Violent behavior in adolescents and parent-child cohabitation


Comportamientos violentos de adolescentes y co-habitabilidad parental-filial



Rogério Lessa HortaI; Bernardo Lessa HortaII; Ricardo Tavares PinheiroIII; Manuela KrindgesIV

IPrograma de Pós-Graduação em Saúde Coletiva. Unidade de Pesquisa e Pós-Graduação. Universidade do Vale do Rio dos Sinos (Unisinos). São Leopoldo, RS, Brasil
IIPrograma de Pós-Graduação em Epidemiologia. Departamento de Medicina Social. Universidade Federal de Pelotas. Pelotas, RS, Brasil
IIIPrograma de Pós-Graduação em Saúde e Comportamento. Escolas de Medicina e Psicologia. Universidade Católica de Pelotas. Pelotas, RS, Brasil
IVCurso de Graduação em Psicologia. Unisinos. São Leopoldo, RS, Brasil





OBJECTIVE: To analyze the association between violent behavior in adolescents and parent-child cohabitation.
METHODS: A population-based cross-sectional study with multiple-stage sampling was performed in the urban area of the city of Pelotas, Southern Brazil, in 2002. A total of 960 adolescents were interviewed using a self-applied questionnaire. The dependent variables (use of weapons and involvement in fights in the previous year were reported by adolescents) and the independent variable (parent-child cohabitation) were analyzed with the chi-square test and prevalence ratios, adjusted for age, sex, socioeconomic level and reporting of alcohol, tobacco or illicit drug use, both recently and throughout life.
RESULTS: Involvement in fights in the previous year was reported by 23% of participants and use of weapons by 9.6%. Prevalence ratios of occurrence of such behaviors was 1.38 (95% CI: 0.71; 2.68, p=0.34) for involvement in fights and 1.68 (95% CI: 1.06; 2.67, p=0.03) for use of weapons, including "adolescents living with the father, mother or both" as reference.
CONCLUSIONS: Parent-child cohabitation must be considered in policies aimed at preventing the use of weapons by children and adolescents, although it is recommended that care should be taken not to stigmatize children and adolescents who do not live with their fathers and mothers.

Descriptors: Adolescent Behavior. Violence. Parent-Child Relations. Family Relations. Socioeconomic Factors. Cross-Sectional Studies.


OBJETIVO: Analizar la asociación entre comportamientos violentos de adolescentes y la co-habitabilidad parental-filial.
MÉTODOS: Estudio transversal de base poblacional realizado en el área urbana de Pelotas, Sur de Brasil, en 2002, con muestreo en múltiples fases. Fueron entrevistados 960 adolescentes con utilización de cuestionario auto-aplicado. Las variables dependientes (uso de armas y envolvimiento en peleas en el último año fueron referidos por los adolescentes) y la variable independiente co-habitabilidad parental-filial fueron sometidas a análisis por la prueba e chi-cuadrado y la tasa de prevalencias, ajustada para edad, sexo, nivel socioeconómico relato de uso en la vida y reciente de alcohol, tabaco o drogas ilícitas.
RESULTADOS: El envolvimiento en peleas en el último año fue referido por 23% de los sujetos y el uso de armas por 9,6%. La tasa de prevalencia de ocurrencia de estos comportamientos fue de 1,38 (IC 95%: 0,71;2,68, p=0,34) para envolvimiento en peleas y de 1,68 (IC 95%: 1,06;2,67, p=0,03) para uso de armas, teniendo como referencia sujetos que co-habitaban con padre, con madre o con ambos.
CONCLUSIONES: La co-habitabilidad parental-filial debe ser considerada en políticas de prevención al uso de armas por niños y adolescentes, pero se recomienda el cuidado de no estigmatizar niños y adolescentes que no co-habitan con padres y madres.

Descriptores: Conducta del Adolescente. Violencia. Relaciones Padres-Hijo. Relaciones Familiares. Factores Socioeconómicos. Estudios Transversales.




