Urban Children's Perceptions of Violence | Pediatrics | JAMA Pediatrics | JAMA Network
[Skip to Navigation]
Sign In
January 2004

Urban Children's Perceptions of Violence

Author Affiliations

From the Division of Pediatric Emergency Medicine, Children's Memorial Hospital (Dr Sheehan and Ms Kim), the Department of Pediatrics, Feinberg School of Medicine, Northwestern University (Dr Sheehan and Ms Kim), and the Chicago Project for Violence Prevention, School of Public Health, University of Illinois at Chicago (Mr Galvin), Chicago, Il.

Arch Pediatr Adolesc Med. 2004;158(1):74-77. doi:10.1001/archpedi.158.1.74

Objective  To determine how preadolescent urban children conceptualize and experience violence in their lives.

Design  This qualitative study reports the results of focus groups designed to examine perceptions of violence among preadolescent urban children. Program directors were trained to conduct the sessions using a semistructured script. All groups were audiotaped or videotaped. The summaries were analyzed for recurring themes.

Setting  A community-based visual arts program for children designed to be a secondary violence-prevention program.

Participants  There were 12 focus groups of volunteer participants. Each consisted of 3 to 6 children aged 8 to 12 years, separated by sex and age. Fifty children participated: 27 boys and 23 girls.

Results  These children defined violence in a broader way than most adults would. Not only did the children identify shootings and stabbings as examples of violence, but they also considered violence to be any act that might hurt someone's feelings (such as cheating and lying) or any act accompanying violence (such as cursing and yelling). The boys and girls were very similar in their views except regarding the issue of intimate-partner violence. The girls were almost universally concerned about this issue, but the boys seemed noticeably unaware that intimate-partner violence was considered a form of violence. Most children felt safe at home, and almost no child felt safe at school. They looked to trusted adults to keep them safe.

Conclusions  Future investigators measuring the effect of violence-prevention activities on preteen children should be aware that their definition of violence may differ from that of young children and should be cognizant of potential sex differences, especially around the topic of intimate-partner violence. Those designing violence-prevention programs for children should consider engaging adult family members as well because children usually turn to them for safety.

FOR NEARLY 20 YEARS, VIOLENCE has been addressed as a public health issue, a health problem that can be understood and intervened upon.1 Our knowledge of the magnitude of the problem and its causes has increased greatly during this time, and a variety of interventions have been designed, tested, and found to be effective.2

Nevertheless, challenges remain in preventing violence, 1 of which is identifying valid, reliable tools to measure the effects of an intervention.3 This is especially problematic for younger at-risk children whose reading and comprehension levels may be low. Available measures also may not be culturally sensitive or specific.

Our own experience has demonstrated these and similar limitations of the assessment techniques most often used to evaluate violence-prevention programs. Chicago Youth Programs Inc (CYP) is a youth development organization that provides health care, tutoring, and recreation for children from birth through young adulthood in 3 Chicago, Ill, neighborhoods: Cabrini Green, Washington Park, and Uptown. The CYP serves 400 extremely poor children from minority backgrounds; most are from single-parent families. There is no cost for participation and no selection criteria for admission; eligibility for enrollment is based on available space. Most children learn about the program from friends and family. One of CYP's recreational activities is KidStART, a violence intervention that engages at-risk urban children in the visual arts. In 2000, a pilot study of KidStART attempted to explore the role of the visual arts in violence prevention and to assess the adequacy of the quantitative tools to measure its effects. Although the selected survey tools were described as appropriate for this age group, the children had a difficult time reading and understanding the questions. In addition, it became apparent that the investigators' definition of violence might not be compatible with the children's. Given the limitations of current measures of violence-prevention efforts, it seems useful to start with a more basic question: How do at-risk children perceive violence? Our study set out to address this question using focus groups, a tool that has been shown to be more effective for obtaining data from at-risk children. The information gathered from these focus groups helped us understand attributes of the intervention participants and gave us insight into future issues to consider in the design and implementation of violence-prevention interventions and measures of their effectiveness.


