that youth exposed to high levels of community violence become desensitized by showing lower than expected levels of emotional distress (e.g., Ng-Mak, Salzinger, Feldman, & Stueve, 2004). Indeed, the existing research demonstrating desensitization outcomes for youth exposed to community violence show that community violence predicts lower levels of emotional distress, suggestive of emotional numbing (Ng-Mak et al., 2004). What is unclear is how variability in community violence exposure among African American youth from low-income, urban neighborhoods is related to differences in desensitization outcomes. Person-based statistical approaches can provide insight into patterns of variability (Masten, 1999) that may provide information about desensitization outcomes based on levels of violence exposure. Thus, the current study employed person-centered analytic procedures, namely latent class analysis (LCA), to identify distinct profiles of community violence exposure in African American youth.

Exposure to Violence in African American Youth

As noted in several reviews of the literature, community violence research often groups youth’s exposure to violence into two main categories: being a victim of violence and witnessing violence that happens to others (Fowler, Tompsett, Braciszewski, Jacques-Tiura, & Baltes, 2009; McDonald & Richmond, 2008; Stein, Jaycox, Kataoka, Rhodes, & Vestal, 2003). Victimization is conceptualized as being the object of intentional acts initiated by another person to cause harm, such as being threatened, robbed, physically attacked, shot, or stabbed, whereas witnessing is conceptualized as hearing, seeing, or hearing about an event that involves loss of property, threat of physical injury, actual injury, or death (Fowler et al., 2009). Compared with youth from other ethnic and socioeconomic status groups, African American youth from under-resourced, urban communities are exposed to disproportionately higher levels of community violence (Zimmerman & Messner, 2013). Prior research shows that between 45% and 96% of African American youth have witnessed violence in their community, ranging from assault to murder (Gaylord-Harden, Cunningham, & Zelencik, 2011; Margolin & Gordis, 2000; Self-Brown et al., 2006), and estimates ranging from 16% to 37% report violent victimization (Farrell & Bruce, 1997; Spano & Bolland, 2013). Exposure is often repeated and ongoing, with 75% of African American youth reporting having witnessed four or more violent events during adolescence (Miller, Wasserman, Neugebauer, Gorman-Smith, & Kamboukos, 1999).

The deleterious effects of exposure to violence on numerous emotional and behavioral outcomes in African American youth are well documented (Gorman-Smith, Henry, & Tolan, 2004; Kliewer et al., 2004; Ozer & Weinstein, 2004; Sullivan, Kung, & Farrell, 2004), consistent with current theory and practice asserting that high levels of community violence exposure lead to more emotional distress and more behavioral problems. However, these cumulative stress models do not explain why some African American youth exposed to high levels of community violence show lower than expected levels of emotional distress (Boxer et al., 2008). Specifically, several studies have found surprisingly low associations between community violence and depressive symptoms (Cooley-Quille, Boyd, Frantz, & Walsh, 2001; Farrell & Bruce, 1997; Fitzpatrick, 1993; Moses, 1999). In other words, higher levels of violence exposure do not always predict higher levels of depressive symptoms.

The unexpected findings for depression may reflect a desensitization process, in which youth “adapt” to repeated exposure by normalizing violence and suppressing emotional distress, such as depressive affect or symptoms (Cooley-Quille & Lorion, 1999; Fitzpatrick & Boldizar, 1993; McCart et al., 2007; Mrug, Loosier, & Windle, 2008). The PAM (Ng-Mak et al., 2004) provides a framework for desensitization to community violence, but focuses on emotional desensitization. Emotional desensitization occurs when an individual experiences a numbing or blunting of emotional reactions to violence exposure, which should typically elicit a strong emotional response (Funk, Baldacci, Pasold, & Baumgardner, 2004). For youth exposed to community violence, affective symptoms of sadness, emptiness, or worthlessness may be blunted. This emotional desensitization process appears adaptive in the short-term for decreasing emotional distress, but places youth at elevated risk for violence perpetration and additional violence exposure (Ng-Mak, Salzinger, Feldman, & Stueve, 2002, 2004). Specifically, youth who experience emotional desensitization may begin to view violent and aggressive behavior as normative, lose inhibitions about using violent behavior, and use more dangerous self-protective behaviors (e.g., carrying a weapon; Gorman-Smith & Tolan, 1998; Guerra, Huesmann, & Spindler, 2003; Jenkins & Bell, 1994; Reese, Vera, Thompson, & Reyes, 2001; Spano, Rivera, & Bolland, 2010), leading to other problems. Indeed, research has shown that higher levels of community violence exposure predict higher levels of delinquent behaviors in youth (e.g., Patchin, Huebner, McCluskey, Varano, & Bynum, 2006) and behavioral problems at school (Aisenberg & Mennen, 2000).

While the PAM provides a framework for researchers to begin to examine emotional desensitization to community violence, the research on desensitization to media violence suggests that individuals may also become desensitized in other ways. For example, physiological desensitization occurs when there is a reduction in physiological reactivity (e.g., heart rate, galvanic skin response) to violence (Carnagey, Anderson, & Bushman, 2007; Funk et al., 2004). Physiological desensitization suggests that individuals experience less physiological arousal, indicative of anxiety or fear, as a result of violence exposure. This form of desensitization has been studied frequently in media violence research, and while this research shows that prolonged exposure to violent video games results in physiological desensitization (Carnagey et al., 2007), the literature on physiological desensitization to community violence is less clear. Specifically, in contrast to depressive symptoms, exposure to community violence has shown relatively consistent associations with higher levels of physiological hyperarousal (e.g., somatic tension, panic, and nervousness; Cooley-Quille et al., 2001; Margolin & Gordis, 2000; Singer, Anglin, Song, & Lunghofer, 1995). For youth in communities with elevated crime rates and violence, repeated exposure to community violence may increase hyperarousal or hypervigilance to help youth recognize and respond to threatening situations (LeDoux, 1993; Schell, Marshall, & Jaycox, 2004). Thus, decreases in depressive affect, indicative of emotional desensitization, coupled with increases in anxiety, indicative of physiological hypersensitization, may be a consequential pattern of outcomes for community violence exposure. Prior research with African American youth demonstrated that the association between community violence and depression was curvilinear (suggestive of emotional desensitization), whereas the association to anxiety was linear and positive (suggestive of physiological desensitization; Gaylord-Harden et al., 2011). These patterns of findings suggest that it may be important to examine separately the effects of community violence on depression and anxiety, as desensitization effects may be specific to depression. In addition, there is a need to understand how much variability exists within African American youth from low-income communities with regard to violence exposure and violence-related outcomes, such as depression and anxiety.

