Co-Occurrence of Interparental Violence and Child Physical Abuse and It’s Effect on the Adolescents’ Behavior

Co-Occurrence of Interparental Violence and Child Physical Abuse and It’s Effect on the Adolescents’ Behavior

Co-Occurrence of Interparental Violence and Child Physical Abuse and It’s Effect on the Adolescents’ Behavior

Permalink:

This study analyzed the impact of the co- occurrence of parental and interparental violence on the behavior of adolescents. Results reveal that the co- occurrence of interparental violence and child physical abuse has a significantly greater negative impact on behavior than does exposure to interparental violence only. Moreover, participants, who are both abused and exposed to interparental violence, exhibit internalized and external- ized symptoms falling within the clinical range more frequently. Exposure exclusively to interparental violence also has definite impact; for example, teens who are only exposed to interparental violence exhibit internalized and externalized symptoms more frequently than do those who have been neither subject to physical abuse from a parent nor exposed to interparental violence. Implications of the findings are discussed.

Keywords Adolescents . Behavior . Co-occurrence of interparental violence and child physical abuse

Studies have indicated that the incidence of exposure to interparental violence is far from marginal. It is estimated that every year in the United States, between 3.3 million (Carlson 1984) and 10 million (Peled et al. 1995) children are at risk of exposure to interparental violence. Between 1994 and 1999, approximately half a million Canadian children saw or heard violence directed toward a parent (Canadian Center for Justice Statistics 2000). According to Sudermann and Jaffe (1999), between 11 and 23% of all children in Canada witness such violence. These research- ers have calculated that in the average class in the school

setting, from two to six children have likely been exposed to interparental violence.

In addition to their exposure to interparental violence, many children are the direct targets of abuse; the average rate of co-occurrence of interparental violence and child abuse lies between 30 and 60% (Edleson 2001; McKay 1994;). A study conducted in Quebec indicated that, of all the reports made to the child protection services, 22.9% included the presence of interparental violence (Chamberland et al. 2007). Interparental violence constituted the primary form of abuse for 40% of these children, while negligence (56%) and physical abuse (32%) were the primary forms of abuse for the rest. Studies on the rate of co-occurrence have been associated with a number of methodological gaps (Appel and Holden 1998; Edleson 1999). For example, Appel and Holden (1998) noted that rates of co-occurrence vary depending on the sample: from 26 to 50% among children involved in reports to child protection agencies; from 4 to 100% among abused women; and from 6 to 21% for the general public.

Studies have revealed the many negative consequences of living in a violent domestic environment. Exposure to interparental violence is associated with the presence of externalized and internalized symptoms in children (Bourassa 1998; Chénard 1994; Christopoulos et al. 1987; Fantuzzo et al. 1991; Fortin et al. 2000; Jaffe et al. 1985, 1990) and adolescents (Bourassa 2002). However, accord- ing to the research available, the co-occurrence of interpar- ental and child abuse is associated with more severe symptoms among children (Davis and Carlson 1987; Fortin et al. 2000; Hughes 1988; Hughes et al. 1989) and teenagers (O’Keefe 1996).

Research is inconsistent regarding differences between boys and girls regarding the impact of exposure to interpar- ental violence on behavior. Some studies indicate that boys

J Fam Viol (2007) 22:691–701 DOI 10.1007/s10896-007-9117-8

C. Bourassa (*) School of Social Work, Université de Moncton, Moncton, New Brunswick, Canada E1A 1E9 e-mail: bourasc@umoncton.ca

show more externalizing symptoms (Jaffe et al. 1986) and that girls exhibit more internalizing symptoms (Christopoulos et al. 1987; Holden and Ritchie 1991; Stagg et al. 1989). However, several studies failed to find differences between boys and girls (Fantuzzo et al. 1991; Gleason 1995; O’Keefe 1994b). The only gender difference in the Fortin et al. (2000) study is that girls show more delinquent behaviors than boys. Therefore, according to those results, it is difficult to establish if girls and boys are affected differently by their exposure to interparental violence.

Moreover, research to date represents only a limited portrait of the consequences of the co-occurrence of interparental and child abuse. First, most studies have focused on young children; only limited knowledge is available with regard to teens, particularly those between the ages of 16 and 19. A study conducted with adolescents allows us to investigate the effects of domestic violence on a population that has probably been exposed for a longer period. In addition, while studies indicate that children who are both abused and exposed to interparental violence exhibit more severe problems than those who are only exposed to interparental violence, many studies have not checked for the presence of child abuse, particularly directed toward teens.

