Journal of Family Psychology 2011, Vol. 25, No.

2, 194 201

2011 American Psychological Association 0893-3200/11/$12.00 DOI: 10.1037/a0022888

Marital Conict Typology and Childrens Appraisals: The Moderating Role of Family Cohesion
Kristin M. Lindahl
University of Miami

Neena M. Malik
University of Miami Miller School of Medicine

Intense and frequent marital conict is associated with greater appraisals of threat and self-blame in children, but little else is known about contextual factors that might affect appraisals. Systemic family theories propose that to understand child adaptation, it is necessary to understand the interconnected nature of family subsystem relationships. In a sample of 257 families with 8- to 12-year-old children, this study examined whether a four-level typology of marital conict management was related to childrens perceptions of marital conict and their appraisals of perceived threat and self-blame. In addition, family cohesion was tested as a moderator of the relationship between marital conict style and childrens appraisals. Observational coding was used to group couples into Harmonious, Disengaged, Conictual-Expressive, and Conictual-Hostile groups. Childrens report of the intensity, frequency, and degree of resolution of interparental discord corresponded well with observers ratings. The relationship between marital conict style and appraisals of threat and self-blame was moderated by family cohesiveness. At high levels of family cohesiveness, no group differences were found for either perceived threat or self-blame, whereas when family cohesiveness was low, threat was higher for the Harmonious and Conictual-Hostile groups, as compared to the Conictual-Expressive group, and self-blame was higher for both conict groups (expressive and hostile), as compared to the Disengaged group. The results provide further evidence of interconnected nature of family subsystem relationships and the importance of distinguishing among different approaches to marital conict management for understanding the complex and perhaps subtle but meaningful effects different family system factors have on child adaptation. Keywords: marital conict, family cohesion, appraisals, self-blame, threat

There is now a large body of research that documents concurrent as well as longitudinal links between marital conict and childrens maladjustment (e.g., Cummings & Davies, 1994; Grych & Fincham, 2001). The degree of association tends to be modest, however, suggesting heterogeneity in how children respond to and cope with disturbances in the marital subsystem (Davies, Harold, GoekeMorey, & Cummings, 2002). Research examining cognitive mechanisms has found support for childrens appraisals as mediators of the relationship between marital conict and adjustment and also has shown links between appraisals of threat and self-blame and externalizing as well as internalizing behavior (e.g., DeBoard-Lucas, Fosco, Raynor, & Grych, 2010; Gerard, Buehler, Franck, & Anderson, 2005; Grych, Fincham, Jouriles, & McDonald, 2000). The two types of appraisals that have received the most attention are threat and self-blame (Grych, Seid, & Fincham, 1992).

Perceived threat refers to fears that the marital conict will escalate, that the child will get drawn into the conict, and that there is little the child can do reduce their anxiety. Self-blame refers to childrens beliefs that they are at fault for the parental discord, or that the conict is about them. The Role of Childrens Appraisals in the Impact of Marital Conict Cognitive models of human development propose that childrens affect and behavior are largely determined by how children structure and perceive the world around them. For children, it is their experiences within their social environment, especially their families, that shape their cognitive processing to a large degree (Vygotsky, 1978). The cognitive-contextual framework (Grych & Fincham, 1990) proposes that childrens appraisals are affected by both the specic properties of the conict (e.g., intensity, content) as well as contextual factors, including family characteristics (Grych et al., 2000). Most studies to date, however, have examined appraisals as mediators of the link between child adjustment and marital conict and relatively few have considered appraisals as the dependent variable of interest. Given links between appraisals and child functioning, however, it appears critical to understand what factors might
194

Kristin M. Lindahl, Department of Psychology, University of Miami; Neena M. Malik, Department of Pediatrics, University of Miami Miller School of Medicine. Correspondence concerning this article should be addressed to Kristin M. Lindahl, Department of Psychology, University of Miami, P.O. Box 249229, Coral Gables, FL 33124. E-mail: Kristin.Lindahl@gmail.com