The Use of Profile Analyses to Examine Desensitization to Community Violence

Person-based statistical approaches, such as LCA, can provide insight into patterns of variability within data (Mandara, 2003; Masten, 1999) by identifying distinct groups or classes of youth. In person-based analysis, the relevant aspect of the analysis is the profile of scores, and not the impact of an individual variable (Bergman & Magnusson, 1997). Despite the utility of person-based analyses, these methods have not been widely used in exposure to community violence research and, to our knowledge, have not been used in an investigation of emotional desensitization to violence. Two existing studies using person-based analyses with African American adolescents from urban communities identified distinct profiles of participants based on community violence exposure (Copeland-Linder, Lambert, & Ialongo, 2010; Ronzio, Mitchell, & Wang, 2011). However, additional research is needed to build upon prior research, as these studies did not specifically assess desensitization to violence outcomes, and Ronzio et al. (2011) focused only on witnessing violence in African American mothers. Given the heightened risk of future violence perpetration by youth who become desensitized, identification of distinct groups of adolescents based on levels of violence exposure will provide information on the prevalence of desensitization, as well as which adolescents may be particularly vulnerable to emotional desensitization and should be targeted for intervention.

Effects of Gender and Age on Violence Exposure

In general, boys of all ethnicities are exposed to higher rates of community violence than girls (Boyd, Cooley, Lambert, & Ialongo, 2003; Chen, 2009; Weist, Acosta, & Youngstrom, 2001), and one in four African American boys report being victimized (e.g., beaten or shot at), compared with 12% of African American females (Chen, 2009), often more than once during adolescence (Gaylord-Harden et al., 2011). Furthermore, gender socialization theories suggest that girls are more inclined to internalize their problems and boys to externalize their problems (Nolen-Hoeksema, 1994). Indeed, male adolescents report more self-protective (e.g., carrying a weapon) and aggressive behaviors in response to witnessing violence, while female adolescents report more depressive symptoms (Jenkins & Bell, 1994; Reese et al., 2001; Self-Brown et al., 2006). Given that males are exposed to higher levels of community violence than females, but show lower levels of internalizing symptoms in response to violence exposure, emotional desensitization in response to violence may be more likely to occur in males than in females (Ng-Mak et al., 2002).

Although some findings show that older youth are exposed to more violence than younger youth (e.g., Weist et al., 2001), reviews of the literature suggest that the association between age and exposure to violence is equivocal (e.g., Stein et al., 2003). Alternatively, recent meta-analytic findings suggest that age may moderate the effects of violence exposure on outcomes. Specifically, younger children are more likely to experience internalizing symptoms in response to community violence, whereas adolescents are more likely to experience externalizing symptoms (Fowler et al., 2009). Therefore, desensitization to community violence may be more likely to occur as children and adolescents get older. Over time, as youth age, they may be more likely to normalize their experiences of violence and, in turn, experience less emotional distress and more aggressive means of coping with violence exposure (Gorman-Smith & Tolan, 1998; Guerra et al., 2003).

The Current Study

As shown above, LCA can categorize youth based on their exposure to violence into different exposure groups and might provide insight into how different naturally occurring “profiles” of exposure to community violence (ECV) relate to outcomes. The identification of subgroups may provide critical information on subgroups of violence-exposed adolescents to target for prevention and intervention efforts (Copeland-Linder et al., 2010; Fowler et al., 2009). Adolescence is an ideal developmental stage to examine such patterns in violence exposure, as behavioral patterns related to violence and aggression, which may emerge during childhood or early adolescence, reach their peak prevalence during this time (e.g., Tolan, Gorman-Smith, & Henry, 2003). Community violence can be disruptive to optimal development by increasing the likelihood of violence perpetration, risky behaviors, and increased associations with deviant peer groups (Fowler et al., 2009; Ingoldsby & Shaw, 2002). The onset of internalizing symptomatology during adolescence, heightened by community violence exposure, predicts a more chronic course for symptoms, which has long-term implications for social, emotional, and cognitive development (e.g., Woodward & Fergusson, 2001).

Thus, the purpose of the current study is to employ person-centered analytic procedures, namely LCA, to identify distinct profiles of community violence exposure in African American adolescents. Consistent with prior research (Copeland-Linder et al., 2010; Ronzio et al., 2011), we expected the LCA to result in at least two classes of participants, based on low and high levels of ECV. However, we did not have predictions about the specific compositions of other possible classes. Also based on prior research, it was predicted that classes with high exposure to community violence would include significantly more males than females and a significantly greater percentage of older youth than younger youth. It was also predicted that classes that report higher levels of ECV would show fewer symptoms of depression and higher symptoms of anxiety, indicative of emotional desensitization and physiological hypersensitization. Finally, it was predicted that males and older adolescents within high exposure classes would show fewer symptoms of depression (indicative of emotional desensitization) than females and younger adolescents in these classes.

Method

Participants

Participants in the current study were 241 (56% female; M age = 12.86, SD = 1.28) African American early adolescents in the sixth through eighth grade (ages 11-15) from five public schools in urban areas. Participants were part of a larger study designed to assess the effects of stress and coping on internalizing symptoms in African American youth from economically disadvantaged communities. Public schools were selected for participation based on high percentages of low-income students and Black students, as gathered from the school system’s annual reports, which include distributions of students’ socioeconomic status and race. The average percentage of Black students at the five schools was 93%, and the average percentage of low-income students, based on eligibility for free or reduced meals programs, was 90%. The participation rate across the five schools was approximately 50%. Thirty-eight percent of the participants were enrolled in the sixth grade, 26% were in the seventh grade, and 36% were in the eighth grade. Approximately 47% resided with their mother, 42% resided with both their parents, and the remaining 11% resided with other relatives (e.g., grandparent).

Procedure

The current study was granted ethical approval by a university-based institutional review board (IRB). Researchers visited the selected schools and met with faculty and staff to introduce the project and describe the procedures of recruitment and data collection. A recruitment letter and consent form were sent home with all sixth- through eighth-grade students. Active parental consent and youth assent were obtained for all study participants. Youth receiving parental consent and providing youth assent were taken to a room designated by the school staff (e.g., classroom, library) to complete a packet of questionnaires in a group setting. At least two research assistants were available to administer the surveys, answer questions, and monitor the progress of participants. Teachers were not present during data collection. To manage the range in reading levels of the participants, the research assistants read all questions aloud. Each participant received a movie pass for his or her participation in the study.

Measures

Demographic form

Participants reported their age, grade, gender, race/ethnicity, primary caregiver, and number of people living in the home.

Victimization and witnessing community violence

Participants’ experiences of victimization and witnessing violence in the community were assessed using the Multicultural Events Schedule for Adolescents (MESA; Gonzales, Gunnoe, Jackson, & Samaniego, 1995). The MESA is an 82-item self-report measure of stress for adolescents living in an urban environment. Participants rate the occurrence of the stressful events that have happened over the past 3 months by selecting “happened” or “did not happen.” The current study utilized the specified time range, rather than lifetime exposure, consistent with the recommendation that researchers interested in minimizing recall bias for reports of victimization should use specified time ranges (Finkelhor, Ormrod, & Turner, 2009). Two-week test–retest reliability of the MESA was .71 (Gonzales et al., 1995), and for African American youth, it was .81 (Gonzales et al., 1995).