Because the lack of applicable, reliable quantitative tools made it difficult to evaluate the effectiveness of KidStART as a violence-prevention program, we decided to conduct a series of focus groups, which have been shown to be more effective in obtaining data from children who may have difficulty understanding or relating to conventional self-report surveys.4 The goal of these focus groups was not to evaluate the effectiveness of the intervention but to better understand the exposure, attitudes, and beliefs of preadolescent at-risk children related to violence. Using focus groups allowed us to follow up on survey questions and allowed the children to qualify their responses to questions. It was also possible for us to observe nonverbal responses such as gestures, smiles, frowns, and actions, which carry information that supplement (and may contradict) the verbal response.

From April 2000 through May 2001, our research team conducted 12 focus groups among participants in CYP. Each focus group consisted of 3 to 6 children aged 8 to 12 years, separated by sex and age. Fifty children participated: 27 boys and 23 girls. The focus groups were moderated by CYP program directors after they had received training from the research assistants. The program directors were chosen to guide the focus groups because they possessed the traits of authority and familiarity that were necessary for moderating small groups of children. Four focus groups were moderated by the Uptown program directors (3 groups of girls and 1 group of boys), 4 by the Cabrini Green program directors (2 groups of boys and 2 groups of girls), and 4 by the Washington Park program directors (1 group of girls and 3 groups of boys). Snacks and inexpensive goody bags were provided to the participants. Written informed consent was obtained from their parents, and the institutional review board of Children's Memorial Hospital, Chicago, reviewed the study.

The focus groups were conducted using a semistructured format. The moderators used a script with open-ended questions. Each focus group lasted between 20 and 30 minutes and was videotaped or audiotaped.

The focus groups were held in the community centers in which the KidStART programs were usually held. Each program moderator explained to the children that the purpose of the focus group was to learn what their thoughts were about violence. The focus group script is presented in the Figure 1. Children were also asked about their satisfaction with the KidStART program. Their responses to this question are not included in this article.

The focus group guide.

The focus group guide.

Each session had a research assistant who took notes. These notes supplemented the videotape or audiotape, and together they provided the data for this study. The research assistants (J.P.G. and L.E.K.) and the primary investigator (K.S.) reviewed the summaries to understand the children's experience with violence, identify themes shared among participants and groups, and identify areas of disagreement. No difference in data interpretation between the research assistants and the primary investigator was identified. This article presents recurring themes found in the data.


The data results were divided into 5 topic areas: definition of violence, experiences with violence, perceptions of safety, acceptability of using violence, and prevention of violence.

Definition of violence

Both boys and girls responded as one might expect in identifying examples of violent acts, naming things such as hitting, stabbing, and shooting. Many respondents added that talking about people, hurting their feelings, cursing, lying, and bullying were also violent acts. Only the girls identified rape (or "sexing") and sexual abuse as forms of violence. The girls were also more likely to describe relationship disputes as examples of violence. One girl described an example: when a girl is "old enough to have a boyfriend . . . he might turn violent on you." The boys tended to frame violence as associated with gang-related activity, playground fighting, and other types of non–intimate-partner violence. When asked, "What type of person commits violent acts?" participants most often responded with these or similar answers: bullies, people who grew up around gangs or in gangs, and "someone who is not happy in their life and has a lot of problems."

Experiences with violence

Several of the children had witnessed severe acts of violence, including shootings or stabbings, in their neighborhood. Some had family members who had been shot. Some participants gave detailed accounts of violence they had witnessed in their school or neighborhood. One participant described a man she had seen stabbed in her building bleeding from his shoulder, who was then taken to the hospital. The participant described being scared and worried. The participant explained that she was worried that if she were "in a fight like that one day, this could probably happen to me. That's why I try to stay away from them."

Many participants knew of other children who brought guns to school. In one such instance, the participant had witnessed another student trying to shoot the principal and a teacher. All the participants had witnessed fistfights in their neighborhood or at school.

Perceptions of safety

When asked where they felt most safe, most participants stated that they felt safest at home or at their grandparents' home. One participant who did not feel safe at home reported to be "scared at home since my mommy and daddy was fighting when I was 7 and now I am 10." Although most participants felt safe in their homes, many participants did not feel safe in their neighborhoods. One participant put it this way: safety was found "not in school, not in the store . . . , [but] at home." Another participant stated that the neighborhood felt unsafe because "people might jump you in the community." However, the most commonly stated unsafe place for the participants was at school. One participant explained, "People threaten each other and . . . bring a gun to school."

