1
Constructive Interparental
Running head: CONSTRUCTIVE CONFLICT STRATEGIES
Constructive Interparental Conflict Strategies
as a Predictor of Adolescent Aggression and Conflict
Julie Ann Gdula
University of Virginia
Distinguished Major Thesis
Advisor: Joseph P. Allen
Second Reader: Robert E. Emery
Abstract
This study examined the relationship between father’s positive reasoning during conflicts with mother and several adolescent outcomes, including aggressive attitudes and conflict strategies with peers and romantic partners. Participants included a diverse community sample of 126 adolescentsand their mothers, peers, and romantic partners in a multimethod, multiple reporter, longitudinal study. Results revealed that father’s positive reasoning with mother at teen age 13 predicted lower adolescent attitudes toward aggression and less antagonistic conflict with a romantic partner at teen age 17 in the long term. Father’s positive reasoning predicted adolescent attitudes toward aggression through a relationship partially mediated by adolescent autonomous relatedness with peers at age 14 and adolescent aggression at age 15.Findings are interpreted as suggesting pathways by which parents may shape their adolescents’ beliefs and behaviors through modeling and socialization, and the need for further research in positive psychology is discussed.
Constructive Interparental Conflict Strategies
as a Predictor of Adolescent Aggression and Conflict
There is little dispute within the field of psychology that interparental abuse and negative conflict style, such as verbal and physical aggression, have adverse consequences for children and adolescents. Interparental conflict has long been known to have effects on both the cognitive and social development of children (Wierson, Forehand, & McCombs 1988). Consistent conflict between parents has been linked to several negative outcomes in children, such as delinquency and increased aggression in peer relationships (Davies & Cummings, 1994). Research shows that adolescents who witness violence between their parents are more likely to engage in aggressive acts (Moretti, Obsuth, & Odgers 2006), and boys exposed to aggressive interparental conflict are more likely to judge aggression as acceptable within romantic relationships (Kinsfogel & Grych, 2004). Though aggressive conflict has been consistently predictive of negative outcomes for teens, some studies suggest that different types of conflict styles lead to very different outcomes in adolescence. For example, a hostile argument style between parents is more closely linked with teen problem behaviors than the frequency of argument (Buehler, Krishnakumar, & Stone 1998). Although no relationship is without conflict, these findings suggest that what matters is not that an argument occurred, but how it is handled by the parents (Du Rocher Schudlich & Cummings 2003).
How are researchers to decide which conflict styles are most predictive of positive outcomes for adolescents? To date, most researchers have focused on children and their immediate emotional responses to conflict to answer this question. Grych and Fincham’s (1990) cognitive-contextual framework presents an interdependent model of several variables which refer to the actual argument, context elements pertaining to the child’s mood and environment, and the child’s emotional and coping behaviors in response. Subsequent research has found support for this model for internalizing problems, but not for externalizing problems, suggesting that further research is necessary to determine a framework for understanding such problems as aggression and delinquency (Dadds, Atkinson, Turner, Blums, & Lendich, 1999). Based on the apparent importance of children’s perception of their parents’ arguments, Grych, Seid, & Fincham (1992) developed a questionnaire to determine children’s perspectives on conflict, called the Children’s Perception of Interparental Conflict Scale (CPIC). Following in this vein, Davies & Cummings (1994) propose an emotional security hypothesis which has driven most further study in the field. It characterizes different interparental conflict tactics by the children’s beliefs about the future stability of their parents’ relationship, or emotional security. In one study based on this theory, if children felt threatened by arguments, they were more likely to exhibit both internalizing and externalizing symptoms one year later (Harold, Shelton, & Goeke-Morey 2004). Using the emotional security hypothesis, Goeke-Morey, Cummings, & Harold (2003) classified parental conflict tactics as either constructive or destructive, according to children’s immediate emotional responses to video representations of strangers in a hypothetical argument situation. Emotional responding produced equivocal results for calm discussion, problem solving, and support. Notably, this study used actors, and not the children’s real parents, to elicit emotions. Thus the emotions were only related to one controlled episode involving people with whom the child had no real relationship.
Few methods and theories other than measurement or observation of children’s emotions have been used in classifying parental conflict styles and their effects on children. Katz & Gottman (1993) made observations of parents’ conflicts and then assessed children’s internalizing and externalizing behaviors three years later. Though the results indicate that different conflict styles do contribute to different outcomes in the children, both patterns studied included hostile, angry, and withdrawn behavior from the parents. Another study used a stress and coping model to explain children’s reactions to parental conflict (Cummings 1998). According to this model, children’s distress leads them to become sensitized to anger, making them incapable of effectively coping with angry situations in the future. Though the author acknowledges a difference between destructive and constructive tactics, he does not provide much detail about the latter. To date, little research has been done on the effects of constructive conflict tactics. Cummings & Davies (2002) purport the need for a process-oriented approach to the effects of marital conflict on children; this requires a deeper exploration of interparental conflict tactics as protective factors, and the effects on children and adolescents over a longer period of time.