The MESA includes four items that assess victimization in the community. A sample item includes, “You were threatened with a knife or gun.” Higher scores indicate a higher level of victimization. In the current sample, Cronbach’s alpha for the four victimization items was .47. However, given that the current study used an item-level LCA, individual items assessing victimization, rather than an overall subscale score, were used in the analyses.

The MESA includes six items that assess witnessing community violence. A sample item includes, “Someone close to you was shot or attacked.” Higher scores indicate a higher level of witnessing violence. In the current sample, Cronbach’s alpha for the six witnessing violence items was .77. However, given that the current study used an item-level LCA, individual items assessing witnessing, rather than an overall subscale score, were used in the analyses.

Depressive symptoms

Depressive symptomatology was assessed with the 27-item Child Depression Inventory (CDI; Kovacs, 1992). For each item, respondents select one of three sentences that most closely describes him or her over the past 2 weeks (“I am sad once in a while, I am sad many times, I am sad all the time”). The 27 items are combined to provide an overall depression score; higher scores indicate higher levels of depression. Reliability and validity of the CDI are well-established (Kovacs, 1992). Internal consistency estimates for the overall depression score in African American youth have ranged from .81 to .84 (Dempsey, 2002; Fitzpatrick, 1993; Johnson & Kliewer, 1999; McMahon & Watts, 2002). Cronbach’s alpha for the overall depression score in the current sample was .85.

Anxiety symptoms

Anxiety symptoms were assessed with the Revised Children’s Manifest Anxiety Scale (RCMAS; Reynolds & Richmond, 1997). The RCMAS contains 37 sentences to which the respondent is asked to respond “yes” or “no.” Of the 37 items, 28 measure anxiety symptoms (“I worry a lot of the time”), and 9 assess socially desirable responding. The 28 items are combined to provide an overall anxiety score. Higher scores indicate higher levels of anxiety. The reliability and validity of the measure are well-established (Reynolds & Richmond, 1997). Internal consistency estimates for the overall anxiety score in African American youth have ranged from .84 to .86 (Kliewer & Kung, 1998; McMahon & Watts, 2002; White, Bruce, Farrell, & Kliewer, 1998). Cronbach’s alpha for the overall anxiety score in the current sample was .89.

Results

Analysis Procedure

LCA was used to analyze community violence exposure among the participants. Similar to cluster analysis, LCA is a person-centered analytical approach that detects unobserved heterogeneous subpopulations, which are latent classes, in the given sample. LCA differs from cluster analysis as classes are based on response probabilities as opposed to distances from the group mean. In essence, this probability model accommodates unequal degrees of variance for each individual cluster.

One of the complexities of LCA is that there are difficulties comparing LCA models with different numbers of classes because they are not “nested” within one another. To determine the optimal number of latent classes, likelihood ratio tests cannot be used to make comparisons, and therefore the Akaike Information Criterion (AIC; Akaike, 1987), Bayesian Information Criterion (BIC; Schwartz, 1978), Adjusted Bayesian Information Criterion (ABIC; Sclove, 1987), Entropy (Ramaswamy, DeSarbo, Reibstein, & Robinson, 1993), and the Bootstrapped Lo–Mendell–Rubin Likelihood Ratio (BLMR LR; McLachlan & Peel, 2000) tests were used to determine the number of latent classes. In the case of the AIC, BIC, and ABIC, lower observed values indicate better model fit. In addition, entropy is a measure of model fit with values closer to 1.00 suggesting better model fit. Finally, a nonsignificant p value for the BLMR LR test indicates that the model with the (K − 1)-class model is preferred to the model with K classes. Research suggests that the BLMR LR as the best discriminator of classes in LCA models (Nylund, Asparouhov, & Muthen, 2007). Therefore, the final classes were determined by small AIC, BIC, and ABIC values comparing each class (K) with each K − 1 classes as well as nonsignificant BLMR LR values.

Model Selection

Using the previously described information criteria, four LCA models with different number of latent classes were estimated sequentially using Mplus Version 7.1 (Muthén & Muthén, 2013). The three-class model suggested the best fit to the data with the lowest observed AIC, BIC, and ABIC values, highest entropy values, and significant BLMR LR. When the number of latent classes was increased from three to four classes, the p value of the BLMR LR test was not statistically significant (p = .667), indicating that the addition of a fourth class did not significantly improve the fit of the model (see Table 1). Therefore, the three-class model was retained, and the three classes were identified as victimization class (n = 95), low exposure class (n = 56), and high violence exposure class (n = 90). The estimated probabilities for violence exposure items for each class are presented in Figure 1.

Table 1.

Information Criteria, Entropy, and Likelihood Ratio Test Values.

	Fit Indices and Entropy of the Growth Mixture Models
Model	Log Likelihood	AIC	BIC	ABIC	Entropy	BLMR LR
1 class	−1,415.432	2,850.864	2,885.711	2,854.014
2 class	−1,226.095	2,494.190	2,567.370	2,500.805	.811	<.001
3 class	−1,200.041	2,464.081	2,575.595	2,474.162	.736	<.001
4 class	−1,190.827	2,467.655	2,617.501	2,481.200	.756	.667

Note. AIC = Akaike Information Criterion; BIC = Bayesian Information Criterion; ABIC = Adjusted Bayesian Information Criterion; BLMR LR = Bootstrapped Lo–Mendell–Rubin Likelihood Ratio.

Figure 1.

Typologies of violence exposure using conditional probabilities.

The largest class of youth, the victimization class (59% female; M age = 12.85, SD = 1.59), reported moderately high levels of direct victimization, defined as having been the object of intentional acts initiated by another person to cause harm (Fowler et al., 2009), and indirect victimization, defined as knowledge of someone close having been the object of an intentional act initiated by another person to cause harm (Finkelhor, Hamby, Ormrod, & Turner, 2005). In contrast, this class reported low rates of witnessing violence, such as hearing gunshots in the community or seeing someone commit a crime.

The low exposure class (46% female; M age = 12.55, SD = 1.04) reported very low rates of witnessing violence in the community and moderately low rates of victimization. While the rates for victimization were relatively higher than witnessing violence for the low exposure class, the rates of violence exposure in this class were considerably lower than the other classes.

The final class, the high exposure class (57% female; M age = 13.04, SD = 0.97), reported very high rates of witnessing, and both direct and indirect victimization. The percentages for violence exposure items for each class are presented in Table 2.

Table 2.

Conditional Probability for Endorsing Violent Events in Each Class.