According to the World Health Organization (WHO),ª violence is a common phenomenon in all countries and any individual has been directly or indirectly involved with situations that characterize a certain form of violence. Adolescents appear in this context as both perpetrators and victims of violent acts. In a study performed in the city of Pelotas, Southern Brazil, among those born in the 1982 cohort, deaths due to external causes were responsible for 64% of deaths in the 15-to-24-year age group.9 Violent behavior is frequently associated with the idea of youth and expected to be present in many situations.3 This may lead to positively valuing the use of force and seeking to develop abilities to achieve this, with an emphasis on beliefs that establish the use of violence as a means to resolve conflicts, in a recursive process.b

The following are among the main characteristics associated with violence: the ability to control one's own emotions; the ability to resolve problems; the feeling of being accepted by the family and community where one lives or not; and, especially, the quality of the relationship with caregivers.12 These characteristics refer to early experiences in the life of children, regulated by family composition and functioning.

At the same time that the family can play the role of psychosocial protection and provide an environment that promotes well-being, it can also be an example of violence. Situations of domestic violence directly affecting children and adolescents or their exposure to situations of conjugal violence can lead to violent acts, committed later in life by adolescents.12,14,17

Currently, there is a great diversity of family structures.2,10 In the new contexts, cohabiting is not a rule for family groups to be formed. It does not guarantee the quality of contact between adults and their children, it only enables the sharing of the same physical space, which can facilitate verbal and non-verbal contacts. Contact with the parents can regulate and be regulated by the development of specific beliefs.6 The belief in violent behavior as an effective and valued alternative for conflict resolution, for example, is an important part of the phenomena associated with the expression of violent acts.

Cohabitation can facilitate or hinder the establishment of a bond of attachment. High scores of attachment to the father who does not live together and/or the stepfather who lives together are found to be protective against the appearance of externalizing problems in adolescents (aggressive or criminal behaviors), although the greatest benefit occurs when there is attachment to both. On the other hand, scores of attachment to the mother who lives together has an inverse relationship with the occurrence of internalizing problems.11

By comparing adolescents who live neither with the father nor the mother with the group who live with both or at least one of them, the recent use of illicit substances (marijuana, cocaine and others) is more frequently reported by those who do not cohabit with any of them. Alcohol use was similarly reported by these groups.8 The use of these substances is associated with the occurrence of violent behaviors.3,6

Parent-child cohabitation seems to have an impact on psychosocial vectors of violent behavior in adolescents and non-cohabitation is ever more present in modern times. However, a review of databases in the last ten years did not find studies that assessed the differences in occurrence of violent behaviors in adolescents who live with their parents or not.

The present study aimed to analyze the association between violent behaviors in adolescents and parent-child cohabitation.



This study was conducted in the city of Pelotas, Southern Brazil,3,8,10 in 2002. A total of 960 adolescents of both sexes, aged between 15 and 18 years, were interviewed.

Multiple-stage sampling was used, based on the 448 census tracts of the city's urban area. In 2001, it was estimated that there were approximately 104,000 households and 18,000 adolescents in the age group selected for this study.c A total of 7,740 households in 90 census tracts were visited by 15 interviewers. In each randomly selected household, all eligible adolescents were interviewed, after written consent was obtained from the parents or responsible adults. The number of interviews conducted per household was not recorded and no institutions were visited.

No interviews were conducted with 79 (7.6%) of the 1,039 adolescents located. Such losses occurred due to refusal of parents or responsible adults to consent to their children's participation, refusal of the adolescents themselves or adolescents' not having been found in the home after three visits.

Interviews were conducted individually and self-applied questionnaires were used, without personal identification data. In the cases of illiterate adolescents or those who reported having difficulty in understanding the questionnaire, questions were read out by the interviewer, who marked the respective responses.

The violent behavior of an adolescent was evaluated with the following items: involvement in fights in the 12 months prior to the interview (yes; no) and use of weapons of any type in the 12 months prior to the interview (yes; no).