Though the long-term implications of constructive conflict tactics have yet to be examined, Cummings, Goeke-Morey, and Papp (2004) found that in the short term, destructive tactics increased aggression while constructive tactics decreased it. This aggression led to delinquent conduct problems later in adolescence, which suggests that constructive interparental conflict tactics may well serve as a protective factor for teens. In addition to conduct problems, aggression in adolescence has been correlated with a series of physical and emotional problems, including cardiovascular disease, internalizing and externalizing behaviors, lowered social status, and negative academic achievement (Smith & Furlong 1994). Adolescents’ attitudes toward aggression have also been identified as a risk factor for violence in romantic relationships (Rickert, Vaughan, & Wiemann 2002). Due to these serious correlates of adolescent aggression, identifying how constructive interparental conflict may serve to protect adolescents against later aggression is crucial.
Even before they begin dating, aggression may adversely affect adolescents through their peer relationships. Aggression has been closely linked with peer rejection (Coie, Dodge, & Kupersmidt 1990), and the literature suggests that parents play a major role in the development of adolescent aggressive behavior and attitudes. Studies indicate that parents’ aggression and conflict styles play a significant role in the establishment of their children’s friendships, within both the interparental relationship and the parent-child relationship. Mother and father aggression in general predicted less peer intimacy and poor social skills in adolescence, with the effect of father aggression evident only in later adolescence (Schlatter, 2001). Overtly aggressive children tend to have parents who exhibit more conflict and aggression within their relationship (Grotpeter, 1998). The parent-child relationship also appears to be a mechanism by which parental aggression influences friendships; adolescents aggressed by their parents are less popular, less well-liked, and described as more externalizing in behavior by their peers, and adolescent victims of physical abuse are more likely to inflict such abuse on peers (Tencer, 2005; Wolfe, Scott, Wekerle, & Pittman, 2001). The serious potential impact of parents on their children in the form of peer relationships requires attention, as it will likely predict future social skills and the success of romantic relationships.
Research indicates that adolescent peer relationships correlate with romantic relationships later on. For example, Feiring, Deblinger, Hoch-Espada, & Haworth (2002) found that girls with less secure friendships were more likely to exhibit aggression toward romantic partners. However, as in parental relationships, conflict is an unavoidable and even necessary part of adolescent romantic relationships. One study found that conflict is particularly characteristic of various types of adolescent relationships and may even serve a developmental purpose as adolescents learn to compromise with others while maintaining autonomy (Laursen 1995). However, as in parental relationships, it is the way conflict is handled that matters most, and there is considerable evidence that teens learn how to do this from their parents. Adolescents who observed aggression between their mothers and fathers reported higher levels of aggression toward their own romantic partners (Moretti, Obsuth, & Odgers 2006), which indicates an indirect, observational relationship between the conflict styles of the parents and the adolescents. Reese-Weber and Bartle-Haring (1998) found this same indirect relationship between the adolescent’s parents’ conflict style and the adolescent’s style with a romantic partner. In addition, they also found a direct relationship; rather than just observing their parents and learning from them how to interact with a romantic partner, adolescents usually used the same conflict style with a romantic partner as they used in conflicts with their parents. This suggests that adolescents learn from their parents both by observing them and by interacting with them. Martin(1990) also suggests that teens learn how to act in their dating relationships based on the specific quality of their relationships with their parents. One study found that less securely attached adolescents demonstrated a more negative affect, less confidence, and a more maladaptive conflict style in arguments with romantic partners (Creasey & Hesson-McInnis, 2001). Thus it seems that both modeling or socialization and attachment theory serve as mechanisms by which interparental conflict tactics lead to the development of children’s conflict styles and aggressive behavior in romantic relationships.
Several of the aforementioned studies revealed different results for boys than for girls. For example, while some studies have found that children of aggressive parents will also display aggression regardless of gender (Crick 1996, Mizokawa 2000), others only find results for boys (Kinsfogel & Grych 2004). The difference lies in the fact that girls use more relational aggression than overt aggression, meaning that girls are more likely to spread rumors or deliberately make others jealous than they are to use physical force or act openly hostile. When aggression is operationalized to include relational aggression, a study will usually show similar levels of aggression in boys and girls. In relationships, however, boys are more likely than girls to endorse aggression toward a partner (Fering, Deblinger, Hoch-Espada, & Haworth, 2002). In adolescence, gender seems to play a role in differentially affecting results of studies on aggression.
