|
Abstract
|
OBJECTIVE--To examine the extent to which adolescents are exposed to various types of violence as either victims or witnesses, and the association of such exposure with trauma symptoms; specifically, the hypotheses that exposure to violence will have a positive and significant association with depression, anger, anxiety, dissociation, posttraumatic stress, and total trauma symptoms. DESIGN AND SETTING--The study employed a survey design using an anonymous self-report questionnaire administered to students (grades 9 through 12) in six public high schools during the 1992-1993 school year. PARTICIPANTS--Sixty-eight percent of the students attending the participating schools during the survey participated in the study (N = 3735). Ages ranged from 14 to 19 years; 52% were female; and 35% were African American, 33% white, and 23% Hispanic. RESULTS--All hypotheses were supported. Multiple regression analyses of the total sample revealed that violence exposure variables (and to a lesser extent, demographic variables) explained a significant portion of variance in all trauma symptom scores, including depression (R2 = .31), anger (R2 = .30), dissociation (R2 = .23), posttraumatic stress (R2 = .31), and total trauma (R2 = .37). CONCLUSIONS--A significant and consistent association was demonstrated linking violence exposure to trauma symptoms within a diverse sample of high school students. Our findings give evidence of the need to identify and provide trauma-related services for adolescents who have been exposed to violence.
VioLit summary:
OBJECTIVE:
The goal of this study by Singer et al. was to investigate the degree to which adolescents are exposed to violence, as victims or witnesses, and how exposure to violence relates to an outcome of trauma symptoms.
METHODOLOGY:
The authors conducted a quasi-experimental, primary analysis, of cross-sectional survey data. A 25 minute self-report questionnaire was administered, by the authors, to all high school students attending six public high schools: Two in Cleveland, Ohio, one in a Cleveland suburb; a small city high school in north-east Ohio; and two in Denver, Colorado. 3,735 students were included in the final sample out of a total of 5,509 students attending the schools. The study was conducted during the 1992-1993 school year during school hours. Participants lived mostly in predominantly lower socioeconomic areas, except for those attending the Cleveland suburban school, which was an upper-middle class community.
The self report instrument was designed to elicit demographic information (e.g., race, gender, parental structure, mother's/father's educational level). Recent physical violence exposure was assessed using a 22 item scale which measured five categories of violent behaviors. These were: threats, slapping/hitting/punching, beatings, knife attacks and shootings. Items for the first three types of violent behavior requested site specific information on where the act had taken place (e.g., at home, school and/or in the neighborhood). The authors reported that items related to knife attacks and shootings were not site specific. Information regarding direct experience of, or being witness to, violence in the past year was also requested. Frequency of violence exposure was measured using a 6 point Likert scale ranging from "never" (0 score) to "almost everyday" (5 score).
Past violence exposure was measured using a ten item scale. These items related to the witnessing of, or experience of, specific violent acts that occurred previous to, and not including, violence witnessed or experienced during the past year. Items requesting site specific information were not included in this scale. Violence frequency was assessed using a 4 point Likert scale, ranging from "never" (0 score), to "very often" (3 score). Sexual abuse/assault exposure was also assessed. Respondents were asked whether they had been forced into a sexual act during the past year, or while growing up but not in the past year. Whether they had witnessed the same type of violation towards someone else was also requested. Similar Likert scales were used. A subject's lifetime experiences or witnessing of sexual abuse/assault were calculated by obtaining the average score for recent and past exposure. The Trauma Symptom Checklist for Children (TSC-C) was used to measure childhood trauma/abuse sequelae in a format understandable to children as young as eight years old. The instrument consists of 54 items and six subscales: anxiety (alpha reliability (r)=.85), depression (r=.89), posttraumatic stress (r=.86), dissociation (r=.83), anger (r=.84) and sexual concerns (r=.68). The overall cronbach's alpha reliability score was .96 for 54 items. A 4 point Likert scale was used, "never" (0) to "almost all the time" (3). An overall cronbach's alpha reliability score for this sample was .95. The authors hypothesized that violence exposure was positively and significantly related to depression, anger, anxiety, dissociation and posttraumatic stress. Overall psychological trauma status was also hypothesized to be positively and significantly associated with violence exposure. Psychological trauma status was defined by the authors as "the degree to which adolescents report symptoms of depression, anxiety, posttraumatic stress, dissociation and/or anger."
The authors conducted principal components analyses on recent (22 items), and past (10 items), violence exposure items to reduce the number of variables. This allowed for the maximum possible amount of variance in violence exposure items to be accounted for. The authors then conducted hierarchical multiple regression analyses on the TSC-C to ascertain the effects of violence exposure on psychological health status.