Acceptability of using violence

None of the participants felt it was acceptable to be the aggressor in violence, but most stated that it was acceptable to be violent in self-defense or if someone threatened a friend or family member. Many participants stated, "If someone hits you, you should hit them back." Some participants reported that they would try to talk to the aggressors first, prior to using physical aggression, or would report them to an adult.

Although most participants said that they would want to help defend friends and family members from violence, many participants seemed to understand the risk of becoming involved in a fight. When asked what he would do if someone were bullying his sister, an 11-year-old boy answered,

I wouldn't physically fight; I would go talk to them. I'd tell him to leave my sister alone. I would say it nicely. I wouldn't say it in a mean way, because that's what's going to make them think I want to fight. I am going to say it calmly. "Don't mess with my sister anymore." There are lots of ways to solve it, not always physical. I wouldn't seek revenge. If I try to do something, I put myself in danger. If I do something to somebody else, then someone in their family might do something back.

A girl reiterated this by explaining that if somebody "helped somebody [fight] but didn't start it, then you could get in trouble, even though you didn't start it."

Prevention of violence

When asked about how they could avoid becoming involved in violence, most participants stated that staying away from the "wrong type of people" and staying in safe places were the best ways. Most of the children said they could turn to a trusted adult or family member for help. Not all participants identified the police or authorities as providers of a safe environment. One girl said that having more police around would prevent violence, but a 10-year-old boy said that the police were violent when they were "checking people" and "searching me." One participant also said that "when you put the person in jail, it doesn't work because they keep on coming back, over and over again."


This study examines how a group of preadolescent children conceptualize and experience violence in their lives. Although the children, for the most part, defined violence through examples that most adults would concur are violent acts (eg, shootings and stabbings), they also included in their concept of violence any act that might hurt someone's feelings and motivate violence (cheating and lying) or might commonly accompany violence (cursing and yelling). That the children included acts such as talking about people and hurting their feelings—acts that are not physically violent and can take place in any environment—may seem incongruous with the grim reality of the severe forms of violence. However, the children's definition of violence may indicate that the environment that surrounds violence has as much weight in a child's mind as the actual incidence of violence.

The boys and girls were very similar in their views except regarding the issue of intimate-partner violence (IPV). The difference here was striking: the girls were quite concerned with this issue, but it did not seem to have even crossed the boys' minds.

Because of the qualitative nature of this study, we cannot make definite conclusions about this difference; however, our experience suggests that preadolescent girls have already formed concepts of IPV. This observation suggests the need for IPV awareness and prevention in early childhood and the need for further investigation into early childhood concepts of IPV. The data from other studies support our hypothesis that IPV awareness begins during childhood. In 1 study, nearly 10% of high school students reported being a victim of dating violence in the past 12 months. This statistic did not differ by sex, race, ethnicity, or grade.5 In Massachusetts, girls in grades 9 through 12 reported a lifetime prevalence of physical and sexual violence of approximately 20%.6

Many of the children in our study had witnessed severe forms of violence near their home or in their community. Although a couple of participants described witnessing domestic violence, most participants stated that their home was the place they felt safest. However, most participants felt that their neighborhood and areas immediately outside their home were not safe. Therefore, the safety of "home" most often referred to a place behind closed doors. Almost no child reported feeling safe at school, and most examples of violence given by participants were set at or near school.

These findings are consistent with an earlier study we performed with a representative sample of children who lived in Cabrini Green.7 Considering the social nature of violence, it is not surprising that most violence exposure for the school-aged participants was at or near school. However, although the participants felt most threatened at school, there is a relatively low incidence rate of violent crime in grade schools in comparison with the incidence of violent crime in their neighborhoods.

The children were remarkably pacifistic in their attitudes about using violence to solve conflict. There is no reason to think they would differ from adults in providing the most socially appropriate answer when asked.8 It was reassuring that the children knew to try to settle disagreements without the use of physical force; however, just because one knows the right thing to do does not always mean one will do it.9 We observed that many of the boys and a couple of the girls expressed how they would react to aggression toward a family member or friend by using nonverbal expressions, such as punching their own hand or mock fighting with two fists. These nonverbal expressions often contradicted their verbal message of pacifism.

The children looked to trusted adults to protect them from harm and sought refuge in safe places. The vulnerability of these children in their dependence on adults for security was underscored by 1 participant, who gave an example of a teacher and a principal who failed to punish a student for fear of retribution. Safety nets such as youth development organizations and other supportive communities (eg, churches and clubs) can perhaps provide caring adults and safe places for children, although it is naive to expect them to be as effective as a warm, nurturing home environment.

