Family Violence and Aggression and Their Associations With Psychosocial Functioning in Jamaican Adolescents

Abstract

The purpose of the current study was to examine the relationships among selected family interaction variables and psychosocial outcomes in a sample of Jamaican adolescents. The authors hypothesized that adolescent psychosocial outcomes would be negatively associated with physical violence, verbal aggression would be more potent than physical violence, and the combined effect of all aggression and violence would be more detrimental than either form of aggression by itself. Overall, the results supported the authors’ hypotheses about the detrimental effects of negative family interactions on children’s well-being. All the parent–child and interadult variables were positively associated with problematic adolescents’ adjustment. However, the verbal aggression variables indicated a greater adverse effect than the physical violence variables. Additionally, the combined effect of parent–child and interadult aggression and violence indicated greater detriment to adjustment than each factor by itself. Interpretation of the findings within the Jamaican context and direction for future research are discussed.

Keywords

adolescents corporal punishment verbal aggression violence psychosocial adjustment

Prevalence data have indicated that children are likely to be exposed to multiple forms of family violence, including parental violence against children and interadult violence, both of which are commonplace across the globe (Chan, Brownridge, Yan, Fong, & Tiwari, 2011; Foshee, Bauman, & Linder, 1999; Shen, 2009; UNICEF, 2009). In light of these data, there is a growing awareness among child and family scientists of the increased potentially harmful effects of these multiple stressors to children’s development (Chan et al., 2011; Margolin & Vickerman, 2007; Osofsky, 2003). For example, data from the United States have shown that children who witness familial interadult violence are invariably also victims of parent to child violence (e.g., hitting, shoving; Edleson, 1999b; Osofsky, 2003). Moreover, victims of physical violence (e.g., hitting) are likely to simultaneously experience verbal aggression (e.g., insults; Straus & Donnelly, 2009; Vissing, Straus, Gelles, & Hartup, 1991). Edleson (1999b) reviewed 35 pertinent research studies and estimated an overlap of interadult violence and child maltreatment in 40% of violent families. Straus, Gelles, and Smith (1990) reported that 50% of the men who battered their partners also frequently abused their children. Children from homes where domestic violence occurs are 15 times more likely than the national average to be abused (Osofsky, 2003) and between 70% and 85% of children who live in violent families witnessed interadult assaults (Jaffe, Wolfe, & Wilson, 1990). An estimated 15.5 million U.S. children reside in families where interparental violence is prevalent (McDonald, Jouriles, Briggs-Gowan, Rosenfield, & Carter, 2007) and 90% of parents use physical punishment as a disciplinary method (Straus & Donnelly, 2009). Moreover, approximately 67% of children experience verbal aggression by their parents; children who are physically punished are likely also to experience verbal aggression (Vissing et al., 1991).

Despite these troubling statistics, there is little systematic research regarding the combined effects of multiple forms of violence and aggression on child outcomes because simultaneous examination of multiple forms of family violence and aggression has been largely neglected by researchers. Furthermore, most of what is known about family processes and their effect on child outcomes emanates from research done in industrialized societies such as the United States. Therefore, it is unclear whether findings from that body of research are applicable to other cultural contexts. The purpose of the current study is to examine the relationships between selected family violence variables and Jamaican adolescents’ socioemotional well-being. The following variables are of special interest to this study: parent to child verbal aggression, parent to child physical violence, interadult verbal aggression, and interadult physical violence.

Definition of Key Terms

The difficulty of defining family violence and aggression has been well documented. Of all the family violence variables, physical violence against children is the most controversial and difficult to define because there is no agreed on definition in the research literature (Edleson, 1999a; Gershoff, 2002). For example, the terms spanking, physical punishment, physical abuse, corporal punishment, and maltreatment are often used interchangeably in the family literature as representing physical violence. However, physical violence, which includes corporal punishment, runs along a continuum of caregiver practices, from spanking to physical abuse. Hence, a major criticism of physical violence against children research is that nonabusive physical punishment (e.g., mild spanking) is often confounded with abusive practices (e.g., beatings) (Gershoff, 2002). For this study, we adopt Straus and Donnelly’s (2009) definition of corporal punishment to conceptualize physical violence (hereafter referred to as corporal punishment): “The use of physical force with the intention of causing a child to experience pain but not injury for the purposes of correction or control of the child’s behavior” (p. 7). Interadult violence is much less controversial since much of the family research accepts Straus and Donnelly’s (2009) definition: “An act carried out with the intention, or perceived intention of causing physical pain or injury to another person” (p. 7). For this study, we use Straus and Donnelly’s perspective but limit the construct to violence between (or among) adults within the home. For this study, verbal aggression is “communication intended to cause psychological pain to another person, or a communication perceived to have that intent” (Vissing et al., 1991, p. 224).

Family Violence Research

Research has established a robust association between interadult family violence and elevated levels of maladjustment in minors (Osofsky, 2003; Zerk, Martin, & Proeve, 2009). Comparable adverse outcomes have also been reported for verbal aggression on children by their caregivers (Teicher, Samson, Polcari, & Greenery, 2006; Vissing et al., 1991), and although contentious, detrimental consequences have been reported for corporal punishment on children (Gershoff, 2002; Straus & Donnelly, 2009). However, the findings from the sparse research on the combined effects of multiple forms of violence on child outcomes are unclear. Some studies (e.g., Edleson, 1999a) have reported greater psychosocial injury for the joint exposure of being abused and witnessing familial violence than for witnessing alone. On the contrary, researchers such as Dawud-Noursi, Lamb, and Sternberg (1998), studying Israeli children, and Kitzmann, Gaylord, Holt, and Kenny (2003), using meta-analytic methodology, observed comparable levels of dysfunctional adjustment patterns among three groups of children who had been exposed to family violence: those who had witnessed the violence, those who had been victims of the violence, and those who had been both witnesses to and victims of the violence. Sternberg, Baradaran, Abbott, Lamb, and Guterman’s (2006) meta-analysis of relevant studies showed that child witnesses of spousal abuse did not differ significantly from children who were themselves abused. However, children who experienced multiple forms of family violence had worse outcomes than those who experienced only one form. More recently, Shen (2009) and Chan et al. (2011) observed that childhood exposure to both interparental violence and physical maltreatment was associated with detrimental emotionality in Taiwanese college students and Chinese adolescents, respectively.