	Low Exposure Class (23%)	Victimization Class (41%)	High Exposure Class (36%)
Violence items
Saw someone commit a crime	.21	.42	.93
Heard gunshots	.4	.16	.61
Saw someone with a weapon	.3	.25	.89
Saw someone being threatened	.18	.62	.97
Someone close was threatened with knife or gun	.22	.79	.97
Someone close was shot or attacked	.1	.56	.87
You were physically attacked	.41	.83	.95
Someone broke into your home	.71	.92	.97
You were threatened with a knife or gun	.55	.93	.100
Someone stole something from you	.26	.53	.69

To determine for differences in relative risk among the latent classes as a function of age and gender, multinomial logistic regression analyses were conducted by regressing the unordered categorical latent variable (i.e., class) on both age and gender. Class membership is assigned to each individual using a pseudo-class draw approach in which each individual’s class is determined through random draws from each individual’s posterior probability distribution (Asparouhov & Muthen, 2010; Wang, Brown, & Bandeen-Roche, 2005). In the present case, the class assignment is used as the dependent variable in the multinomial logistic regression analysis. To allow for assessment of change in risk, the regression coefficients were converted to relative risk ratios by exponentiating the linear coefficients computed in the multinomial logistic regression analyses. The results suggest age was significantly associated with an increased risk of being a member of the high exposure class (OR = 1.58, p = .013) relative to the low exposure class. However, there was no increase in risk as a function of age between the victimization and low exposure classes (OR = 1.40, p = .081) or victimization and high exposure classes (OR = 0.88, p = .478). Interestingly, there were no significant differences in relative risk between classes as a function of gender.

To test for differences in depression and anxiety across the three clusters, the equality of means for each outcome was compared across the latent classes. To compare the invariance of the means across the classes, the previously described pseudo-class draws approach was used (Asparouhov & Muthén, 2007; Wang et al., 2005). Once class assignments were determined, tests of mean equality were conducted. In the present analysis, each mean was constrained to be equal, and a significant chi-square indicated significantly different means. The tests for differences in mean scores of depression indicate that depressive symptoms were significantly lower for the high exposure class in comparison with the victimization class, χ²_diff(1) = 4.04, p = .044. Also, the high exposure class demonstrated significantly lower depressive symptoms in comparison with the low exposure class, χ²_diff(1) = 6.74, p = .009. Finally, depressive symptoms were not significantly different for the low exposure class in comparison with the victimization class χ²_diff(1) = 0.38, p = .539. The final class means for depressive symptoms are high exposure (M = 7.85, SE = 0.77), victimization (M = 10.30; SD = 0.85), and low exposure (M = 11.19, SD = 1.04). The overall test for differences in anxiety scores across the three classes did not indicate significant differences in anxiety across classes, χ²_diff(2) = 1.03, p = .598. The final class means for anxiety symptoms were high exposure (M = 37.11, SE = 0.78), high victimization (M = 38.33, SE = 0.99), and low exposure (M = 38.65, SE = 1.52).

Discussion

The results of the current study reveal that there are classes or “profiles” of adolescents that demonstrate distinct patterns of victimization and witnessing community violence. The findings suggested that there were three classes of youth based on violence exposure. Consistent with predictions, there was a high exposure class and a low exposure class; however, results also revealed a victimization class. The high exposure class reported high rates of all types of community violence, including witnessing violent events in the community and victimization. The rates of exposure to violence in this class were higher than the other two classes for all items. The low exposure class reported very low rates of witnessing and low rates of victimization. The largest class of youth, the victimization class, reported high levels of high direct and indirect (vicarious) victimization.

Consistent with our predictions, depression scores were lowest in the high exposure class. This finding lends support for emotional desensitization theories, such as the PAM, which posit that youth exposed to high levels of community violence experience lower than expected levels of depressive symptoms, which may reflect emotional numbing (Ng-Mak et al., 2004). Emotional numbing is regarded as a volitional and acquired coping mechanism in traumatized youth (Kerig, Bennett, Thompson, & Becker, 2012). Youth may begin to employ emotional numbing of affective symptoms as a means of coping with the traumatic nature of witnessing violent acts and being victimized (Kerig et al., 2012). Interestingly, youth exposed to very high levels of both victimization and witnessing were most likely to show an emotional desensitization effect, as evidenced by lower depressive symptoms that the other classes. Conversely, youth exposed almost solely to high rates of victimization do not show the same emotional desensitization effect. Thus, it may be that experiencing a wider range of violent events is most likely to lead to emotional desensitization effects. An alternative explanation may be that there is something unique about witnessing violence that results in emotional desensitization. For example, seeing or hearing about violence that happens to others results in a lower likelihood of personal threat or fear than being directly victimized (e.g., Russo & Roccato, 2010). Consequently, it may be easier to desensitize oneself to seeing or hearing about the victimization of others, whereas being a victim heightens threat sensitivity to a level that is resistant to desensitization.

It should be noted that, although it was predicted that the high exposure class would show higher rates of anxiety than other classes, the three classes did not differ on anxiety symptoms. As noted earlier, exposure to community violence may result in physiological hypersensitization, rather than physiological desensitization. Adolescents exposed to violence may experience heightened hypervigilance to help recognize and respond to threatening situations (LeDoux, 1993; Schell et al., 2004). While no differences across groups for anxiety suggests that physiological hyperarousal is not an outcome of high levels of violence exposure, but this would contradict existing research (e.g., Fowler et al., 2009). The unexpected finding may be due to the measurement of anxiety in the current study. In particular, the findings may suggest that understanding the role of hyperarousal or hypersensitization to community violence may require the use of physiological markers of stress reaction (e.g., cortisol, heart rate), rather than self-reports of anxiety.

Although there were no specific predictions about differences between a victimization class and other classes, the between-the-classes differences are worth discussing. The existing literature provides few clues to understand why the high exposure and the victimization class differ with regard to desensitization outcome of depressive symptoms. Recent quantitative reviews show that direct victimization, in comparison with witnessing, is more detrimental to youth’s psychosocial functioning (e.g., Fowler et al., 2009). Specifically, victimization may be more related to increased depression because the violent events happen to that individual or someone close to them, as opposed to witnessing violence in the community that happens to a stranger/acquaintance (Lambert, Boyd, Cammack, & Ialongo, 2012). Therefore, symptoms may not be muted in youth who are victimized. However, research shows that victimization is more likely than other forms of violence to lead to emotional numbing (Fowler et al., 2009). Although the exact reasons for the difference between the high exposure class and the victimization class remain unclear, the findings underscore the need for additional research using person-based analyses. Interestingly, there was no difference between the victimization class and low exposure class on depressive symptoms. Research with youth of color from low-income urban communities (e.g., Grant et al., 2000) demonstrates that they are disproportionately exposed to multiple types of stressors that are predictive of higher levels of depressive symptoms. Perhaps youth with low exposure to violence are more vulnerable to the effects of other stressors because they lack the numbing or muting of emotions associated with an emotional desensitization effect.