A few theories can help better understand the impact of interparental violence on the presence of internalized and externalized symptoms. According to social learning theory, the observation of others constitutes the central component of learning (Bandura 1973; Patterson et al. 1984). The more significant the model (e.g. one’s parents), the greater the chance that the observer will reproduce a given behavior. The chances that an observer will repro- duce a model’s behavior also increase if the model is a member of the same gender (Bandura 1969; Margolin and Patterson 1975). The observation of violent behaviors can lead a child to adopt attitudes that favor violence and, eventually, to imitate these violent behaviors (Anderson and Cramer-Benjamin 1999). Also, the observation of victim behaviors can lead children to imitate those behaviors. Moreover, if a child is subjected to child physical abuse in addition to exposure to aggressive models, then the probability of imitation of the violent or victim behaviors may increase (Kashani et al. 1991; O’Keefe 1996).

Another theoretical explanation is the psychiatric model who indicates that the adaptation problems of children and adolescents exposed to interparental violence can be explained by post-traumatic stress (Anderson and Cramer-Benjamin 1999; Margolin and John 1997). The psychiatric model suggest that the traumatic experiences (e.g., earthquake, war, interparental violence) provokes an intense fear and a helpless feeling and this could be manifested as internalizing problems (somatic difficulties, sleep problems, depression, anxiety) and externalizing problems (Anderson and Cramer-

Benjamin 1999). Children exposed to interparental violence live in a family environment in which they are scared for their protection and that of the parent who is victim.

Thus violence that occurs at the hands of one parent and that victimizes the other parent may create a world for children that is frightening, confusing, and lacking security and safety. Interparental violence transforms the home into a dangerous and unpredictable environ- ment. As concluded by Jafnoff-Bulmann (1993), «The most devastating negative life events on children are likely to be those that involve victimization by the very people who are looked to for protection and safety» (Margolin and John 1997, p. 91).

According to Rutter’s (1981) cumulative risk hypothe- ses, the risk of children experiencing problems doubles or quadruples with the presence of two or more stressors. Therefore, being exposed to domestic violence with the stress of being physically abused interact to contribute to an increase in externalized and internalized symptoms.

The results presented in this article are part of a study which seeks to extend our knowledge about the impact of interparental violence on teens aged between ages 15 and 19, a group age neglected in most studies on exposure to interparental violence. It was predicted that teens who were both victim of child abuse and exposed to interparental violence would show more behavioral problems (internalized and externalized symptoms) than those who are only exposed to interparental violence. It was also predicted that those exposed to interparental violence would exhibit more behav- ioral problems than those not exposed to that violence or not victim of child abuse. Because of the inconsistencies regarding the effect of interparental violence by gender, it is difficult at this point to elaborate an hypothesis on that aspect.

Materials and Methods

Participants and Procedure

The population studied was composed of teens attending a secondary school in New Brunswick, Canada. Three school districts were selected based on two criteria: representation of three geographic areas in the province and accessibility to the researcher. One secondary school in each district was subsequently identified, and a total of 984 French teens aged 16 to 18 were invited to take part in the project. Of this number, 296 (30.1%) opted not to respond to the questionnaire. Among the completed questionnaires, 198 (20.1%) were rejected due to an excessive number of missing responses. A total of 490 teens, aged between 15 and 19 years old (M=16.9), formed the final sample (49.8% of original respondents): 269 girls and 221 boys.

692 J Fam Viol (2007) 22:691–701

Overall, 78.4% of the sample resided in two-parent families. Eighteen percent of youth lived in a single parent or a reconstituted family. Eleven percent of responds had no siblings, while the rest had two or more siblings.

The number of refusal and rejected questionnaires may affect the generalization and external validity of the study. While the response rate was somewhat low, it compares to the rates typically recorded in studies of this nature. For example, Dumas and Beauchesne (1993) obtained a re- sponse rate of 52.1% in a study of paid employment among secondary school students.

Many reasons can explain the number of refusal or rejected questionnaires. Some may live severe violence at home and not want to think about it during class. Others may not live violence at home and, therefore, not feel concerned by the research. Also, the questionnaires were ad- ministered during flu season and there are many absences. Finally, because the questionnaires were administered during a free period, some preferred to use that time to finalize a paper or study for a test.