Consequences of Verbal Aggression

Because verbal aggression is often subsumed under the general rubric of constructs such as psychological abuse, emotional abuse, and emotional maltreatment (Vissing et al., 1991), very little is known about its effect on child outcomes. However, the limited literature indicated that the insidious nature of verbal aggression makes it more detrimental to child outcomes than other forms of aggressive behaviors (Teicher et al., 2006). Johnson et al. (2001) found that child victims of maternal verbal abuse were more than three times likely as their nonvictimized peers to exhibit symptoms of personality disorders in adolescence and adulthood. Teicher et al. (2006) reported that the detrimental effects of parental verbal aggression were “comparable to those associated with witnessing domestic violence or nonfamilial sexual abuse and larger than those associated with familial physical abuse” (p. 993). Additionally, the cumulative effect of being the victim of parental verbal abuse and witnessing domestic violence had a greater adverse effect on children than either dynamic by itself (Teicher et al., 2006). Similarly, an earlier study (Vissing et al., 1991) found that children subjected to frequent parental verbal aggression exhibited high rates of adjustment problems but the combined effect of verbal aggression and physical abuse predicted greater problematic adjustment than either variable by itself. Undoubtedly, more research is needed to help decipher the source of these conflicting findings. Whether the findings reported here will be supported in the Jamaican context is unknown.

Familial Interactions and Child Outcomes

Human ecological systems theory has been identified as one of the preeminent perspectives in assessing and understanding human development (Reiss & Roth, 1993). The framework emphasizes the saliency of the interaction between the individual and the social environment and underscores the inseparableness of the two in effecting ontogenetic outcomes (Bronfenbrenner & Ceci, 1994). The perspective maintains that developmental outcomes emanate from a complex web of exchanges between the individual and the environment. These exchanges take place within a system of proximal (i.e., close social relationships) and distal contexts (e.g., economic climate, mass media) to either promote or stifle positive developmental outcomes. However, although all systems work together to influence what the individual becomes, the family context, being the most fundamental unit of society, exerts the most direct and profound impact on the individual (Bronfenbrenner & Ceci, 1994).

Family research using the human ecological perspective to examine the impact of the family on child outcomes has identified several intrafamilial risk factors including ineffective parenting (Gershoff, 2002, Margolin & Vickerman, 2007, 2011; Smith, Springer, & Barrett, 2011) and angry interadult conflict (Kitzmann et al., 2003; Margolin & Vickerman, 2007, 2011; Osofsky, 2003; Zerk et al., 2009) as key determinants of child and adolescent well-being. For example, voluminous data have suggested that children and youth reared in homes where harsh disciplinary practices (e.g., corporal punishment) predominate show increased propensity to a wide-ranging constellation of adjustment problems such as poor parent–child relationship, impaired cognition, low self-esteem, depression, anxiety, and antisocial behaviors (Afifi, Brownridge, Cox, & Sareen, 2006; Gershoff, 2002; Straus & Donnelly, 2009). For instance, Straus and Donnelly (2009) reported that adolescents who had experienced physical punishment had 23% more depressive symptoms and exhibited a greater propensity to suicidal ideation than their peers who were not physically punished. Similarly, a robust association between interadult family violence and a variety of adjustment problems including various forms of internalizing and externalizing symptomatology has been reported (Kitzmann et al., 2003; Osofsky, 2003; Zerk et al., 2009). According to Margolin and Vickerman (2007, 2011), intrafamilial violence can be conceived as a trauma-eliciting stimulus that exacerbates stress and potentiates the risk of maladaptive functioning on all dimensions of well-being: cognitive, emotional, behavioral, and psychobiological. In this study, we use the human ecological model to guide our understanding of the role of selected family processes on the functioning of Jamaican adolescents. Examination of distal processes is beyond the scope of this study.

A second important paradigm comes from the child risk and resiliency literature. This model highlights the potency of multiple adversities in shaping children’s realities and developmental processes (Masten, 1997). Familial violence is a stressful and traumatizing experience for children, whether they are merely witnesses or victims (Margolin & Vickerman, 2007, 2011; Osofsky, 2003), but the trauma intensifies when children are both witnesses and victims of that experience (Margolin & Vickerman, 2011). Indeed, research has shown a strong negative association between cumulative stress and psychosocial adjustment in children and families (Masten, 1997; Osofsky, 2003; Vissing et al., 1991). For example, child outcomes worsen and resiliency decreases as chronic adversity and risks multiply (Masten, 1997). Because family violence, more often than not, is chronic (Edleson, 1999a, 1999b; Margolin & Vickerman, 2011; Straus & Donnelly, 2009), it is plausible that exposure to familial verbal aggression and physical violence, a phenomenon referred to as the “double whammy” effect (Edleson, 1999a, 1999b), may stifle positive socioemotional well-being (Foshee et al., 1999; Gershoff, 2002; Margolin & Vickerman, 2007).

The Jamaican Context

Within the last decade, Jamaica, a small island nation (4,411 square miles) of 2.8 million people, has experienced a serious upsurge in violence, perpetrated mainly by youth aged 16 to 25 years (Smith & Green, 2007; United Nations, 2006). Violence is the leading cause of death among young males in that age cohort and the fifth leading cause of death among all age groups (Smith & Green, 2007). Undoubtedly, this worrying escalation of maladaptive behaviors (e.g., aggression and violent behavior) has signaled an urgent need to examine the impact of various social contexts on children’s developmental outcomes (Smith & Green, 2007; Smith et al., 2011), with an eye to identifying potential sources of problematic child and adolescent adjustment. This kind of inquiry is vital to the Caribbean setting where there is a critical paucity of research regarding the associations between familial processes and child outcomes (Smith et al., 2011). For example, the notion that many Jamaican homes lack the appropriate parent–child interactions for optimal child development has been well documented (Brown & Johnson, 2008; Smith et al., 2011) and a high incidence of harsh punishment (Brown & Johnson, 2008; Samms-Vaughan, 2008; UNICEF, 2009), parent to child verbal aggression (Brown & Johnson, 2008; UNICEF, 2009), and intimate partner violence (United Nations, 2006) has been confirmed. However, little is known about the consequential impact of those dynamics on children’s functioning. A contributing factor to the dearth of pertinent research may be based in the prevailing historic view that family violence is a private matter and an accepted cultural convention (United Nations, 2006; UNICEF, 2009). We found only one study that specifically examined the effect of parent–child physical aggression on child outcomes, but none on the effect of interadult aggression. Such studies are urgently needed especially in light of reports that, globally, the Latin America and the Caribbean region has the highest rate of violence affecting women and children (UNICEF, 2009). The current study will help fill this obvious gap in the family and child development literature.