Although the PAM suggest that youth exposed to high levels of violence may begin to show lower levels of emotional distress, the complexity of the experience of community violence exposure warrants the consideration of alternative explanations. For example, African American youth may be less likely to exhibit affective symptoms associated with dysphoric mood (Politano, Nelson, Evans, Sorenson, & Zeman, 1986), and for African American adolescents living in high-crime communities, expressions of sadness or low self-esteem may increase vulnerability to direct victimization (Reynolds, O’Koon, Papademetriou, Szczygiel, & Grant, 2001; White & Farrell, 2006). Rather than not experiencing depressive symptoms, which may signal emotional desensitization, adolescent males may be suppressing depressive symptoms to facilitate their ability to navigate dangerous communities (Cassidy & Stevenson, 2005). For instance, African American male adolescents reported that they were more likely to respond to victimization in the community by acting “tough” or carrying a weapon for protection rather than feeling depressed (Jenkins & Bell, 1994; Reese et al., 2001). Unfortunately, research has shown that this “acting tough” behavior actually increases the risk of victimization (e.g., Stewart, Schreck, & Simons, 2006). Also, exposure to high levels of violence may lead to post-traumatic stress disorder (PTSD) symptoms in adolescents (e.g., Dempsey, Stacy & Moely, 2000), and, in turn, youth who experience avoidance or numbing symptoms of PTSD may underreport their symptoms of depression (McCart et al., 2007; Mrug et al., 2008).

Inconsistent with the second prediction, the gender composition across the classes were equivalent, suggesting that males were not more likely than females to be classified in the high exposure class. Furthermore, desensitization effects were equivalent across gender. Although our findings are not in agreement with other reports of increased ECV and increased likelihood of desensitization among boys (e.g., Ng-Mak et al., 2002), our use of LCA differs from prior research and may more accurately reflect the naturally occurring patterns of ECV within the sample. Interestingly, supplemental analyses with the current sample demonstrate that males report higher levels of victimization than females. The finding that older youth were exposed to more violence in the community is consistent with other reports (Weist et al., 2001), but this finding is contested and may vary by specific community (Bell & Jenkins, 1993). Further research is needed to validate the findings for gender and age.

The results of the current study suggest that there may be systematic variability in exposure to community violence among African American youth in economically disadvantaged, urban neighborhoods. Prior variable-based research has consistently shown that this class is more likely to be exposed to violence in their communities than are other classes of youth. Our findings demonstrated that, while all the participants were exposed to at least one violent event, those participants exposed to especially high levels of community violence did not constitute the majority of the sample (Copeland-Linder et al., 2010; Ronzio et al., 2011). Similar to Copeland-Linder et al. (2010), future studies on profiles of community violence exposure should include protective factors as well as vulnerability factors to determine malleable factors that may predict class membership or distinguish between classes and can be targeted for intervention or prevention.

In addition, the variation among profiles in the current underscores the need to distinguish between witnessing and experiencing community violence. Specifically, although the low exposure class reported very low rates of witnessing violent acts in the community, they reported slightly higher (but still low) rates of direct victimization. Furthermore, the victimization class reported higher levels of direct and indirect victimization, but lower levels of other forms of violence. For example, this class was more likely to be threatened with a knife or gun or know that someone close to them was threatened with a knife or gun, rather than to see a crime in their neighborhood or hear gunshots. Distinguishing between different types of violent events is critical because prior research has shown that victimization is more detrimental to youth functioning than witnessing violence (Fowler et al., 2009; O’Donnell, Schwab-Stone, & Muyeed, 2002) or that victimization and witnessing are associated with different outcomes (McDonald & Richmond, 2008).

Limitations and Strengths

The current study is not without limitations. First, the current study is limited by reliance on youth self-report. Although youth’s self-report data of exposure to violence (e.g., Cooley-Strickland et al., 2009) and internalizing symptoms (e.g., Abela & Hankin, 2011) have shown adequate reliability, reliance on self-report of constructs may increase the likelihood of shared method variance. Due to the variability obtained, it does not appear that the problem of shared method variance affected the results of the current study. Nonetheless, future research would benefit from inclusion of multiple informants (e.g., parents, teachers) to ensure a broader perspective on the variables examined. Furthermore, the current study utilized cross-sectional data, thus providing little information about the stability of class membership over time. Furthermore, the cross-sectional nature does not account for the possibility of bi-directional associations. Specifically, youth with lower levels of emotional distress, as assessed by depressive symptoms, may be more likely to be in situations that are more dangerous and result in violence exposure. In addition, the use of a binary scale to report on violence exposure does not differentiate between experiencing an event only once and repeatedly experiencing the event, which may limit the ability to understand how frequency of exposure relates to emotional desensitization in the current study.

Finally, the generalizability of the findings are limited to African American early adolescents who reside in urban, low-income neighborhoods, as African American adolescents who reside in middle- to higher income neighborhoods are exposed to less community violence and associated stressors. With regard to generalizability, it should also be noted that our participation rate may limit generalizability within our target population. Our review of participation rates in 12 similar research studies with youth of color from urban communities suggests that rates range from 29% (e.g., Allison et al., 1999) to 93% (e.g., Cooley-Quille et al., 2001), but the average rate is about 65%. Research suggests that achieving high rates of participation using active parental consent in urban, under-resourced communities requires multiple contact with youth, multiple methods to obtain parental consent, and substantial incentives for schools, teachers, and students (O’Donnell et al., 1997; Secor-Turner, Sieving, Widome, Plowman, & Vanden Berk, 2010). Furthermore, research suggests that African American youth and families from urban, economically disadvantaged communities may experience research engagement barriers that require significant relationship building and multiple methods of community engagement (Breland-Noble, Bell, Burriss & Poole, 2012). While our participation rate was lower than the average, it should be noted that we relied on one method to obtain consent, and we were granted only two delivery attempts to students. Future research should incorporate multiple approaches to engage students and families. In light of the limitations, the current study extends prior research by utilizing person-based analyses to provide information on the natural patterns of violence exposure in a sample of communities with high rates of violence exposure. Also, few existing studies utilize both depressive and anxiety symptoms as outcomes for community violence exposure (Fowler et al., 2009) and as proxies of desensitization to violence (Gaylord-Harden et al., 2011). Thus, the current study both extends and supports prior research on desensitization to community violence exposure in African American youth.

Conclusion

In conclusion, the disparities in violence exposure experienced by the segment of African American youth residing in communities with elevated crime and limited resources are well documented. For these youth, exposure to community violence has a significant influence on daily life and impinges negatively upon optimal development (Kuther & Wallace, 2003), but adolescents who become desensitized to violence show a higher likelihood to exhibit violent or retaliatory behavior, without feeling emotional distress, thus perpetuating the perilous cycle of violence that plagues these communities. Future research on desensitization should incorporate both internalizing and externalizing problems to provide a more complete picture of the effects of community violence. Given the evidence that adolescence is a critical period for violence exposure, programs that can preempt or reduce desensitization outcomes in youth exposed to violence have the potential to yield large benefits.