A comparison was done between the respondents by, whom the questionnaire was completed and those whom the questionnaire was rejected according to the gender, grade, number of children in the family and family type. The comparison is based on the 490 youths whom the questionnaire was retained and 143 of the 198 youths whom the questionnaire was rejected. The Chi-Square tests indicated that the respondents who had the questionnaire retained and those with the rejected questionnaires present a similar profile; there were no significant differences for gender (χ2=0.136, ns), grade (χ2=1.95, ns) and family type (χ2=1.74, ns). However, there was a significant difference with regards to the number of children in the family (χ2=10.81, p<0.05).

The questionnaires, which featured four rating scales, were distributed to the participants during regular school hours. At two of the schools, the questionnaires were distributed either by the researcher herself or by her research assistant. At the third school, the questionnaires were distributed by the teachers due to the high number of grade 11 and 12 classes at the school. A letter was forwarded to all parents in advance to advise them of the study and to encourage them to contact the researcher if they needed additional information. Participants were free to complete the questionnaire if they chose, and they completed a consent form. A list of resources available within the community was also sent out with the questionnaire.

Measures

The presence of internalized and externalized symptoms was assessed using the French versions of the internalized and externalized symptoms scales of the Youth Self Report

(YSR) (Achenback 1991): Rapport personnel des jeunes. These scales enable measurement of the perception of children or teenagers (aged 11 to 18) in terms of their behavior and social skills (Achenback 1991). The internal- ized disorder scale covers the subscales of withdrawal, somatic complaints, and anxiety/depression. The external- ized disorder scale is composed of subscales to measure delinquent and aggressive behavior. Achenback (1991) prescribed cutoff points based on subject gender for determining whether problems were within the normal range or exceeded the clinical threshold. The psychometric qualities of the French version of the YSR have not been evaluated; the authors of the French version instead use the standards from the U.S. version, whose psychometric properties are viewed as excellent, particularly when applied to teens between the ages of 15 and 18 (Achenback 1991).

Interparental violence was assessed using the French versions of the psychological and physical aggression scales from the Relationships Between My Parents questionnaire (Straus 2000, CTS2-CA). The CTS2-CA serves in assessing a person’s perception of conflicts between his or her parents. The CTS2-CA is a version of the Revised Conflict Tactics Scale, couple version (Straus et al. 1996; CTS2). Therefore, subsequent to the authors’ approval of the original tool, the CTS2-CA was translated based on the French version of the CTS2 (Lussier 1997, Questionnaire sur la résolution des conflits conjugaux [Resolution of partners’ conflicts Quec- tionnaire], unpublished data). In the CTS2-CA and the CTS2, the psychological aggression scale contains 16 items and the physical aggression scale contains 24 items. The respondent answers on a Likert scale ranging from 0 (never) to 6 (more than 20 times). The psychometric properties of the CTS2-CA have not been evaluated; however, the CTS2 is known to have superior psychometric qualities (Lussier 1997, Questionnaire sur la résolution des conflits conjugaux [Resolution of partners’ conflicts Quectionnaire], unpub- lished data; Straus et al. 1996).

The French version of the physical aggression subscale of the Parent–Child Conflict Tactics Scales, children version (Straus 2000, CTSPC-CA) was used to evaluate the physical violence to which respondents had been exposed. The CTSPC-CA enabled assessment of the persons’ perception of the physical violence to which they had been subjected by their parents. The CTSPC-CA is a version of the Revised Parent–Child Conflict Tactics Scales, parent version, which serve in the assessment of the disciplinary practices of parents. Again, subsequent to the authors’ approval of the original tool, the CTSPC-CA was translated based on the French version of the Revised Parent–Child Conflict Tactics Scales, parent version (Fortin et al. 1996). The respondent answers on a Likert scale ranging from 0 (never) to 6 (more than 20 times). Data are not currently available on the psychometric properties of

J Fam Viol (2007) 22:691–701 693

the CTSPC-CA. However, in the study carried out by Straus et al. (1998), the internal consistency coefficient for the physical aggression subscale of the PCCTS was 0.55. The PCCTS also appears to have a high degree of construct validity (ibid. 1998).

Finally, a number of questions were used to collect information on the sociodemographic characteristics of the participants, including age, gender, number of children in the family, family type and parents’ educational level. However, the educational level of parents was not taken into account in this study since a high number of respondents did not provide this information.