Purpose of the Study

The purpose of the current study is to investigate the relationships among family violence, verbal aggression, and child psychosocial outcomes in the Jamaican context. In Jamaica, like many cultures across the globe, family violence is reportedly a socially accepted norm at all levels of society (United Nations, 2006, UNICEF, 2009). Consequently, not only are Jamaican children and adolescents targets of the family violence but also are witnesses to most of that violence (Samms-Vaughan, 2008). Empirical data from industrialized societies have shown positive associations between children’s exposure to family violence and an elevated propensity to adjustment problems (McDonald et al., 2007; Zerk et al., 2009). However, some researchers have suggested that the severity of the outcome may be a function of the form of violence (Edleson, 1999a; Kitzmann et al., 2003) and whether children are the targets or indirect victims of the violence (Teicher et al., 2006; Vissing et al., 1991).

We designed the study to investigate the relationships of corporal punishment, parental verbal aggression, interadult physical violence, and interadult verbal aggression on adolescents’ psychosocial status. In accordance with human ecological theory, cumulative adversity framework, and existing research, we hypothesized that (1) increased corporal punishment would be significantly related to poorer adolescent outcomes (Bronfenbrenner & Ceci, 1994), (2) parent–child verbal aggression would have a greater adverse affect on adolescent outcomes than would corporal punishment (Vissing et al., 1991), (3) increases in interadult verbal aggression and physical violence would be significantly related to poorer adolescent outcomes (Bronfenbrenner & Ceci, 1994), (4) interadult verbal aggression would have a greater adverse effect on adolescent psychosocial adjustment than interadult physical violence (Teicher et al., 2006; Vissing et al., 1991), and (5) the combined effect of parent–child and interadult variables would have a larger negative effect on adolescent outcomes than either alone (Margolin & Vickerman, 2011; Masten, 1997).

Method

Participants

The sample was drawn from a larger convenience sample of adolescents who participated in the Jamaican Adolescent Project. Participants in the current study were 171 males (54.2%) and females (45.8%) attending a secondary school in the metropolitan area of the capital city, Kingston. Participants ranged in age from 12.25 to 18.83 years (M = 15.14, SD = 1.40) and reported their parents’ marital status as single (34.1%), married (43.9%), divorced (9.1%), and living together but not legally married (12.8%). The household size ranged from two to five (M = 1.62, SD = 0.82). This sample was selected because it constituted the largest single homogenous grouping of the larger sample; it was chosen to decrease the likelihood of the confounding effect of socioeconomic status.

Procedure

Permission and approval to conduct the study were given by the local Ministry of Education and the administration, acting in loco parentis, of the schools that participated in the study. The study was approved by the institutional review board of the authors’ home institution. Participants completed the self-administered surveys in group settings in space designated by the school administration. Prior to data collection, consent was obtained from individual students after they had been informed that their participation was voluntary and that they could discontinue their involvement at any time. Participants were informed of the researchers’ efforts to ensure anonymity of their responses. They were instructed not to place any identifying information on the survey. Additionally, every attempt was made to provide ample spacing between respondents and their peers to keep their responses as private as possible. Respondents were asked not to discuss or share their responses with anyone. On return of the survey, students were presented with a token packet, which included a pen and pencil.

Measures

Demographics

Participants indicated their gender, age, family size, and parental marital status. Age was measured in years, and family size was assessed by the number of individuals residing in the household. Marital status was recorded as single (never been married), married, divorced, or living together (not legally married).

Parent–child verbal aggression and corporal punishment

Ten items representing aggressive and violent parent–child interactions were drawn from the empirical literature. Adolescents indicated on a 4-point Likert-type scale (0 = never, 1 = rarely, 2 = often, 3 = always) the type and frequency of verbal and physical interactions to which they are currently or had been (in the past) subjected to at home. Five items represented verbal aggression (e.g., shaming, threatening to harm) and five items represented corporal punishment (e.g., hitting, punching). The items on each category were totaled to assess each dimension. Cronbach αs for the sample were .78 for verbal aggression and .82 for corporal punishment.

Interadult verbal aggression and physical violence

Six items representing verbal aggression and physical violence were drawn from the extant literature and used to assess interadult violence. Adolescents indicated on a 4-point Likert-type scale (0 = never, 1 = rarely, 2 = often, 3 = always) the degree to which adults in their families engaged in aggressive verbal or physically violent relational interactions. The items conceptually assessed verbal aggression (three items) and physical violence (three items). Participants indicated the type of hostile interaction adults in their families exhibited toward each other. The items on each category were totaled to assess each dimension: verbal aggression (e.g., yell and scream at each other in anger) and physical violence (e.g., fight a lot—hit each other). Cronbach αs for the sample were .84 and .90 for corporal punishment and verbal aggression, respectively.

Given that we are unaware of any studies that examined the factor structure of these measures of parent–child and interadult violence in a Jamaican sample, we conducted a confirmatory factor analysis (CFA) using LISREL 8.72 (Jöreskog & Sörbom, 2005) to verify that the model fit the data provided by this sample. Based on Kline’s (2005) recommendations, four fit indices were used to determine the extent to which the specified model fit the data: the minimum fit function (MFF) χ², which should demonstrate a ratio to degrees of freedom (df) of approximately 3 or less (Kline, 2005); the Bentler–Bonnet nonnormed fit index (NNFI; Bentler & Bonnet, 1980), which should be at least .90; the comparative fit index (CFI; Bentler, 1990), which also should be at least .90; and the standardized root mean square residual (SRMR), which should be less than .10 (Kline, 2005). The overall CFA for the factor weights and latent variable covariances showed a good fit of the model to the data: MFF χ²(265.27) to df(98) ratio = 2.71; NNFI = .93; CFI = .95; and SRMR = .071.