FINDINGS/DISCUSSION:
The authors found that white students were underrepresented in the Ohio school. Females were overrepresented and african-americans under-represented in one of the Denver schools. In the other Denver school, Hispanics were underrepresented. The authors found that the mean age for the total sample was 16 years (SD, 1.2 years, range, 14-19 years). Females comprised 52% of the sample. 35% of the respondents were african-american, 33% were white, 23% hispanic and 9% were other. Approximately (28%) of the students had at least one college graduated parent. 56% of students had at least one high school graduated parent. Neither parent had graduated from high school for 16% of the students. 53% of students lived with both parents. since levels of victimization varied by school, the authors presented percentage ranges as their findings. The authors found that 33% to 44% of males reported being slapped/hit/punched at school, and 3% to 22% had been beaten or mugged in their neighborhoods within the past year. Three percent to 33% of males reported being shot or shot at in the previous year and 6% to 16% stated they had been attacked or stabbed with a knife. Female victimization rates were: being slapped/hit/punched at home, 34% to 56%; attacked/stabbed with a knife, 0% to 9%; shot or shot at, 0.5% to 12%; and sexually abused/assaulted, 12% to 17%. Overall, apart from physical victimization and sexual abuse/assault, females reported lower victimization rates. Frequencies of witnessing recent acts of violence for males varied across schools and specific acts included: someone else sexually abused/assaulted, 9% to 21%; someone else beaten/mugged at school, 32% to 82%; or in the neighborhood, 11% to 72%; someone else attacked/stabbed with a knife, 14% to 46%; and someone else shot at or shot, 5% to 62%. For females, results revealed that for someone else sexually abused/assaulted, 15% to 20%; someone else beaten up or mugged at school, 24% to 82%; someone else attacked/stabbed with a knife, 7% to 44%; and someone else shot or shot at, 5% to 49%.
Principal component analyses revealed that for recent violence exposure, five variable clusters accounted for 57% of the variance. These included: 1) being witness to neighborhood violence (3 items), 2) being witness to, or victim of, violence at home (6 items), 3) being a witness of violence at school (3 items), 4) being a witness or victim of a shooting/knife attack (4 items), and 5) being a victim of neighborhood or school violence (3 items). For past violence exposure, three variable clusters were extracted and accounted for 64% of the variance. These clusters included: 1) being a witness of past violence (3 items), 2) being exposed to very serious violence (4 items), and 3) being a victim of past violence (3 items). Multiple regression analyses revealed that after controlling for demographic variables, the violence exposure variable accounted for 29% of the variance in the TSC-C score and ranged from 19% to 27% across symptom subscales. The authors reported that females had the greatest scores on the total TSC-C (B=-.26, p<.01), and on the anxiety (B=-.32, p<.01), depression (B=-.31, p<.01), posttraumatic stress (B=-.25, p<.01), dissociation (B=-.16, p<.01), and anger (B=-.11, p<.01) subscales. The authors stated that the most important violence exposure variables were: being a witness or victim of violence at home (B range, .16 to .23, p<.01); having a history of being threatened, slapped/hit/punched and/or beaten/mugged (B range, .14 to .18, p<.01); and being the victim of sexual abuse/assault. (B range, .08 to .20, p<.01). Further important variables included exposure to a shooting or knife attack (B=.21, p<.01) and being a witness of past violence (B=.21, p<.01).
The authors found that adolescents, in this study, had been exposed to substantial levels of violence. The authors stated that adolescent males, attending city schools, experienced the greatest victimization, and were witnesses of severe violence. The authors reported that youths in suburban areas experienced relatively low severe violence exposure, however, exposure to less severe violence at school was greater. It was stated that females, for the most part, experienced greater levels of violence exposure at home, as well as, being witness to, or a victim of, sexual abuse/assault. Female gender was the greatest predictor of trauma symptoms.
The authors concluded that being exposed to violence was reliably, statistically, associated with trauma symptoms. The authors stated that their hypotheses were supported.
AUTHORS' RECOMMENDATIONS:
The authors recommended that health care professionals (included school guidance counsellors) be trained to 1) screen youths for violence exposure, 2) to explore how trauma occurred if adolescents are carrying physical injuries, and 3) to be responsive to the needs of those victimized. The authors stated that the results of their study have supported previous empirical research which has shown a relationship between violence exposure and psychological distress symptoms. Also, it was contended that future studies should be longitudinal.
(CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado)
Exposure to Violence
Witnessing Violence Effects
Juvenile Witness
Juvenile Victim
Late Adolescence
Early Adolescence
Ohio
Colorado
Senior High School Student
Psychological Victimization Effects
Post-Traumatic Stress Disorder
Juvenile Depression
Juvenile Anger
Juvenile Anxiety
Anger Causes
Anxiety Causes
Depression Causes
Witness Depression
Witness Adjustment
Victim Adjustment
Victim Anger
Victim Anxiety
Victim Depression
Juvenile Adjustment
Emotional Adjustment
Language: en |