MARITAL TYPOLOGY AND CHILDRENS APPRAISALS

195

shape threat and self-blame appraisals. It has been shown that marital conict that is intense and frequent is linked to greater threat and self-blame (Grych, Harold, & Miles, 2003), but beyond that, very little is known about how the qualitative nature of marital conict or how other family contextual factors might impact appraisals. Much of the research to date on marital conict and its impact on childrens adjustment has relied on self-report measures of marital conict and has operationalized marital conict as a continuous variable, varying in intensity from low to high. Marital research, however, suggests that how couples manage disagreements in their relationship also varies along other dimensions, such as levels of animosity and engagement (e.g., Gottman, 1993). The importance of differentiating between confrontational and hostile, avoidant and distancing, and more amicable and reciprocal approaches to conict management in marital interactions is well established in the marital literature (e.g., Bradbury, Fincham & Beach, 2000; Fincham, 2003; Roberts, 2000), but is less well integrated into the study of family functioning. Longitudinal research shows that couples can be volatile and highly emotionally expressive, but still have stable marriages, whereas hostile couples (both those conictengaged as well as those emotionally detached) are at higher risk for divorce (Gottman, 1993). How different couple relationship management styles might be perceived by children, however, is not well understood. Research suggests that parental and child perceptions of marital conict are related, though, as is often the case when different reporters perceptions are compared, where associations are often modest. Grych, Seid, and Fincham (1992) found childrens report of marital conict properties (i.e., the frequency, intensity, and degree of resolution of the conict) to correlate with parental report of conict, to a moderate degree (correlations .30 .39). What has not yet been studied, however, is how childrens perceptions of marital conict might differ depending on the nature of the marital conict itself. One of the aims of the present study was to examine childrens report of marital conict properties across couples that were highly conictual and combative, couples that were emotional and expressive but not acrimonious, couples that were disengaged and distant from one another, and couples that experienced relatively little conict. It is likely, though not yet empirically tested, that different marital conict styles relate in different ways not only to childrens perception of the conict, but to childrens attributions about the conict as well. Conict that is hostile and intense is likely to be threatening, and research shows children to be vulnerable to blaming themselves (Grych, Harold, & Miles, 2003). Conict varies, however, in its level of hostility, and conict that is acrimonious and antagonistic is likely to be more threatening than conict that is volatile and expressive, rather than verbally aggressive. The implications of conict that is characterized by aloofness, disengagement, and tension are less clear, though Sturge-Apple, Davies, and Cummings (2006a) did nd interparental withdrawal to show stronger, more direct associations with internalizing and externalizing maladjustment

than interparental hostility. A disengaged and avoidant marital dyad is likely to present some level of threat, as there are observable signs that the marital subsystem is not completely healthy, but it may not necessarily lead to selfblame. The present study examined multiple forms of conict management in the marriage (Harmonious, Disengaged, Conictual-Expressive, and Conictual-Hostile) and examined how childrens perceptions of these four styles might differ and also how they are related to childrens reports of feeling threatened by or to blame for marital difculties. Family-Level Moderators as a Link between Conict and Appraisals Very few studies have examined what contextual variables might moderate the relationship between marital conict and childrens appraisals. Although we are beginning to learn more about family and parenting factors that are associated with childrens appraisals, little is known about how the family context might affect the relationship between marital conict style and childrens appraisals. Central to the emotional security hypothesis is the conceptual nesting of interparental processes within the larger family system (Davies & Cummings, 1994). This framework would suggest that childrens perceptions of marital conict are likely to vary, depending on the larger family context in which the marital discord is embedded. The cognitivecontextual model more specically focuses on childrens appraisals and argues that they are inuenced by the broader context in which the marital conict occurs, including characteristics of the family (Grych & Fincham, 1990). Marital conict that occurs in the context of a cohesive family system and supportive parent-child interactions is likely to be perceived differently than conict occurring in the context of a web of emotionally distant or conictual family relationships (Fosco & Grych, 2007). Family cohesiveness is one of the family characteristics proposed by the emotional security hypothesis to serve a protective function for children exposed to marital conict (Davies et al., 2002). Family cohesion refers to the emotional bonding family members have toward one another (Olson, Russell, & Sprenkle, 1982). Cohesive families are generally described as exhibiting warm, close, and harmonious interactions across multiple family subsystem relationships (e.g., Minuchin, 1974). As Davies, Cummings, and Winter (2004) suggest, marital discord may be perceived more benignly by children when it occurs in the context of cohesive and supportive relationships in the family, whereas the harmful implications of marital discord may be heightened for children by family processes that reect high levels of discord, chaos, or disengagement. Witnessing marital disputes in the context of cohesive, warm family relationships is thought to increase childrens condence in their parents ability to resolve disagreements or problems in the marriage (Davies et al., 2002), thereby potentially reducing appraisals of threat or self-blame. Only a couple of studies could be located that examined the moderating role of the larger family system on chil-