Psychosocial adjustment

Four scales from the Massachusetts Youth Screening Instrument: Version 2 (MAYSI-2) assessed internalized distress. The 52-item paper-and-pencil self-administered protocol was designed for identifying potential mental health problems in adolescents. The measure identifies adolescents’ thoughts and feelings experienced or behaviors exhibited in the past few months. Respondents circle “Yes” or “No” to indicate whether each item on the measure is true or not true of them. The MAYSI-2 is scored by summing the positive (“yes”) responses on each scale. The scales selected for this study were the (a) Alcohol/Drug (eight items), which identifies respondents’ frequency and degree of alcohol and drug use (e.g., “Have you used alcohol or drugs to help you feel better?”); (b) Angry-Irritable (nine items), which assesses feelings of preoccupying anger and vengefulness and addresses respondents’ propensity toward irritability, and tension related anger (e.g., “Have you felt angry a lot?”); (c) Depressed-Anxious (nine items), which evaluates symptoms of depression and anxiety (e.g., “Have you given up hope for your life?”); and (d) Suicide ideation (five items), which identifies specific thoughts and feelings that are symptomatic of suicidal intent and injury (e.g., “Have you felt that life was not worth living?”). Reported internal consistency (α coefficients) values among the scales range from .61 to .89 (average = .75); test–retest reliability ranged from .55 to .93. These coefficients are similar to those reported for scores on comprehensive psychological measures such as the well-established Child Behavior Checklist-YSR measure and the Millon Adolescent Clinical Inventory (Grisso & Barnum (2006). For this sample, Cronbach α for the five scales ranged from .74 to .81. Similar to the measures of violence above, we conducted a CFA to verify that the model fit the data provided by this sample for this measure. The overall CFA for the factor weights and latent variable covariances showed a good fit of the model to the data: MFF χ²(604.96) to df(369) ratio = 1.64; NNFI = .90; CFI = .91; and SRMR = .081.

Additionally, adolescents responded to seven items, drawn from the extant literature as conceptually representing behavior problems. The items were set on a 4-point Likert-type scale on which adolescents specified (0) never, (1) once, (2) a few times, and (3) to indicate the frequency of involvement in antisocial and delinquent activities (e.g., stolen things, carried a weapon for protection). Cronbach α for the sample was .87.

Results

Data Analytic Plan

First, descriptive analyses were conducted to examine sample characteristics. Second, Pearson correlations were calculated to examine associations between demographic variables and adolescent psychosocial outcomes to determine whether demographic variables would need to be controlled in subsequent analyses. Third, additional Pearson correlations were calculated to test Hypotheses 1 and 3 to determine the associations between violence and aggression variables and adolescent psychosocial outcomes. Finally, a series of regression analyses was conducted to examine the independent and combined effect of the aggression and violence variables on predicting adolescent psychosocial adjustment.

Approximately 62% (62.4) of the sample reported having received corporal punishment at home; 73% of males indicated that they had received corporal punishment compared with 63.1% of the females, whereas nearly 90% (88.5) of the sample (90.7% of males and 85.7% of females) indicated that they had been victims of familial verbal aggression. Regarding interadult violence, approximately 43.9% (40% of males and 49.3% of females) reported witnessing interadult physical violence. A total of 81.3% of the sample (77.8% of males and 85.5% of females) reported witnessing interadult verbal aggression. Chi-square analyses showed no significant gender differences in prevalence on any dimension of physical violence or verbal aggression. Furthermore, none of the demographic variables, except gender and age, showed significant relationships with any of the predictor variables or dependent variables. Specifically, gender was significantly related to interadult physical violence (r = .16, p = .045), age was related to alcohol/drug use (r = .20, p = .013) and behavior problems (r = .26, p = .001). However, adjusting for multiple comparisons, only the association between age and conduct problems remained significant. Therefore, age was entered as a control variable in all regression analyses involving conduct problems.

Hypothesis 1 indicated that increases in parent to child verbal aggression and corporal punishment and would be significantly related to poorer adolescent outcomes. As shown in Table 1, all, but one, correlations (i.e., corporal punishment and alcohol/drug use) were significant after employing a Bonferroni correction, indicating that corporal punishment and verbal aggression were significantly and positively correlated with poorer adolescent psychosocial outcomes after correcting for multiple comparisons.

Table 1.

Correlations Among Predictor and Outcome Variables.

Dimensions	Parent–Adol. Corporal	Parent–Adol. Verbal	Interadult Physical	Interadult Verbal
Alcohol/Drug use	.24**	.31***	.32***	.15*
Anger-Irritable	.36***	.45***	.35***	.43***
Depress-Anxious	.44***	.48***	.45***	.53***
Suicide Ideation	.39***	.43***	.42***	.43***
Behavior Adjustment	.43***	.42***	.31***	.32***
Parent–Adol. Corporal		.73***	.46***	.43***
Parent–Adol. Verbal	.73***		.45***	.51***
Interadult Physical	.46***	.45***		.66***
Interadult Verbal	.43***	.51***	.66***

Note: Parent–Adol. = parent–adolescent; Corporal = corporal punishment; Verbal = verbal aggression.

p < .05. **p < .01. ***p < .001. Using a Bonferroni correction with 32 correlations, p values less than .0015 are considered significant.

Hypothesis 2 stated that child parent–child verbal aggression would have a greater adverse affect on adolescent outcomes than would corporal punishment. We conducted a series of regression analyses where corporal punishment and parent–child verbal aggression were entered together to compete for variance in predicting each of the adolescent outcomes. Consistent with the stated hypothesis, results showed that after controlling for corporal punishment, parent–child verbal aggression significantly predicted all five psychosocial outcomes (ts from 2.80 to 3.83, ps < .001). In addition, after controlling for parent–child verbal aggression, corporal punishment was not significantly related to any of the psychosocial outcomes (ts from .08 to 1.95, ns). See Table 2.

Table 2.