Results

Respondent Profile Based on Independent Study Variables: Interparental Violence and Child Physical Abuse

Of the 490 adolescents comprising the final sample, 253 (51.6%) had been exposed at least once to interparental violence within the past five years, while 237 (48.4%) had not been exposed to this type of violence. Among the

respondents who had been exposed to violence, 101 (20.6%) had been exposed to both psychological and physical violence, while 152 (31%) had been exposed only to psychological violence. Girls reported more frequent exposure to interparental violence than boys. T tests with Bonferonni correction (p<0.007) revealed that this differ- ence was significant with regard to physical violence perpetrated by the mother [t (488)=1.91, p<0.007; see Table 1], psychological violence perpetrated by the mother [t (488)=1.95, p<0.007], psychological violence perpetrat- ed by the father [t (488)=2.51, p<0.007], total psycholog- ical violence [t (488)=2.32; p<.007], total physical violence [t (488)=1.24, p<0.007). However, there was no significant difference between boys and girls for physical violence perpetrated by the father [t (488)=0.49, ns], total physical violence [t (488)=1.24, ns] and total interparental violence [t (488)=2.15, ns].

There was a positive and significant correlation between psychological violence perpetrated by the mother and that perpetrated by the father as well as between physical violence perpetrated by the mother and that perpetrated by the father (see Table 2). In addition, a positive and significant correlation is noted between total psychological violence and

Table 1 Interparental violence by the aggressor’s gender and youth’s gender: mean and standard deviations

Variables Girls (n=269) Boys (n=221)

M SD M SD

Mother to father violence Psychological violence 15.30 27.55 10.76* 23.93 Physical violence 4.46 17.68 2.06* 9.62

Father to mother violence Psychological violence 14.48 26.05 8.91* 23.11 Physical violence 3.51 15.31 2.84 14.79

Violence by both parents Psychological violence 29.77 50.79 19.67* 45.50 Physical violence 7.97 31.62 4.90* 23.14 Total interparental violence (psychological + physical) 37.75 73.34 24.57* 62.59

* p<0.05

Table 2 Correlations among different forms of interparental violence

Interparental violence 1 2 3 4 5 6 7

Psychological violence Mother to father – 0.83** 0.96** 0.54** 0.51** 0.55** 0.90** Father to mother – 0.95** 0.47** 0.59** 0.56** 0.90** Violence by both parents – 0.53** 0.57** 0.58** 0.94** Physical violence Mother to father – 0.79** 0.95** 0.76** Father to mother – 0.95** 0.79** Violence by both parents – 0.82** Interparental violence in total –

** p<0.01

694 J Fam Viol (2007) 22:691–701

total physical violence. Thus, the more the youth is exposed to psychological violence and physical violence perpetrated by the mother, the more likely he is exposed to psycholog- ical violence and physical violence perpetrated by the father. Given the elevated correlations among the variables, the subsequent analyses will take only total interparental violence into account. Meanwhile, although violence perpe- trated by female and male parents has been grouped, the researcher does not claim that these acts of aggression have the same goals or identical consequences.

Of the 490 respondents in the final sample, 32 (6.5%), that is, 18 girls and 14 boys had been victims of child physical abuse but had not been exposed to interparental violence, 131 (26.7%), including, 79 girls and 52 boys had been exposed to interparental violence but had not been victims of child physical abuse and 122 (24.9%) more specifically, 74 girls and 48 boys had been both abused and exposed to interparental violence. According to the results of the T tests with Bonferonni correction (p<0.01), there was no significant difference between boys and girls (see Table 3) for violence perpetrated by the mother [t (488)=1.53, ns], violence perpetrated by the father [t (488)=0.81, ns] and violence perpetrated by both parents [t (488)=1.17, ns].

Table 4 illustrates the correlations between physical violence perpetrated by the father and that perpetrated by the mother as well as the correlations between child physical abuse and interparental violence. First, with regard to child physical abuse, violence perpetrated by the mother and violence perpetrated by the father are closely associat- ed. The more the youth is victim of violence by the mother, the more likely he is victim of violence by the father. On these grounds, subsequent analyses take only total child physical abuse into account. Next, a positive and significant

correlation is noted between total interparental violence and total child physical abuse; as such, the greater the participant’s exposure to interparental violence, the more likely it is that the participant was also subjected to child physical abuse.