196

LINDAHL AND MALIK

drens appraisals, and only one that examined whole family processes specically. Fosco and Grych (2007) studied patterns of emotional expression in the family and how they impact the meaning of interparental conict for children. Children were more likely to blame themselves for marital disputes, but not to perceive greater threat, when parents reported that the family emotional climate was characterized by high levels of negative affect and low levels of positive affect. This study is an important rst step in better understanding in how the family context might be related to childrens appraisals, but is limited in its somewhat narrow denition of family emotional climate, which was operationalized by parent report of their expression of negative affect (e.g., criticism, anger) and positive affect (e.g., giving praise, expressing affection) toward other members of the family. To more fully examine the family-level context in which marital conict occurs, a broader, more holistic, more integrative dimension of family functioning was targeted in this study. By focusing on family cohesion, this study attempted to examine a family-level construct that could tap into the type of general relationship patterning in the family system that emerges when marital, coparenting, and parentchild subsystems are simultaneously at play (Davies, Cummings, & Winter, 2004). The Present Study In the present study, couples and families were observed interacting. Four different marital conict styles were observationally derived, and comparisons across the four groups were made in terms of childrens perceptions of marital conict characteristics and childrens appraisals of threat and self-blame. In addition, family cohesion was tested as a moderator of the relationship between marital conict style and appraisals of self-blame and threat. The following specic hypotheses were tested. First, it was expected that childrens reports of the intensity, frequency, and duration of marital conict would correspond with the observational coding scheme, such that the ConictualHostile group was expected to receive the highest scores on these measures, and the Harmonious group was expected to receive the lowest scores. Second, it was hypothesized that a Conictual-Hostile marital conict style would be associated with the highest levels of child-reported threat and self-blame. Low levels of threat were expected for the Harmonious and Conictual-Expressive groups. Third, it was expected that a cohesive family atmosphere would attenuate reports of threat and self-blame for all groups. Method Participants The participants were part of a larger study and included 257 two-parent families with a child between 8 and 12 years of age (144 boys, 113 girls). Of the families expressing interest in the study and who met criteria (child aged 8 to 12; both parents required to participate), 85% agreed to participate. Families were recruited through newspaper ad-