Using person-based analyses to examine desensitization outcomes provides useful information for prevention and intervention efforts (Copeland-Linder et al., 2010), as it helps to identify a specific subgroup of youth who may be more likely to show lower rates of emotional distress (i.e., desensitization) in the context of community violence. Specifically, the identification of three distinct violence exposure classes supports the notion that exposure to violence and associated outcomes vary within group and suggests that there is a need to match the intensity and nature of interventions to the level of exposure, as well as the severity and type of youth’s emotional outcomes (Walker & Shinn, 2002). Furthermore, the class of participants experiencing the highest levels of violence exposure was not the largest of the three classes, which suggests that there may be a range of protective factors that are embedded in the families and communities that socialize these youth (Gaylord-Harden, Burrow, & Cunningham, 2012). Work similar to Copeland-Linder et al. (2010) is needed to identify factors that may provide protection against both exposure to violence and subsequent emotional and behavioral difficulties. In addition, future research should consider examining the stability of classes, as understanding how these processes develop and change over time may be crucial to prevention and intervention efforts.

Footnotes

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: This research was supported by the Grant on Empirical Advancement of Ethnic Minority Clinical Child and Adolescent Psychology from Division 53 of the American Psychological Association.

Author Biographies

Noni K. Gaylord-Harden is an associate professor of psychology at Loyola University Chicago. Her primary research interests have been in the investigation of stress, coping, and psychological symptoms in African American adolescents. Her more recent work focuses on exposure to community violence as a stressor for youth and how youth cope specifically with community violence exposure. She has received funding from Division 53 of the American Psychological Association, the Institute for Education Sciences, and the National Institute for Child Health and Human Development for her research efforts.

Daniel Dickson is a graduate student in the Clinical Psychology Doctoral Program at Loyola University Chicago. His main research interests include gender-linked patterns of psychosocial well-being versus ill-being, and the impact of stressors, coping, and cognitive processes on depressive affect.

Cynthia Pierre is a graduate student in the Clinical Psychology Doctoral Program at Loyola University Chicago. Her main research interests include understanding the different types of stressors faced by African American youth, the impact of these stressors on psychosocial functioning, and social support as a coping strategy.

References

Abela

J. R. Z.

Hankin

B. L.

(2011). Rumination as a vulnerability factor to depression during the transition from early to middle adolescence: A multiwave longitudinal study. Journal of Abnormal Psychology, 120, 259-271.

Aisenberg

Mennen

F. E.

(2000). Children exposed to community violence: Issues for assessment and treatment. Child and Adolescent Social Work Journal, 17, 341-360.

Akaike

(1987). Factor analysis and AIC. Psychometrika, 52, 317-332.

Allison

K. W.

Burton

Marshall

Perez-Febles

Yarrington

Kirsh

L. B.

Merriwether-DeVries

(1999). Life experiences among urban adolescents: Examining the role of context. Child Development, 70, 1017-1029.

Asparouhov

Muthén

B. O.

(2007). Wald test of mean equality for potential latent class predictors in mixture modeling. Retrieved from http://www.statmodel.com/download/MeanTest1.pdf

Bell

C. C.

Jenkins

E. J.

(1993). Community violence and children on Chicago’s southside. Psychiatry: Physical and Biological Processes, 56, 46-54.

Bergman

L. R.

Magnusson

(1997). A person-oriented approach in research on developmental psychopathology. Development and Psychopathology, 9, 291-319.

Boxer

Morris

A. S.

Terranova

A. M.

Kithakye

Savoy

McFaul

(2008). Coping with exposure to violence: Relations to aggression and emotional symptoms in three urban samples. Journal of Child and Family Studies, 17, 881-893.

Boyd

R. C.

Cooley

M. R.

Lambert

S. F.

Ialongo

N. S.

(2003). First-grade child risk behaviors for community violence exposure in middle school. Journal of Community Psychology, 31, 297-314.

10.

Breland-Noble

A. M.

Bell

C. C.

Burriss

Poole

H. K.

(2012). The significance of strategic community engagement in recruiting African American youth & families for clinical research. Journal of Child and Family Studies, 21, 273-280.

11.

Busby

D. R.

Lambert

S. F.

Ialongo

N. S.

(2013). Psychological symptoms linking exposure to community violence and academic functioning in African American adolescents. Journal of Youth and Adolescence, 42, 250-262.

12.

Carnagey

N. L.

Anderson

C. A.

Bushman

B. J.

(2007). The effect of video game violence on physiological desensitization to real-life violence. Journal of Experimental Social Psychology, 43, 489-496.

13.

Cassidy

R. F.

Stevenson

H. C.

(2005). They wear the mask: Hypervulnerability and hypermasculine aggression among African American males in an urban remedial disciplinary school. Journal of Aggression, Maltreatment, & Trauma, 11, 53-74.

14.

Chen

(2009). Adolescents’ internalizing behaviors after extreme violence exposure: A comparison of race and gender for African American and Asian American youth. Families in Society, 90, 145-152.

15.

Cooley-Quille

Boyd

R. C.

Frantz

Walsh

(2001). Emotional and behavioral impact of exposure to community violence in inner-city adolescents. Journal of Clinical Child Psychology, 30, 199-206.

16.

Cooley-Quille

Lorion

(1999). Adolescents’ exposure to community violence: Sleep and psychophysiological functioning. Journal of Community Psychology, 27, 367-375.

17.

Cooley-Strickland

Quille

T. J.

Griffin

R. S.

Stuart

E. A.

Bradshaw

C. P.

Furr-Holden

(2009). Community violence and youth: Affect, behavior, substance use, and academics. Clinical Child and Family Psychology Review, 12, 127-156.

18.

Copeland-Linder

Lambert

S. F.

Ialongo

N. H.

(2010). Community violence, protective factors and adolescent mental health: A profile analysis. Journal of Clinical Child and Adolescent Psychology, 39, 176-186.

19.

Dempsey

(2002). Negative coping as mediator in the relation between violence and outcomes: Inner-city African American youth. American Journal of Orthopsychiatry, 72, 102-109.

20.

Dempsey

Stacy

Moely

(2000). “Approach” and “avoidance” coping and PTSD symptoms in innercity youth. Current Psychology, 19(1), 28-45.

21.

Farrell

A. D.

Bruce

S. E.

(1997). Impact of exposure to community violence on violent behavior and emotional distress among urban adolescents. Journal of Clinical Child Psychology, 26, 2-14.

22.

Finkelhor

Hamby

S. L.

Ormrod

Turner

(2005). The Juvenile Victimization Questionnaire: Reliability, validity, and national norms. Child Abuse & Neglect, 29, 383-412.

23.

Finkelhor

Ormrod

R. K.

Turner

H. A.

(2009). Lifetime assessment of poly-victimization in a national sample of children and youth. Child Abuse & Neglect, 33, 403-411.

24.

Fitzpatrick

K. M.

(1993). Exposure to violence and presence of depression among low-income African American youth. Journal of Consulting and Clinical Psychology, 61, 528-531.

25.