Regression Analyses for Parent–Child Verbal Aggression and Corporal Violence on Adolescent Outcomes.

Variable	Β	SE Β	β	R ²
Alcohol Use regressed on
Parent–Child Verbal Aggression	.05	.02	.32**
Parent–Child Corporal Punishment	.00	.02	.01	.10
Anger regressed on
Parent–Child Verbal Aggression	.33	.09	.40***
Parent–Child Corporal Punishment	.05	.09	.05	.20
Depression regressed on
Parent–Child Verbal Aggression	.26	.07	.36**
Parent–Child Corporal Punishment	.14	.08	.18	.26
Suicide regressed on
Parent–Child Verbal Aggression	.19	.06	.35**
Parent–Child Corporal Punishment	.07	.06	.13	.20
Conduct Problems regressed on
Parent–Child Verbal Aggression	.07	.02	.31**
Parent–Child Corporal Punishment	.04	.02	.20	.23

p < .01. ***p < .001. Using a Bonferroni correction for five regressions, p values less than.01 are considered significant.

Hypothesis 3 indicated that increases in interadult physical violence and child verbal aggression would be significantly related to poorer adolescent outcomes. As can be seen in Table 1, all but one correlation (i.e., interadult verbal and alcohol/drug use) were significant. Specifically, interadult verbal aggression and physical violence were significantly and positively correlated with poorer adolescent psychosocial outcomes, except that interadult verbal aggression was not significantly associated with adolescent alcohol use.

Hypothesis 4 proposed that interadult verbal aggression would have a greater adverse effect on adolescent outcomes than interparental physical violence. We conducted a series of regression analyses where interadult verbal aggression and interadult physical violence were entered together to compete for variance in predicting each of the adolescent outcomes. Consistent with the stated hypothesis, after controlling for interadult physical violence, interadult verbal aggression significantly predicted greater anger, depression, and suicidal ideation (ts from 2.98 to 4.75, ps < .01). In addition, after controlling for interadult verbal aggression, interadult physical violence was not related to these three psychosocial outcomes (ts from 1.16 to 2.53, ps > .01, not significant after correcting for multiple comparisons). Neither interadult verbal aggression nor physical violence uniquely predicted conduct problems. Finally, and contrary to the stated hypothesis, interadult physical violence predicted alcohol/drug use after controlling for interadult verbal aggression (t = 3.91, p < .001), whereas interadult verbal aggression was unrelated to alcohol/drug use (t = −1.11, ns). See Table 3.

Table 3.

Regression Analyses for Interadult Verbal Aggression and Physical Violence on Adolescent Outcomes.

Variable	Β	SE Β	β	R ²
Alcohol Use regressed on
Interadult Verbal Aggression	−.02	.02	−.11
Interadult Physical Violence	.10	.03	.39	.11
Anger regressed on
Interadult Verbal Aggression	.32	.08	.36***
Interadult Physical Violence	.14	.12	.11	.19
Depression regressed on
Interadult Verbal Aggression	.33	.07	.41**	.30
Interadult Physical Violence	.22	.10	.19*
Suicide regressed on
Interadult Verbal Aggression	.16	.06	.27**
Interadult Physical Violence	.20	.08	.23	.22
Conduct Problems regressed on
Interadult Verbal Aggression	.04	.02	.19
Interadult Physical Violence	.05	.03	.15	.16

p < .05. **p < .01. ***p < .001. Using a Bonferroni correction for five regressions, p values less than .01 are considered significant.

Hypothesis 5 stated that the combined effect of parent–child and interadult variables would have a larger negative effect on adolescent psychosocial outcomes than either variable alone. We conducted a series of hierarchical regression analyses by entering interadult variables in Step 1 predicting each adolescent outcomes followed by entering parent–child variables in Step 2 to determine whether adding these variables accounted for an additional significant amount of variance in predicting adolescent outcomes. Consistent with the stated hypothesis, results showed that adding the parent–child variables accounted for a significant amount of additional variance in all but one of the psychosocial outcomes. After correcting for multiple comparisons, the effect of adding the parent–child variables accounted for 4.5% additional variance in predicting alcohol/drug use (F = 4.03, p = .02). Results are shown in Table 4.

Table 4.

Regression Analyses for Impact of Interadult and Parent–Child Corporal Punishment.

Variable	Β	SE Β	β	R² change
Alcohol Use regressed on
Block 1
Interadult Verbal Aggression	−.01	.02	.06
Interadult Physical Violence	.09	.03	.35**	10***
Block 2
Parent–Child Verbal Aggression	.05	.02	.29*
Parent–Child Corporal Punishment	−.01	.02	−.05	.05*
Anger regressed on
Block 1
Interadult Verbal Aggression	.32	.09	.35**
Interadult Physical Violence	.14	.12	.11	.19***
Block 2
Parent–Child Verbal Aggression	.22	.09	.27*
Parent–Child Corporal Punishment	.02	.09	.02	.05**
Depression regressed on
Block 1
Interadult Verbal Aggression	.31	.07	39***
Interadult Physical Violence	.24	.11	.20*	.30***
Block 2
Parent–Child Verbal Aggression	.14	.08	.18
Parent–Child Physical Violence	.11	.08	.14	.06**
Suicide regressed on
Block 1
Interadult Verbal Aggression	.16	.06	.27**
Interadult Physical Violence	.19	.08	.23*	.21***
Block 2
Parent–Child Verbal Aggression	.12	.06	.22*
Parent–Child Corporal Punishment	.04	.06	.07	.05**
Conduct Problems regressed on
Block 1
Interadult Verbal Aggression	.04	.02	.19
Interadult Physical Violence	.06	.03	.19	.12***
Block 2
Parent–Child Verbal Aggression	.06	.02	.26*
Parent–Child Corporal Punishment	.04	.02	.18	.11***

p < .05. **p < .01. ***p < .001. Using a Bonferroni correction for five regressions, p values less than .01 are considered significant.