vertisements and iers in schools and were paid $70 for their participation. The sample was diverse: 49% HispanicAmerican, 39% white, not Hispanic, and 12% AfricanAmerican/Caribbean Islander. The parents average age was 38 and the childrens average age was 9.5 (SD 1.43). The average length of marriage was 11 years, and 35% of the men and 39% of the women were on their second or third marriage. The families were middle-class, overall (monthly income: M $3,566, SD $1,838). Procedure Families participated as part of a larger study. Institutional Review Board approval of the study was secured, and parental consent and child assent were obtained. After completing questionnaires, the couples were videotaped engaging in a marital problem discussion task, and families (mother-father-child) were videotaped discussing a recent family argument. The order of the videotaped interactions was counterbalanced. For the marital discussion, couples were instructed to take 12 min and discuss one of the top three problem areas in their relationship, including how they would like to see the problem resolved. For the family discussion, families were asked to take 12 min to discuss a recent family argument that involved both parents and their child. They were instructed to describe what happened during the argument and to try to reach a solution to the problem. Measures
Dimensions of interparental conict. The Frequency, Intensity, and Resolution subscales from the CPIC (Grych, Seid, & Fincham, 1992) were used to assess childrens perceptions of interparental conict. The Frequency scale (6 items; range of total scores 0 to 10) examines how often parents argue. The Intensity scale (7 items; range of total scores 0 to 14) measures the verbal and physical intensity of marital conicts. The Resolution scale (6 items; range of total scores 0 to 12) assesses how well parents recover from and resolve conicts. Adequate internal consistency was obtained: Frequency ( .90), Intensity ( .83), and Resolution ( .82). Appraisals of interparental conict. The Self-Blame and Threat subscales from the CPIC (Grych, Seid & Fincham, 1992) were used to assess childrens appraisals of interparental conict. The Self-Blame scale (9 items; range of total scores 0 to 18) reects the degree to which children blame themselves for their parents conict and perceive the conict to involve child-related issues. The Threat scale (12 items; range of total scores 0 to 24) examines the extent to which children feel threatened by, and uncertain or not condent in, their ability to cope with marital conict. Adequate internal consistency was found: Self-Blame ( .84) and Threat ( .79). Observational measure of marital conict style. Marital interactions were coded with the System for Coding Interactions in Dyads (SCID; Malik & Lindahl, 2004) and Marital Conict Style (MCS) from the SCID is a categorical

MARITAL TYPOLOGY AND CHILDRENS APPRAISALS

197

code that represents a typology of couple conict management. Couples were coded as Harmonious, Disengaged, Conictual-Expressive, or Conictual-Hostile. Harmonious couples (n 123; 48% of the sample) had marital problem discussions that proceeded smoothly, without signicant conict. Disengaged couples (n 48; 19% of the sample) appeared passive, disconnected, or avoidant. Interactions were tense or awkward, though not hostile or highly conictual. Conictual-Expressive couples (n 44; 17% of the sample) appeared quick-tempered, volatile, and sometimes stubborn, but were not antagonistic or mean-spirited and usually ended their discussions in a positive manner. Conictual-Hostile couples (n 42; 16% of the sample) tended to be angry, negative, and critical, and the spouses tended to bicker and make limited progress in resolving the problem they were discussing. Adequate interrater reliability was established (kappa .82). The four marital conict types were not found to differ across the ethnic groups. Observational measure of family cohesion. The family interactions were coded with the System for Coding Interactions and Family Functioning (SCIFF; Lindahl & Malik, 2000). The family interactions were coded for cohesiveness. Cohesion represented the level of unity, togetherness, and closeness within the family. Cohesion was rated on a 1 (very low) to 5 (high) scale. Adequate inter-rater reliability was achieved ( .90). The videotaped interactions were coded by research assistants who were blind to all information about the families and also unaware of the hypotheses tested in the study. The marital and family interactions were coded by separate, nonoverlapping teams of coders. Both the SCID and the SCIFF have been found to be reliable with multiethnic samples (Lindahl & Malik, 2000; Malik & Lindahl, 2004). Coders received a minimum of 15 hr of training and watched each interaction at least three times before making global ratings of the interactions. Performance of coders was monitored, and feedback was given weekly to minimize coder drift. Forty percent of the families were randomly selected for the reliability analyses and were coded by three independent coders.