Fitzpatrick

K. M.

Boldizar

J. M.

(1993). The prevalence and consequences of exposure to community violence among African American youth. Journal of the American Academy of Child & Adolescent Psychiatry, 32, 424-430.

26.

Fowler

P. J.

Tompsett

C. J.

Braciszewski

J. M.

Jacques-Tiura

A. J.

Baltes

B. B.

(2009). Community violence: A meta-analysis on the effect of exposure and mental health outcomes of children and adolescents. Development and Psychopathology, 21, 227-259.

27.

Funk

J. B.

Baldacci

H. B.

Pasold

Baumgardner

(2004). Violence exposure in real-life, video games, television, movies, and the Internet: Is there desensitization? Journal of Adolescence, 27(1), 23-39.

28.

Gaylord-Harden

N. K.

Burrow

Cunningham

J. A.

(2012). A cultural-asset framework for investigating successful adaptation to stress in African American youth. Child Development Perspectives, 6, 264-271.

29.

Gaylord-Harden

N. K.

Cunningham

J. A.

Zelencik

(2011). Effects of exposure to community violence on internalizing symptoms: Does desensitization to violence occur in African American youth? Journal of Abnormal Child Psychology, 39, 711-719.

30.

Gibson

C. L.

Morris

S. Z.

Beaver

K. M.

(2009). Secondary exposure to violence during childhood and adolescence: Does neighborhood context matter? Justice Quarterly, 26(1), 30-57.

31.

Gonzales

N. A.

Gunnoe

M. L.

Jackson

K. M.

Samaniego

R. Y.

(1995, June). Validation of a multicultural events scale for urban adolescents. Paper presented at the Biennial Conference of the Society for Community Research and Action, Chicago, IL.

32.

Gorman-Smith

Henry

D. B.

Tolan

P. H.

(2004). Exposure to community violence and violence perpetration: The protective effects of family functioning. Journal of Child Clinical and Adolescent Psychology, 33, 439-449.

33.

Gorman-Smith

Tolan

(1998). The role of exposure to community violence and developmental problems among inner-city youth. Development and Psychopathology, 10, 101-116.

34.

Grant

K. E.

O’Koon

J. H.

Davis

T. H.

Roache

N. A.

Poindexter

L. M.

Armstrong

M. L.

. . . McIntosh

J. M.

(2000). Protective factors affecting low-income urban African American youth exposed to stress. The Journal of Early Adolescence, 20, 388-417.

35.

Guerra

N. G.

Huesmann

L. R.

Spindler

(2003). Community violence exposure, social cognition, and aggression among urban elementary school children. Child Development, 74, 1561-1576.

36.

Ingoldsby

E. M.

Shaw

D. S.

(2002). Neighborhood contextual factors and early-starting antisocial pathways. Clinical Child and Family Psychology Review, 5(1), 21-55.

37.

Jenkins

E. J.

Bell

C. C.

(1994). Violence among inner city high school students and post-traumatic stress disorders. In Friedman

(Ed.), Anxiety disorders in African-Americans (pp. 76-88). New York, NY: Springer.

38.

Johnson

P. D.

Kliewer

(1999). Family and contextual predictors of depressive symptoms in inner-city African American youth. Journal of Child and Family Studies, 8, 181-192.

39.

Kerig

P. K.

Bennett

D. C.

Thompson

Becker

S. P.

(2012). “Nothing really matters”: Emotional numbing as a link between trauma exposure and callousness in delinquent youth. Journal of Traumatic Stress, 25, 272-279.

40.

Kliewer

Cunningham

J. R.

Diehl

Parrish

K. A.

Walker

J. M.

Atiyeh

. . . Mejia

(2004). Violence exposure and adjustment in inner-city youth: Child and caregiver emotion regulation skill caregiver-child relationship quality, and neighborhood cohesion as protective factors. Journal of Clinical Child and Adolescent Psychology, 33, 477-487.

41.

Kliewer

Kung

(1998). Family moderators of the relation between hassles and behavior problems in inner-city youth. Journal of Clinical Child Psychology, 27, 278-292.

42.

Kovacs

(1992). Children Depression Inventory manual. North Tonawanda, NY: Multi-Health Systems.

43.

Kuther

T. L.

Wallace

S. A.

(2003). Community violence and sociomoral development: An African American cultural perspective. American Journal of Orthopsychiatry, 73, 177-189.

44.

Lambert

S. F.

Boyd

R. C.

Cammack

N. L.

Ialongo

N. S.

(2012). Relationship proximity to victims of witnessed community violence: Associations with adolescent internalizing and externalizing behaviors. American Journal of Orthopsychiatry, 82(1), 1-9.

45.

LeDoux

J. E.

(1993). Emotional networks in the brain. In Lewis

Haviland

J. M.

(Eds.), Handbook of emotions (pp. 109-118). New York, NY: Guilford Press.

46.

Mandara

(2003). The typological approach in child and family psychology: A review of theory, methods, and research. Clinical Child and Family Psychology Review, 6, 129-146.

47.

Margolin

Gordis

E. B.

(2000). The effects of family and community violence on children. Annual Review of Psychology, 51, 445-479.

48.

Masten

A. S.

(1999). Ordinary magic: Resilience processes in development. American Psychologist, 56, 227-238.

49.

McCart

M. R.

Smith

D. W.

Saunders

B. E.

Kilpatrick

D. G.

Resnick

H. S.

Ruggiero

K. J.

(2007). Do urban adolescents become desensitized to community violence? Data from a national survey. American Journal of Orthopsychiatry, 77, 434-442.

50.

McDonald

C. C.

Richmond

T. R.

(2008). The relationship between community violence exposure and mental health symptoms in urban adolescents. Journal of Psychiatric and Mental Health Nursing, 15, 833-849.

51.

McLachlan

Peel

(2000). Finite mixture models. New York, NY: Wiley.

52.

McMahon

S. D.

Watts

R. J.

(2002). Ethnic identity in African American youth: Exploring links with self-worth, aggression, and other psychosocial variables. Journal of Community Psychology, 30, 411-431.

53.

Miller

L. S.

Wasserman

G. A.

Neugebauer

Gorman-Smith

Kamboukos

(1999). Witnessed community violence and antisocial behavior in high-risk, urban boys. Journal of Clinical Child Psychology, 28, 2-11.

54.

Moses

(1999). Exposure to violence, depression, and hostility in a sample of inner city high school youth. Journal of Adolescence, 22, 21–32.

55.

Mrug

Loosier

Windle

(2008). Violence exposure across multiple contexts: Individual and joint effects on adjustment. American Journal of Orthopsychiatry, 78, 70-84.

56.

Muthén

(2013). Mplus 7.1. Los Angeles, CA: Muthen & Muthen.

57.

Ng-Mak

D. S.

Salzinger

Feldman

R. S.

Stueve

A. C.

(2002). Normalization of violence among inner-city youth: A formulation of research. American Journal of Orthopsychiatry, 72, 92-101.

58.