In sum, higher levels of corporal punishment and verbal aggression and interadult violence and verbal aggression were independently associated with adolescents’ increased propensity to all the psychosocial factors assessed. However, when taken together, parent–child verbal aggression seemed to exact the greatest negative toll on adolescent psychosocial functioning. Specifically, whereas parent–child verbal aggression predicted adolescents’ increased susceptibility to both psychological and conduct problems, corporal punishment indicated no negative effects. Regarding interadult violence, whereas verbal aggression predicted psychological problems (i.e., anger, depression, suicide ideation) it did not predict alcohol/drug use or conduct problems. Although interadult violence was related to alcohol/drug use, it did not predict any of the other psychosocial problems. Finally, the combined negative behaviors (i.e., parent–child verbal and corporal punishment and interadult verbal and physical violence) were better predictors of negative adolescent outcomes than each one by itself, an indication that the more types of violence adolescents experienced the worse their outcomes. Moreover, being a witness to interadult negative interactions as well as being a victim of corporal punishment engendered greater adverse psychosocial functioning than being either a witness or a victim.

Discussion

Guided by the human ecological theory, stress pile up framework, and existing research, the study examined the relationships between family violence and psychosocial outcomes in a sample of Jamaican adolescents. The study is valuable in that it extends existing research on the impact of problematic family processes in the Jamaican context. Of particular importance is the information gleaned from examining the role of the unique and combined effects of witnessing and being a victim of family violence, common experience of Jamaican children, on adolescent functioning. In addition, the findings are significant because they not only support past research demonstrating the adverse effect of harsh family contexts on children’s development but, more specifically, provide well-needed information on the relevance of family behaviors and interactions to Jamaican children and adolescents’ adjustment.

The study results confirm our hypotheses that parent–child corporal punishment and interadult violence would be associated with Jamaican adolescents’ problematic adjustment and that parent–child and interadult verbal aggression would be more detrimental to adolescents’ psychosocial outcomes than physical violence. In addition, the hypothesis that the combined effect of all forms of aggression and violence would be more harmful to adolescent adjustment than each form by itself was supported. Overall, the findings are consistent with human ecological theoretical supposition regarding the saliency of the proximal social context on individual development (Bronfenbrenner & Ceci, 1994), particularly the potential detrimental effects of negative family interactions on child outcomes (Gershoff, 2002; Margolin & Vickerman, 2007; Sternberg et al., 2006; Vissing et al., 1991). The results also provide some degree of validation to the adversity pile up paradigm regarding the “double whammy” effect of multiple adverse events—in this case, verbal aggression and physical violence—on children’s developmental processes (Edleson, 1999a; Masten, 1997; Shen, 2009; Sternberg et al., 2006). Furthermore, the findings concur with existing research that verbal aggression might be more injurious to children and adolescent well-being than other types of aggression (Johnson et al., 2001; Teicher et al., 2006; Vissing et al., 1991).

The more potent effect of verbal aggression, compared with corporal punishment, regardless of whether adolescents were the victims of the verbal aggression or witnessed it, is quite instructive. According to Vissing et al. (1991), the common children’s rhyme “Sticks and stones may break my bones, but names will never hurt me” is illustrative of the view that verbal aggression is less injurious to the psyche than physical violence and, therefore, may lead caregivers to believe that caustic words hurt less or not at all. This line of reasoning is especially relevant to the Jamaican environment where much attention regarding child maladjustment is being focused on corporal punishment (Smith & Green, 2007; United Nations, 2006) at the expense of verbal aggression. Johnson et al. (2001) argued that verbal abuse (from parents) inflicts and leaves deeper scars than other types of abuse “because children tend to identify with their parents and therefore tend to internalize and self-inflict abusive statements against themselves throughout their lives” (p. 21) may be tenable in the Jamaican context. Our study did not differentiate between parents and nonparents or between family members and nonfamily members; adolescents merely reported the behaviors of parents/guardians. However, it is worthy of note that parents are not the only “significant others” in children’s lives. The extended family (e.g., grandparents, aunts, cousins, teachers) is a vital component of Jamaican families. Therefore, in that context, children could identify with and internalize messages from any adult with whom they are closely associated. Further research is warranted to delineate the possible differential effects verbal aggression by parents and others.

The finding of the cumulative effect of violence on adjustment outcomes warrants special mention. The finding supports research using the stress pile-up model, which purports that multiple adverse events lead to greater adjustment difficulties than a single event (e.g., Edleson, 1999a; Osofsky, 2003). Indeed, the child development literature suggests that any one of the adverse family events discussed here is by itself devastating to psychosocial outcomes, but the combined effect is worse (Vissing et al., 1991). We speculate that the effects may arise from two pathways. Physical violence, such as verbal aggression, can lead to low self-esteem, which in turn fosters feelings of incompetence, worthlessness, and a sense of deserving the punishment meted out (Chan et al., 2011; Shen, 2009). Relatedly, witnessing familial interadult verbal aggression and violence may foster, in children, low self-efficacy, arising from feelings of helplessness brought on by the perceived inability to protect the self (or other victims) from the perpetrator (Chan et al., 2011; Shen, 2009). These dynamics are especially distressing to children and adolescents when family is the source of the trauma. According to Margolin and Vickerman (2011, p. 65), “family violence has the unfortunate consequence of undermining parents [or guardians] as protectors and sources of support.” Furthermore, fear and feelings of helplessness and vulnerability may be particularly devastating to minors’ functioning “when the threatened or actual injury is caused by one parent and directed toward the child, the other parent, or a sibling” (p. 65). Our data do not allow for assessing those psychological constructs; however, diminished self-esteem and self-efficacy have been shown to engender serious negative psychosocial outcomes in children and youth (Chan et al., 2011; Shen, 2009).

Another pertinent issue to the overall findings is the potential role of individual characteristics. In ecological theory, individual characteristics play a salient function in dealing with the social environment (Bronfenbrenner & Ceci, 1994). Coping was not examined in the current study or in relevant past research. However, research studies have suggested that the coping technique individuals use to deal with stressful events is a robust predictor of psychosocial outcomes. For instance, the cognitive strategy of self-blame has featured prominently in the developmental literature as a common coping mechanism to deal with adversity at all life stages (i.e., childhood, adolescence, adulthood). In comparison with peers who employ alternative coping mechanisms, individuals who self-blame exhibit more maladaptive symptoms (Chan et al., 2011), experience a lower quality of life, and face a greater risk of revictimization than peers who employ alternative coping mechanisms (Klein, Turvey, & Pies, 2007). It is plausible that adolescents in the present study may blame themselves for the familial violence they experience. The phenomenon of attributing the negativity of one’s immediate social context to self-identified faults, misdeed, and negative traits has been well documented (Chan et al., 2011; Margolin & Vickerman, 2011).