Because this was a qualitative, nonrepresentative sample, its results cannot be generalized. However, despite the inherent limitations of qualitative studies, they can provide a rich understanding of a subject to create hypotheses, develop appropriate survey tools, and refine research questions.10 Focus groups complement self-report surveys by allowing participants to explain their views and qualify their answers. The information collected from our focus groups suggests that self-report surveys among children should be culturally sensitive or specific, using comprehensible words and language familiar to young children. Future studies measuring the effect of violence-prevention activities on preteen at-risk children should consider a broader definition of violence and should be aware of possible sex differences, especially around the issue of IPV. We feel that the use of focus groups can be a valuable tool for evaluating beliefs about and exposure to violence among young at-risk children—the study of whom has been challenged by an unfortunate paucity of valid, reliable tools. The importance of intervening in early childhood for violence prevention has been established,2 but this can be achieved only with the appropriate intervention tools to measure success.

This qualitative study revealed that the participants are dependent on the safety of their homes and the presence of responsible, caring adults. Therefore, as adults, we are reminded of the need for strong families and the importance of providing safe havens and supportive community organizations for those children lacking a healthy family environment.


What This Study Adds

An evaluation of a violence-prevention program requires valid, reliable tools to measure the effectiveness of an intervention. The tools also need to be presented at the appropriate reading level and should be culturally specific. Unfortunately, among urban, poor preadolescent children, who are on the verge of becoming the highest risk group for violence, there is a severe lack of such appropriate measures.

To our knowledge, this is one of the first studies to qualitatively explore this population's concept of and experience with violence. The major contribution of this study is for future investigators. In designing violence-prevention studies involving this population, investigators can incorporate our findings when designing suitable evaluation tools. They can also use our report to help design their intervention.

Corresponding author: Karen Sheehan, MD, MPH, Children's Memorial Hospital, 2300 Children's Plaza No. 62, Chicago, IL 60614 (e-mail: ksheehan@northwestern.edu).

Accepted for publication July 14, 2003.

The implementation and evaluation of the KidStART program was funded in part by a grant from the Illinois Violence Prevention Authority, Chicago.

We thank Patricia Heinicke, Jr, for review of the manuscript.

Mercy  JARosenberg  MLPowell  KEBroome  CVRoper  WL Public health policy for preventing violence.  Health Aff (Millwood). 1993;127- 29Google ScholarCrossref
US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, National Institutes of Health, National Institute of Mental Health, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services, Youth Violence: A Report of the Surgeon General.  Rockville, Md US Dept of Health and Human Services2001;
Dahlberg  LLToal  SBBehrens  CB Measuring Violence-Related Attitudes, Beliefs, and Behaviors Among Youths: A Compendium of Assessment Tools.  Atlanta, Ga Centers for Disease Control and Prevention, National Center for Injury Prevention and Control1998;
Stewart  DWShamdasani  PN Focus Groups: Theory and Practice.  Thousand Oaks, Calif Sage Publications Inc1990;
Not Available, Dating violence page. Child Trends Databank Web site. Available at:http://www.childtrendsdatabank.org/indicators/66DatingViolence.cfmAccessed February 17, 2003
Silverman  JGRaj  AMucci  LA  et al.  Dating violence against adolescent girls and associated substance abuse, unhealthy weight control, sexual risk behavior, pregnancy, and suicidality.  JAMA. 2001;286572- 579Google ScholarCrossref
Sheehan  KDiCara  JALeBailly  SChristoffel  K Children's exposure to violence in an urban setting.  Arch Pediatr Adolesc Med. 1997;151502- 504Google ScholarCrossref
Dillman  D Mail and Telephone Surveys: The Total Design Method.  New York, NY John Wiley & Sons Inc1978;
Thompson  RSSacks  JJ Evaluating an injury intervention or program. Rivara  FPCummings  PKoepsell  TDGrossman  DCMaier  RVeds. Injury Control: A Guide to Research and Program Evaluation. Cambridge, England Cambridge University Press2001;196- 216Google Scholar
Crabtree  BFMiller  WL Doing Qualitative Research.  Thousand Oaks, Calif Sage Publications Inc1999;