Results Demographic Variables Associations between demographic and dependent variables were examined. Family income and child age were signicantly and negatively related to childrens reports of the frequency, intensity, and resolution of marital discord (rs 0.14 to 0.20, ps .05). Family income also was related to appraisals of threat (r .24, p .001) and to family cohesion (r .24, p .001), though not self-blame (r .11, p .075). Family income was included as a covariate in all analyses involving perceived threat. Age was not signicantly related to threat, self-blame, or family cohesion (rs 0.01 to 0.10, ps .06). Gender was unrelated to dimensions of marital conict or threat (rs .09, ps .12), but was correlated with self-blame (r .13, p .027), indicating increased self-blame for boys. No ethnic differences were found. Mean levels of threat (M 10.26, SD 4.80) and self-blame (M 3.50, SD 3.31) were similar to those found in other community-based samples (e.g., Fosco & Grych, 2007; Richmond & Stocker, 2007), and the constructs were moderately associated with one another (r .32, p .001). Conict Management Style and Childrens Perceptions of Marital Conict To examine differences in childrens perceptions of dimensions of marital conict across the four MCSs, a oneway multivariate analysis of covariance (MANCOVA; covarying income and child age) was performed and a main effect of conict style was found, F(9, 598) 4.25, p .001, 2 .05. Follow-up ANOVAs showed signicant differences for all three dimensions of interparental discord. As seen in Table 1, childrens reports of marital conict properties were consistent with the coding typology. Posthoc comparisons, using a Bonferroni correction, indicated the lowest frequency of conict for the Harmonious group, whereas the Disengaged, Conictual-Expressive, and Conictual-Hostile groups did not differ from one another. The two high-conict groups, Conictual-Expressive and

Table 1 Means and Standard Deviations, Results of Analyses of Variance, and Group Comparisons for Appraisals by Marital Conict Style
Marital conict style Harmonious Variable Conict properties Frequency Intensity Resolution (n 123) 3.88 (2.51) 4.54 (3.37) 2.02 (2.54) Disengaged (n 48) 5.71 (3.00) 5.68 (3.22) 4.19 (3.21) C-express (n 44) 6.28 (2.43) 6.18 (2.92) 2.82 (1.93) C-hostile (n 42) 6.67 (2.57) 8.38 (2.87) 5.66 (2.65) F 9.74 5.68 7.31 2 .12 .08 .09 Group comparisonsa H D, C-Express, C-Hostile H, D C-Hostile H D, C-Hostile; C-Express C-Hostile

Note. Frequency, Intensity, and Resolution are subscales from the Childrens Perception of Interparental Conict scale (CPIC). df 3, 251. H Harmonious; D Disengaged; C-express Conictual-Expressive; C-hostile Conictual-Hostile. Only signicant results are reported under group comparisons (all ps .05). p .01. p .001.

198

LINDAHL AND MALIK

Conictual-Hostile, were similar in frequency and intensity of conict, though they differed in resolution ability, with more resolution difculties reported for the ConictualHostile group. The Disengaged group shared characteristics with both the Harmonious (low intensity) and the Conictual-Hostile (poor resolution) groups. Family Cohesion as a Moderator One of the main aims of the study was to examine appraisals of threat and self-blame across the four MCS groups and to determine whether cohesion moderated the association between MCS and childrens appraisals. The General Linear Model was used, which is an extension of the more restrictive ANCOVA (Keppel & Wickens, 2007), so that it would be possible to test the moderation effects of a continuous predictor (e.g., family cohesion) on a categorical independent variable (i.e., MCS), while still controlling for family income as a covariate. MCS was entered as a between-subjects factor to test for a main effect of MCS, and the interaction between MCS and family cohesion was entered to test for moderation. Family cohesion and income were entered as covariates. The nature of the signicant interaction effects were interpreted with pairwise comparisons of estimated marginal means at specied levels of family cohesion. Threat. A main effect of MCS was found for perceived threat [F(3, 263) 3.99, p .008, 2 .05], but this needs to be interpreted in light of a signicant MCS Cohesion interaction term [F(3, 263) 2.75, p .052, 2 .03]. To investigate the nature of the moderation of MCS on threat by family cohesion, follow-up pairwise comparisons of the estimated marginal means across the four MCS groups were run at both low (cohesion 1) and high (cohesion 5) levels of family cohesion. This allows us to determine whether the differences among the means for each MCS group would differ in families with different values of cohesion, without articially (and with loss of information) categorizing cohesion. At high levels of family cohesion (i.e., family conict 5), no signicant group differences are found. At low levels of family cohesion, however, an inspection of the pairwise comparisons shows the Harmonious (M 14.70, p .054) and Conictual-Hostile (M 17.65, p .003) groups to report signicantly higher levels of threat than the Conictual-Expressive group (M 10.01). No other comparisons were signicant. Thus, when family cohesion is low, marital conict is more threatening for the Harmonious and Conictual-Hostile groups, as compared to the Conictual-Expressive group, whereas when cohesion is high, perceived threat is low for all for groups (sample M 8.79). Self-blame. A main effect of MCS also was found for self-blame [F(3, 264) 5.04, p .002, 2 .05], though this also is qualied by a signicant MCS x Cohesion interaction term [F(3, 264) 3.66, p .012, 2 .04]. Pairwise comparisons of the estimated marginal means across the four MCS groups were run at both low (cohesion 1) and high (cohesion 5) levels of family cohesion. At a high level of family cohesion (i.e., family con-