In the current study, corporal punishment indicated no negative effect on adolescent outcomes after controlling for verbal aggression, a finding contrary to an earlier study (Smith et al., 2011) that reported physical punishment to be significantly associated with adverse psychological and behavioral consequences in a Jamaican sample. The conflictual findings might relate to methodological factors. The earlier study investigated a single factor, physical punishment, and made no attempt to delineate other types of family process variables; it is plausible that other factors might have confounded the results. In addition, the current study used a significantly smaller sample. Therefore, we must be careful not to underemphasize the possible negative effect of corporal punishment on Jamaican children’s development. Besides, the more potent effect of verbal aggression does not necessarily mean that adolescents were not significantly affected by the corporal punishment they experience. It could very well be that adolescents may be developing (or have developed) coping strategies and behavior patterns that allow them to suppress or delay overt psychosocial difficulties (Grych et al., 2002) emanating from the corporal punishment.

Limitations

Despite the significance and value of this study, its limitations must be acknowledged. First, using data consisting solely of adolescents’ reports may have introduced an element of unmeasured bias. Adolescents may not have accurately reported their experiences and their perceptions of treatment may have differed from other sources (e.g., parents). However, other studies (e.g., Brown & Johnson, 2008), using both parent and child reports regarding corporal punishment in the Jamaican setting, have shown relatively high levels of agreement. Moreover, adolescents’ perceptions of their experiences and emotional status are as important, if not more important, as those of their parents (Osofsky, 2003).

Second, the study did not assess other types of adverse conditions that might be occurring in children’s lives. For example, the high rate of community violence that supposedly engenders problematic psychosocial well-being in children and adolescents (Smith & Green, 2007) was not examined here. Therefore, we cannot ascertain the role of the distal contexts on Jamaican family interactions and, ultimately, on child outcomes.

Third, the cross-sectional nature of the data does not allow for causal interpretations of the findings; an association between variables does not mean that one variable caused the other. Nevertheless, given the limited research on family processes in the Jamaican context, the findings presented here can be meaningful to professional practice and provide a base from which future research can be launched. A direction for future research would be to employ longitudinal data to explore the possible pathways from negative family experiences to children and adolescents’ adjustment and well-being.

Finally, any attempt to interpret behavior must be done with an appreciation of the setting in which the behavior occurs. Hence, the generality of the findings is limited to the Jamaican context and specifically to the current sample, which was drawn from a larger convenience sample.

Directions for Future Research

This study warrants replication to increase confidence in the findings presented here; the reports of this selected group of adolescents may not be representative of the wider adolescent population. In light of the limitations presented here, it might be prudent for future research to explore possible intervening variables in the association between family processes and child outcomes in the Jamaican context. Past research, emanating from other societies, has shown that personal characteristics may be salient to ontogenetic outcomes. For example, perception of self-worth and efficacy, personal appraisals of stressful events, coping strategy, and relationships with extrafamilial others have been shown to promote resiliency in minors experiencing chronic adversity (Chan et al., 2011; Masten, 1997; Shen, 2009). Consequently, more research into those variables may yield meaningful information on how to improve outcomes in children who experience violence in their families. Furthermore, it is important to recognize that family violence takes place within a complex social context. For example, poverty, social isolation, substance abuse, community violence, and other social challenges are salient factors of the social milieu in which many families operate (Margolin & Vickerman, 2007). Therefore, those dynamics should be addressed in future studies in an attempt to tease out the apparent complex relationships among possible confounding factors. Also, it might be helpful for future research to add a complementary qualitative component. Having adolescents talk through their experiences, perceptions, and feelings about events occurring in their lives may be instructive and provide information not easily gleaned from quantitative inquiries.

Footnotes

Acknowledgements

The authors wish to thank Dr. Sheila Barrett for her help and support throughout the data collection process. We are also grateful to the administration and students of the schools from which the data were collected, for their participation in the study.

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 study was supported by grants from the Graduate School, the College of Education, Health, and Human Sciences, and the Department of Child and Family Studies at University of Tennessee, Knoxville.

References

Afifi

T. O.

Brownridge

D. N.

Cox

B. J.

Sareen

(2006). Physical punishment, child abuse and psychiatric disorders. Child Abuse & Neglect, 30, 1093-1103. doi:10.1016/j.chiabu.2006.04.006

Bentler

P. M.

(1990). Comparative fit indices in structural models. Psychological Bulletin, 107, 238-246.

Bentler

P. M.

Bonett

D. G.

(1980). Significance tests and goodness-of-fit in the analysis of covariance structures. Psychological Bulletin, 88, 588-600.

Brown

Johnson

(2008). Childrearing and child participation in Jamaican families. International Journal of Early Years Education, 16, 31-40. doi:10.1080/ 09669760801892110

Bronfenbrenner

Ceci

S. J.

(1994). Nature-nurture reconceptualized in developmental perspective: A bioecological model. Psychological Review, 101, 568-586.

Chan

K. L.

Brownridge

D. A.

Yan

Fong

D. Y.

Tiwari

(2011). Child maltreatment poly-victimization: Rates and short-term effects on adjustment in a representative Hong Kong sample. Psychology of Violence, 1, 4-15. doi:10.1037/a0020284

Dawud-Noursi

Lamb

M. E.

Sternberg

K. J.

(1998). The relations among domestic violence, peer relationships, and academic performance. In Feiring

Lewis

(Eds.), Families, risk and competence (pp. 207-230). Mahwah, NJ: Erlbaum.

Edleson

(1999a). Children’s witnessing of adult domestic violence. Journal of Interpersonal Violence, 14, 839-870. doi:10.1177/088626099014008004

Edleson

(1999b). The overlap between child maltreatment and woman battering. Violence Against Women, 5, 134-154. doi:10.1177/107780129952003

10.