Ng-Mak

D. S.

Salzinger

Feldman

R. S.

Stueve

A. C.

(2004). Pathologic adaptation to community violence among inner-city youth. American Journal of Orthopsychiatry, 74, 196-208.

59.

Nolen-Hoeksema

(1994). An interactive model for the emergence of gender differences in depression in adolescence. Journal of Research on Adolescence, 4, 519-534.

60.

Nylund

K. L.

Asparouhov

Muthen

B. O.

(2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling, 14, 535-569.

61.

O’Donnell

D. A.

Schwab-Stone

M. E.

Muyeed

A. Z.

(2002). Multidimensional resilience in urban children exposed to community violence. Child Development, 73, 1265-1282.

62.

O’Donnell

L. N.

Duran

R. H.

San Doval

Breslin

M. J.

Juhn

G. M.

Stueve

(1997). Obtaining written parent permission for school-based health surveys of urban young adolescents. Journal of Adolescent Health, 21, 376-383.

63.

Ozer

E. J.

Weinstein

R. S.

(2004). Urban adolescents’ exposure to community violence: The role of support, school safety, and social constraints in a school-based sample of boys and girls. Journal of Clinical Child and Adolescent Psychology, 33(3), 463-476.

64.

Patchin

J. W.

Huebner

B. M.

McCluskey

J. D.

Varano

S. P.

Bynum

T. S.

(2006). Exposure to community violence and childhood delinquency. Crime & Delinquency, 52, 307-332.

65.

Politano

P. M.

Nelson

W. M.

Evans

H. E.

Sorenson

S. B.

Zeman

D. J.

(1986). Factor analytic evaluation of differences between black and Caucasian emotionally disturbed children on the Children’s Depression Inventory. Journal of Psychopathology and Behavioral Assessment, 8, 1-7.

66.

Ramaswamy

DeSarbo

W. S.

Reibstein

D. J.

Robinson

W. T.

(1993). An empirical pooling approach for estimating marketing mix elasticities with PIMS data. Marketing Science, 12, 103-124.

67.

Reese

L. R. E.

Vera

E. M.

Thompson

Reyes

(2001). A qualitative investigation of perceptions of violence risk factors in low-income African American children. Journal of Clinical Child Psychology, 30, 161-171.

68.

Reynolds

C. R.

Richmond

B. O.

(1997). What I think and feel: A revised measure of children’s manifest anxiety. Journal of Abnormal Child Psychology, 25(1), 15-20.

69.

Reynolds

L. K.

O’Koon

J. H.

Papademetriou

Szczygiel

Grant

K. E.

(2001). Stress and somatic complaints in low-income urban adolescents. Journal of Youth and Adolescence, 30, 499-514.

70.

Ronzio

C. R.

Mitchell

S. J.

Wang

(2011). The structure of witnessed community violence amongst urban African American mothers: Latent class analysis of a community sample. Urban Studies Research, 2011, 1-9.

71.

Russo

Roccato

(2010). How long does victimization foster fear of crime? A longitudinal study. Journal of Community Psychology, 38, 960-974.

72.

Schell

T. L.

Marshall

G. N.

Jaycox

L. H.

(2004). All symptoms are not created equal: The prominent role of hyperarousal in the natural course of posttraumatic psychological distress. Journal of Abnormal Psychology, 113, 189-197.

73.

Schwartz

(1978). Estimating the dimension of a model. The Annals of Statistics, 6, 461-464.

74.

Sclove

(1987). Application of model-selection criteria to some problems in multivariate analysis. Psychometrika, 52, 333-343.

75.

Secor-Turner

Sieving

Widome

Plowman

Vanden Berk

(2010). Active parent consent for health surveys with urban middle school students: Processes and outcomes. Journal of School Health, 80(2), 73-79.

76.

Self-Brown

S. R.

LeBlanc

Kelley

M. L.

Hanson

Laslie

Wingate

(2006). Effects of community exposure to violence and parental mental health on the internalizing problems of urban adolescents. Violence and Victims, 21, 183-189.

77.

Singer

M. I.

Anglin

T. M.

Song

Lunghofer

(1995). Adolescents’ exposure to violence and associated symptoms of psychological trauma. Journal of the American Medical Association, 273, 477-482.

78.

Spano

Bolland

(2013). Disentangling the effects of violent victimization, violent behavior, and gun carrying for minority inner-city youth living in extreme poverty. Crime & Delinquency, 59, 191-213.

79.

Spano

Rivera

Bolland

(2010). Are chronic exposure to violence and chronic violent behavior closely related developmental processes during adolescence? Criminal Justice & Behavior, 37, 1160-1179.

80.

Stein

B. D.

Jaycox

L. H.

Kataoka

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.

81.

Stewart

E. A.

Schreck

C. J.

Simons

R. L.

(2006). “I ain’t gonna let no one disrespect me”: Does the code of the street reduce or increase violent victimization among African American adolescents? Journal of Research in Crime and Delinquency, 43, 427-458.

82.

Sullivan

T. N.

Kung

E. M.

Farrell

A. D.

(2004). Relation between witnessing violence and drug use initiation in early adolescence: Parental monitoring and support as protective factors. Journal of Clinical Child and Adolescent Psychology, 33, 488-498.

83.

Tolan

P. H.

Gorman-Smith

Henry

D. B.

(2003). The developmental ecology of urban males’ youth violence. Developmental Psychology, 39, 274-291.

84.

Walker

H. M.

Shinn

M. R.

(2002). Structuring school-based interventions to achieve integrated primary, secondary, and tertiary prevention goals for safe and effective schools. In Interventions for academic and behavior problems II: Preventive and remedial approaches (pp. 1-25). Silver Spring, MD: National Association of School Psychologists.

85.

Wang

C. P.

Brown

H. C.

Bandeen-Roche

(2005). Residual diagnostics for growth mixture models: Examining the impact of a preventive intervention on multiple trajectories of aggressive behavior. Journal of the American Statistical Association, 100, 1054-1076.

86.

Weist

M. D.

Acosta

O. M.

Youngstrom

E. A.

(2001). Predictors of violence exposure among inner-city youth. Journal of Clinical Child & Adolescent Psychology, 30, 187-198.

87.

White

K. S.

Bruce

S. E.

Farrell

A. D.

Kliewer

(1998). Impact of exposure to violence on anxiety: A longitudinal study of family social support as a protective factor for urban children. Journal of Child and Family Studies, 7, 187-203.

88.

White

K. S.

Farrell

A. D.

(2006). Anxiety and psychosocial stress as predictors of headache and abdominal pain in urban early adolescents. Journal of Pediatric Psychology, 31, 582-596.

89.

Woodward

L. J.

Fergusson

D. M.

(2001). Life course outcomes of young people with anxiety disorders in adolescence. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 1086-1093.

90.

Zimmerman

G. M.

Messner

S. F.

(2013). Individual, family background, and contextual explanations of racial and ethnic disparities in youths’ exposure to violence. American Journal of Public Health, 103, 435-442.