The independent variables used were: sex (male; female); age (in complete years); socioeconomic level (A, B, C, D and E), defined according to the Scale of Social Class Indicators of the Brazilian Association of Population Studies (ABEP) and based on the scores achieved by the sum of material assets;3 alcohol, tobacco and illicit drug use (throughout life; recent use in the 30 days prior to the interview) and parent-child cohabitation (living with father, mother or both; not living with either parent).

A total of 5% of interviews were randomly selected and certain questions were chosen to be repeated by another interviewer for quality control of the study. No cases of falsification of interviews or participants refusing the first contact with the research group were identified.

The chi-square test with a 5% significance level was used for the analysis to test differences in proportions and prevalence ratios (PR), with 95% confidence intervals (95% CI). Prevalence ratio was controlled for sex, age, adolescent's social class and reporting of alcohol, tobacco or illicit drug use, whether recently or throughout life.

The EpiInfo program was used to analyze data. This research project was approved by the Research Ethics Committee of the Universidade Católica de Pelotas - UCPel, Process 0347-2/01-EM. Adolescents who showed signs of psychological stress were instructed on the possibility of seeking care at the Clínica Escola do Curso de Psicologia da UCPel.



A total of 960 questionnaires were completed, of which 87.3% of adolescents lived with their father, mother or both.

Table 1 shows the absolute and relative frequencies of involvement in fights or use of weapons in the last 12 months prior to the interviews. Involvement in fights in the last 12 months prior to interviews was reported by 23.0% of participants, while 9.6% reported use of weapons.



Table 2 shows that parent-child cohabitation was not associated with involvement in fights, even after control for sex, age and socioeconomic level of adolescents. In contrast, use of weapons showed an association. After adjustment for confounding factors (socioeconomic level, sex and age), the PR was 1.68 (95% CI: 1.06;2.67), p=0.03. The possibility that the greater use of weapons among adolescents who do not live with their parents was being mediated by illicit drug use was tested, although there was no significant change in the PR of 1.57 (95% CI: 0.99; 2.50) after controlling for this variable. A total of 33 adolescents reported living with the father only, thus reducing power by stratifying the analysis, considering cohabitation with the father and mother separately. With regard to use of weapons, the groups who lived with the father or mother exclusively had a behavior similar to that of the group who lived with both parents, and different from that of the group who did not live with either. The prevalence ratios for use of weapons, including the group who lived with both parents as reference, were 1.78 (95% CI: 1.09;2.90) (p=0.01) for those who did not live with either, 0.72 (95% CI: 0.17;3.00) (p=0.65) for those who lived with the father exclusively, and 1.20 (95% CI: 0.75;1.91) (p=0.44) for those who lived with the mother exclusively.




Considering only those adolescents who lived with at least one parent and those who did not live with either, it is observed that reporting the use of weapons in the previous year was significantly associated with non-parent-child cohabitation, which was not found for involvement in fights. If "use of weapons" is seen as predisposition to violence and "involvement in fights" and "use of weapons" are considered to be associated with each other in this same group of adolescents,16 it is understood that adolescents who do not live with any of their parents are more predisposed to involvement in violent situations, such as the behaviors studied here. Data from the present study and its cross-sectional design are not sufficient to conclude about the direction of such association.

Weapons can be associated with the search for a subjective feeling of security, in addition to other symbolic values. Violence takes on several forms in different population groups: moral, psychological, physical, verbal and sexual violence and violence in robberies and traffic, among others. All external causes of deaths involve a direct or indirect form of violence. Involvement in fights, here understood as a dispute or misunderstanding that includes several forms of physical aggression between two or more individuals, whether inside or outside the home,19 is the most common and simplest form of violence experienced by children and adolescents. The use of firearms or cutting weapons does not represent an act of violence in itself, but rather the expectation of a violent situation or the predisposition to experiences of violence. Even among professionals who own weapons, such as police officers or security guards, the ownership of such involves their preparation for situations of use of full force.3,7,13

The family and the home are also characterized as environments where situations of violence are experienced and where learning from such experiences occurs. For a long time, studies on families and family relations aimed to evaluate effects or possible effects of divorce of parents on child behavior in general. For this reason, these studies compare cohabitation between children and their father and mother and that between children and their father or mother exclusively.16 The results point to the need of considering parent-child cohabitation and its effects as present when the children live with either of their parents.