Foshee

V. A.

Bauman

K. E.

Linder

G. F.

(1999). Family violence and the perpetration of adolescent dating violence: Examining social learning and social control processes. Journal of Marriage and the Family, 61, 331-342.

11.

Gershoff

E. T.

(2002). Corporal punishment by parents and associated child behaviors and experiences: A meta-analytic and theoretical review. Psychological Bulletin, 128, 539-579. doi:10.1037//0033-2909.128.4.539

12.

Grisso

Barnum

(2006). Massachusetts Youth Screening Instrument Version 2: Users’ manual and technical report. Sarasota, FL: Professional Resource Press.

13.

Grych

J. H.

(2002). Marital relationships and parenting. In Bornstein

M. H.

(Ed.), Handbook of parenting (Vol. 4, 2nd ed., pp. 203-226). Mahwah, MJ: Erlbaum.

14.

Jaffe

P. G.

Wolfe

D. A.

Wilson

S. K.

(1990). Children of battered women. Newbury Park, CA: Sage.

15.

Johnson

J. G.

Cohen

Smailes

E. M.

Skodol

A. E.

Brown

Oldham

J. M.

(2001). Childhood verbal abuse and risk for personality disorders during adolescence and early adulthood. Comprehensive Psychiatry, 42, 16-23. doi:10.1053/comp.2001.19755

16.

Jöreskog

K. G.

Sörbom

(2005). LISREL 8.72. Chicago, IL: Scientific Software.

17.

Kitzmann

K. M.

Gaylord

N. K.

Holt

A. R.

Kenny

E. D.

(2003). Child witnesses to domestic violence: A meta-analytic review. Journal of Consulting and Clinical Psychology, 71, 339-352. doi:10.1037/0022-006X.71.2.339

18.

Klein

Turvey

D. L.

Pies

C. J.

(2007). Relationship of coping styles with quality of life and depressive symptoms in older heart failure patients. Journal of Aging and Health, 19, 22-38. doi:10.1177/0898264306296398

19.

Kline

R. B.

(2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: Guilford Press.

20.

Margolin

Vickerman

K. A.

(2007). Posttraumatic stress in children and adolescents exposed to family violence: Overview and issues. Professional Psychology: Research and Practice, 38, 613-619. doi:10.1037/0735-7028.38.6.613

21.

Margolin

Vickerman

K. A.

(2011). Posttraumatic stress in children and adolescents exposed to family violence: Overview and issues. Couple and family Psychology: Research and Practice, 1(Suppl.), 63-73. doi:10.1037/2160-4096.1.S.63

22.

Masten

A. S.

(1997). Resilience in children at risk. Center for Applied Research and Educational Improvement (CAREI) Newsletter, 5(1). Retrieved from http://www.cehd.umn.edu/carei/reports/rpractice/Spring97/resilience.html

23.

McDonald

Jouriles

E. N.

Briggs-Gowan

M. J.

Rosenfield

Carter

A. S.

(2007). Violence toward a family member, angry adult conflict, and child adjustment difficulties: Relations in families with 1- to 3-year-old children. Journal of Family Psychology, 21, 176-184.

24.

Osofsky

(2003). Prevalence of children’s exposure to domestic violence and child maltreatment: Implications for prevention and intervention. Clinical Child and Family Psychology Review, 6, 161-170.

25.

Reiss

A. J.

Jr. Roth

J. A.

(1993). Understanding and preventing violence: Vol. 1. Panel on the Understanding and Control of Violent Behavior. Washington, DC: National Academies Press.

26.

Samms-Vaughan

(2008). Comprehensive longitudinal studies of child health, development and behaviour in Jamaica: Findings and policy impact. West Indian Medical Journal, 57, 639-644.

27.

Shen

A. C.

(2009). Self-esteem of young adults experiencing interparental violence and child physical maltreatment: Parental and peer relationships as mediators. Journal of Interpersonal Violence, 24, 770-779. doi:10.1177/0886260508317188

28.

Smith

D. E.

Green

(2007). Violence among youth in Jamaica: A growing public health risk and challenge. Pan American Journal of Public Health, 22, 417-424.

29.

Smith

D. E.

Springer

Barrett

(2011). Physical discipline and socioemotional adjustment among Jamaican adolescents. Journal of Family Violence, 26(1), 51-61. doi:10.1007/s10896-010-9341-5

30.

Sternberg

K. J.

Baradaran

L. P.

Abbott

C. B.

Lamb

M. E.

Guterman

(2006). Type of violence, age, and gender differences in the effects of family violence on children’s behavior problems: A mega-analysis. Developmental Review, 26, 89-112. doi:10.1016/j.dr.2005.12.001

31.

Straus

M. A.

Donnelly

D. A.

(2009). Beating the devil out of them: Corporal punishment in American families and its effects on children. New Brunswick, NJ: Transaction.

32.

Straus

M. A.

Gelles

R. J.

Smith

(1990). Physical violence in American families: Risk factors and adaptations to violence in 8,145 families. Brunswick, NJ: Transaction.

33.

Teicher

M. H.

Samson

J. A.

Polcari

McGreenery

C. E.

(2006). Sticks, stones, and hurtful words: Relative effects of various forms of childhood maltreatment. American Journal of Psychiatry, 163, 993-1000. doi:10.1176/appi.ajp.163.6.993

34.

UNICEF. (2009, July). Child abuse: A painful reality behind closed doors. Challenges, 9, 1-9. Retrieved from http://www.unicef.org/lac/Boletin-Desafios9-CEPAL-UNICEF_eng(1).pdf

35.

United Nations. (2006). Common country assessment for Jamaica: 2006-2010. Retrieved from http://www.jm.undp.org/files/CCA-_Jamaica_3-final-unct.pdf

36.

Vissing

Y. M.

Straus

M. A.

Gelles

R. J.

Harrop

J. W.

(1991). Verbal aggression by parents and psychosocial problems of children. Child Abuse Neglect, 15, 223-238.

37.

Zerk

D. M.

Mertin

P. G.

Proeve

(2009). Domestic violence and parental reports of young children’s functioning. Journal of Family Violence, 24, 423-432. doi:10.1007/s10896-009-9237-4