Gender may also play a role for parents of adolescents. Most child and adolescent research has focused on mothers, and fathers have been largely ignored; however, recent research suggests that fathers become increasingly important in adolescence, and their impact should not be overlooked. Kempton, Thomas, and Forehand (1989) found that fathers’ aggressive conflict tactics were influential in adolescent functioning, but mothers’ were not. Allen, Hauser, and Bell(1994) found a correlation between fathers’ displays of autonomy and relatedness toward their adolescents and the adolescents’ psychosocial development. Research also indicates that fathers’ psychopathology is related to adolescent psychopathology (Phares and Compas, 1992). Though the reason for the increasingly important role of the father as his child enters adolescence is unclear, some researchers posit that the father is the parent who mediates a child’s transition into adulthood and the larger community outside the family. Consistent with this, one study finds that mothers are more family-oriented than fathers or adolescents (Jurich, Schumm, & Bollman 1987); this suggests that mothers may serve as more accurate reporters of conflict tactics between family members.
Previous studies have provided a solid empirical foundation for the questions investigated in the present study, but none have examined the long-term effects of constructive marital conflict style on adolescent aggression and conflict in peer and dating relationships. The present study evaluates change over a span of five years – from early to late adolescence. Instead of evaluating the emotions of participants immediately after viewing a video in a controlled setting, the current study focuses on general behaviors and attitudes and their relationship to real parents’ conflict styles. To eliminate social desirability effects and reporter bias, the study employs a multi-informant design, combining information from adolescents, their mothers, their peers, and their romantic partners. In the current study, we predict that fathers’ positive reasoning in arguments with mothers will be associated with a lower level of teen aggression by late adolescence. We also hypothesize that it will predict improvements in the quality of the teen’s future peer and romantic relationships. We propose that positive reasoning in interparental conflicts will function as a protective factor against the adolescent’s development of less effective conflict strategies, such as the endorsement of aggressive conflict tactics and maladaptive behavior in arguments with peers and romantic partners.
Method
Participants
This report is drawn from a larger longitudinal investigation of adolescentsocial development in familial and peer contexts. Participants included126seventh and eighth graders (52 boys and 74 girls) and their mothers. Participants returned for subsequent annual visits with their peers and romantic partners. Adolescents completed initial interviews at approximately age 13 (M=13.3, SD =.61) and were then reinterviewed annually for the next six years. The sample wasracially/ethnically and socioeconomically diverse: 67% identified themselves as Caucasian, 21% as African American, and 12% as mixed race or other. Adolescents’ parents reported a median familyincome in the $30,000 –$39,999 range. At each wave, adolescents’ were also asked to nominate their “closestfriend” of the same gender to be included in the study. Closefriends were defined as “people you know well, spend time with and whoyou talk to about things that happen in your life.” For adolescents who haddifficulty naming close friends, it was explained that naming their “closest”friends did not mean that they were necessarily very close to these friends,just that they were close to these friends relative to other acquaintancesthey might have. Adolescents were recruited from the 7th and 8th grades at a publicmiddle school drawing from suburban and urban populations in the southeasternUnited States. An initial mailing to parents of students in therelevant grades in the school gave them the opportunity to opt out of anyfurther contact with the study. Only 2% of parents opted out of suchcontact. Of all families subsequently contacted by phone, 63% agreed toparticipate and had an adolescent who was able to come in with both aparent and a close friend. This sample appeared generally comparable tothe overall population of the school in terms of racial/ethnic composition(37% non-White in sample vs. approximately 40% non-White in school)and socioeconomic status (mean household income =$44,900 in samplevs. $48,000 for community at large). The adolescents provided informedassent, and their parents provided informed consent before each interviewsession. The same assent/consent procedures were also used for collateralpeers and their parents. Interviews took place in private offices within auniversity academic building.At the first wave of data collection, adolescents came in for two visits,the first with their parents, and the second with their identified closest friend. Approximately 1 year later, adolescents again came in for twovisits, the first alone and the second with the person whothey named as their current closest friend in the wave 2 individualinterview. Parents, adolescents, and peers were all paid for their participation.
Of the original 184 participants, 126 of them had father figures about whom their mothers completed questionnaires in wave 1. This sample differs significantly from the subsample of participants without father figures, in gender and income. Boys comprise 58.6% of participants without father figures but only 41.3% of participants with father figures. The median family income of participants without father figures was in the $15,000-19,999 range, while the median family income for participants with father figures was in the $30,000 –$39,999 range.
The sample that participated in both waves 1 and 2 of the study wasa subsample of 112 adolescents who had complete data at Wave 1. Adolescents reported knowing their close friends for a mean of 4.73 years. Attritionanalyses revealed no significant differences between the 112 adolescents who did versus the 14 adolescents who did not return for the second waveof the study on any of the demographic or substantive measures in thestudy.
In addition to selecting close friends, adolescents had the option to select romantic partners as well, starting with the fifth wave of the study. To qualify, partners must have been dating for at least two months at the time of the visit. The sample that both participated in wave 1 of the study and selected a romantic partner was 49. Attritionanalyses revealed no significant differences between the 49 adolescents who did versus the 77 adolescents who did not select romantic partners on any of the demographic or substantive measures in thestudy.
The subsample of participants with complete data at wave 6 of the study included 95 of the original 126 participants. Attrition analyses revealed no significant differences between this subsample and the original subsample.