The family, its dynamics and the environment of shared life of familial groups may represent both a protective structure and a social group of risk.14 In this sense, one of the limitations of the present study is the fact that it did not qualify the participants' shared family environment as violent or not. Situations of family violence suffered or witnessed by children and adolescents in the domestic environment are an important part of the onset of violent behaviors in adolescent populations.17,d Cohabitation may or may not be involved with the dynamics of abusive families. The family does not lose its social functions and its members do not necessarily lose the possibility of a shared life because they do not live in the same home.18 A violent family environment may not have its effects changed by non-cohabitation. In addition, no sub-groups were characterized among the families, in terms of the maintenance or not of contact among fathers, mothers and adolescents. Moreover, no cases of non-cohabitation due to death of parents were emphasized. Possible differences among these sub-groups cannot be analyzed. Not living in the same home may be associated with specific dynamics, although it does not necessarily mean lack of contact.

The possibility of information being omitted by adolescents must be considered, due to the nature of the theme. The variables studied are seen as inadequate behaviors by society and tend to be repressed. The use of a self-applied and confidential questionnaire and the choice for only two variables with objective responses that are easily understandable sought to reduce the occurrence of classification errors. Lack of knowledge about the number of adolescents interviewed per household prevented the intraclass correlation from being controlled in the analysis. It is estimated that such correlation was small, because institutions were not visited and because of the relationship between the number of households visited and the number of adolescents found. This relationship corresponds to less than one adolescent aged between 15 and 18 years per household.

Firearms are not considered as the cause of violence, although they must be seen as an instrument used for violent purposes. They can be used as an instrument of affirmation and show of ability and power. In addition to this, there is the ease of access to weapons and the fact that the culture is favorable, valuing shows of force and power.19 Even if owning and using a weapon is associated with arguments for one's defense, this is invariably a situation of conflict with the law in the case of such population group, because access to weapons is prohibited to younger than 18 years, the maximum age of participants in this study. Even if not cohabiting does not imply abandon or loss of contact, a condition of lack of protection seems to be present, once the legal determination does not prevent access to weapons. It should be emphasized that parent-child cohabitation does not stop one from reporting the use of weapons or involvement in fights, as they do not seem sufficiently cautious. The occurrence of violent behaviors among adolescents in this study has also been reported elsewhere.7

Assuming that the behaviors studied indicate a situation of violence experienced (involvement in fights) and predisposition to violent acts (use of weapons), a relevant situation of lack of protection is what affects this population. Not only did these adolescents become involved in fights and have access to weapons, but they were not discouraged from doing so either.

One must be careful when analyzing information such as this and only consider lack of protection in the perspective of the individual and their family or social network. Likewise, one should be careful to avoid distorted interpretations due to class or economic interests.15 The fact that adolescents do not live together with their parents does not necessarily indicate that they are unprotected. The greatest occurrence of use of weapons among adolescents who did not live with their father or mother indicates that something in the structure that replaces cohabitation contributes to the occurrence of such behavior. This aspect does not seem to be alcohol or other drug use, although the literature frequently indicates its association with use of weapons.13

This population of adolescents is not subject to campaigns of registration of firearms or campaigns of disarmament involving legal procedures, because their contact with weapons is illegal.

Protection of adolescents may be characterized as more evident levels of discouragement of the use of violent means to resolve conflicts, associated with the promotion of the ability of resolving crises and tensions. The development of social abilities, which can be taught or encouraged in community programs and schools, may be the action of choice. Initiatives to promote a reconnection with the land and the local culture, which also raise self-esteem, can also have an impact on this phenomenon.4

The importance of putting existing laws into effect and actual government action in daily life should be emphasized. Additional repressive measures of an individual nature are not necessary. Moreover, severe punishments in family environments do not need to be promoted, because they are usually associated with psychological problems.1 Brazil has higher indices of occurrence of behaviors that violate the law than Canada, where repressive and punitive measures, similar to the Brazilian ones, are applied, thus suggesting that sanctions alone are not sufficient to resolve this issue.5 The government needs to be present in the communities, preventing the illegal access to weapons in different areas of the country and listening and encouraging local participation, which can achieve the goals of valuing and recovering bonds and self-esteem.

The ownership and use of weapons by adolescents needs to be the focus of specific actions and the parent-child cohabitation variable can be used as an indicator of higher probability of the occurrence of such behavior. The child-adolescent populations that do not live with their parents can be identified and provided special care, although not exclusive. Attention must be paid so that there is no stigmatization of the young population that does not live with their parents, because the reporting of use of weapons was also present among those who live with their parents. Distorted data interpretations of this study can lead to simplified rules, disregarding a relevant number of adolescents.

Stigmatization can also affect adolescents in general, especially due to the high occurrence of such behaviors in this population. Stating that the aggressiveness and violence committed by young people are expected is something that can also cause these behaviors to become common, thus strengthening and spreading them. The growing concern about the understanding of this phenomenon should consider each adolescent as an individual and avoid generalizations. This makes it necessary for the analysis of violent events to include socio-environmental factors, such as family, gender, social class, nationality and ethnicity, among others.3,14

In conclusion, the "involvement in fights" outcome was not dependent on living with the father or mother, whereas use of weapons was. The prevalence of such outcomes and their magnitude, especially in terms of use of weapons, is alarming and should serve as reference for other studies. It is recommended that perspectives that cause violence to become common, such as the indication that violent behaviors are expected, be avoided.



1. Bordin IAS, Paula CS, Nascimento R, Duarte CS. Severe physical punishment and mental health problems in an economically disadvantaged population of children and adolescents. Rev Bras Psiquiatr. 2006;28(4):290-6. DOI:10.1590/S1516-44462006000700008        

2. Costa LF, Féres-Carneiro T. Reorganizações familiares: as possibilidades de saúde a partir da separação conjugal. Psicol Reflex Crit. 1992;8(Supl):495-504.         

3. Cruzeiro ALS, Silva RA, Horta BL, Souza LDM, Faria AD, Pinheiro RT, et al. Prevalência e fatores associados ao transtorno da conduta entre adolescentes: um estudo de base populacional. Cad Saude Publica. 2008; 24(9):2013-20. DOI:10.1590/S0102-311X2008000900007        

4. De La Barra F, Toledo V, Rodriguez J. Estudio de salud mental en dos cohortes de niños escolares de Santiago occidente. I: Prevalencia y seguimiento de problemas conductuales y cognitivos. Rev Chil Neuro-Psiquiatr. 2002;40 (1):9-21. DOI:10.4067/S0717-92272002000100002        

5. Gallo AE. Atuação do psicólogo com adolescentes em conflito com a lei: a experiência do Canadá. Psicol Estud. 2008;13(2):327-34. DOI:10.1590/S1413-73722008000200015        

6. Garbarino J. Why are adolescents violent? Cienc Saude Coletiva. 2009;14(2):533-8. DOI: 10.1590/S1413-81232009000200021        

7. Guralh SA, Migliorini SRO. Adolescência - violência: desperdícios de vidas. Rev Bras Educ. 2008;13(38):399-400. DOI:10.1590/S1413-24782008000200016        

8. Horta RL, Horta BL, Pinheiro RT. Drogas: Famílias que protegem e que expõem adolescentes ao risco. J Bras Psiquiatr. 2006;55(4):268-72. DOI:10.1590/S0047-20852006000400002        

9. Horta BL, Gigante DP, Candiota JS, Barros FC, Victora CG. Monitorização da mortalidade na coorte de nascimentos de 1982 a 2006, Pelotas, RS. Rev Saude Publica. 2008;42(Supl 2):108-14. DOI:10.1590/S0034-89102008000900015        

10. Horta RL, Horta BL, Pinheiro RT, Morales B, Strey MN. Tabaco, álcool e outras drogas entre adolescentes em Pelotas, Rio Grande do Sul, Brasil: uma perspectiva de gênero. Cad Saude Publica. 2007;23(4):775-83. DOI:10.1590/S0102-311X2007000400005        

11. King V. The Antecedents and Consequences of Adolescents Relationships With Stepfathers and Nonresident Fathers. J Marriage Fam. 2006;68(4):910-28. DOI:10.1111/j.1741-3737.2006.00304.x        

12. Phebo L, Moura ATMS. Violência urbana: um desafio para o pediatra. J Pediatr (Rio J.). 2005;81(5):187-96. DOI:10.1590/S0021-75572005000700009        

13. Pillon SC, O'brien B, Chavez KAP. The relationship between drugs use and risk behaviors in brazilian university students. Rev Latino-Am Enfermagem. 2005;13(Spec 2):1169-76. DOI:10.1590/S0104-11692005000800011        

14. Robinson J, Haydes P, Mantz-Simons L. Children's story stem responses: A measure of program impact on developmental risks associated with dysfunctional parenting. Child Abuse Neglect. 2000;24(1):99-110. DOI:10.1016/S0145-2134(99)00117-9        

15. Santos MFS, Alessio RLS, Silva JMMN. Os adolescentes e a violência na imprensa. PsicTeor Pesq. 2009;25(3):447-52. DOI:10.1590/S0102-37722009000300020        

16. Silva RA, Jansen K, Godoy RV, Souza LDM, Horta BL, Pinheiro RT. Prevalência e fatores associados a porte de arma e envolvimento em agressão física entre adolescentes de 15 a 18 anos: estudo de base populacional. Cad Saude Publica. 2009;25(12):2737-2745. DOI:10.1590/S0102-311X2009001200020        

17. U.S. Preventive Services Task Force. Screening for family and intimate partner violence: recommendation statement. Ann Fam Med. 2004;2(2):156-60. DOI:10.1370/afm.128        

18. Wagner A, Ribeiro LS, Arteche AX, Bornholdt EA. Configuração familiar e o bem-estar psicológico dos adolescentes. Psicol Reflex Critica. 1999;12(1):147-56. DOI:10.1590/S0102-79721999000100010        

19. Zaluar A, Leal MC. Violência extra e intramuros. Rev Bras Ci Soc. 2001;16(45): 145-64. DOI:10.1590/S0102-69092001000100008        



Bernardo Lessa Horta
R. Triunfo, 3080 - Laranjal
96090-790 Pelotas, RS, Brasil
E-mail: blhorta@uol.com.br

Received: 11/29/2009
Approved: 4/23/2010
The present research was funded by the Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul (FAPERGS - Process 01/1552.9).



The authors declare that there are no conflicts of interest.
a Organización Mundial de la Salud. Informe mundial sobre la violencia y la salud: resumen. Washington; 2002.
b Abramovay M, Waiselfisz JJ, Andrade CC, Rua MG. Gangues, galeras, chegados e rappers: juventude, violência e cidadania nas cidades da periferia de Brasília. Rio de Janeiro: Garamond; 2004.
c Instituto Brasileiro de Geografia e Estatística. Resultados da amostra do Censo Demográfico 2000 - Malha municipal digital do Brasil: situação em 2001. Rio de Janeiro; 2004.
d Minayo MCS. Contextualização do debate sobre violência contra crianças e adolescentes. In: Ministério da Saúde. Violência faz mal à saúde. Brasília; 2004. p.13-6.

Faculdade de Saúde Pública da Universidade de São Paulo São Paulo - SP - Brazil
E-mail: revsp@org.usp.br