ict 5), no signicant group differences were found (sample M 3.51). An inspection of the pairwise comparisons at low family cohesion, however, shows the Conictual-Hostile (M 7.69, p .001) and the Conictual-Expressive (M 7.01, p .051) groups to report signicantly higher levels of self-blame than the Disengaged group (M 3.62). No other comparisons were signicant. Thus, though no differences in self-blame are found when families are highly cohesive, when family cohesion is low, children from the Conictual-Hostile and Conictual-Expressive groups blame themselves more for marital conict than children from the Disengaged group. Discussion Some degree of conict is endemic to human relationships, and conict may threaten relationship strength and personal adjustment more than any other aspect of communication (Canary & Cupach, 1988). Not all conict is destructive, however, and certain types of conict may erode relationships and negatively affect children more than others. Research has demonstrated the importance of studying childrens appraisals in understanding the impact of interparental conict on childrens adjustment. When children are more threatened, they are at higher risk for internalizing problems, and when they feel more responsible for the conict between their parents, the data are mixed, but children appear to be at greater risk for both internalizing as well as externalizing difculties (Dadds, Atkinson, Turner, Blums, & Lendrich, 1999; Grych, Harold, & Miles, 2003; McDonald & Grych, 2006; Stocker, Richmond, Low, Alexander, & Elias, 2003). Limited research exists, however, on the contextual factors that might be related to the appraisals children make. As predicted by the emotional security hypothesis and the cognitive-contextual framework, in this study, both the qualitative nature of the marital conict and the family context in which it occurs were related to childrens appraisals of threat and self-blame. Three important ndings emerge from the present study. For one, childrens report of marital conict was consistent with behavioral observations, thus providing some initial validity for the marital conict typology. Children from the Harmonious group reported interparental discord to be infrequent, of low intensity, and well resolved. Just the opposite was the case for children in the Conictual-Hostile group, who reported high intensity and the lowest level of conict resolution of all of the groups. In fact, the total report of marital conict frequency, intensity, and degree of resolution for the Conictual-Hostile group was double that typically found in community samples and approached that reported by a domestic violence shelter sample (see Grych, Fincham, Jouriles, & McDonald, 2000). What distinguished the three groups for which conict or tension was apparent to the observers (i.e., the Disengaged, Conictual-Expressive, and Conictual-Hostile groups) was not the frequency of conict, as the three groups were similar in that regard, but rather intensity and resolution. As one might predict based on the group labels, intensity of marital conict was lower in the Disengaged group, as

MARITAL TYPOLOGY AND CHILDRENS APPRAISALS

199

compared to the Conictual-Hostile group, but children from both types of families described their parents as exhibiting difculty in their ability to successfully resolve disagreements. The Conictual-Expressive and ConictualHostile groups did not differ in childrens reports of frequency or intensity of conict, with both presumably exhibiting intense are-ups at home on a fairly frequent basis, but conict resolution was lower for the latter group. True to the group label, children in the families where parents were observed to fall into the Conictual-Expressive group appear to perceive their parents as emotionally expressive and perhaps volatile, but ultimately able to work through areas of disagreement successfully. The Disengaged group shared characteristics with both the Harmonious (low intensity) and the Conictual-Hostile (high frequency, poor resolution) groups. Like the trained coders, children from the Disengaged group appear to perceive their parents as experiencing low-grade, but chronic, tension or conict that is not easily resolved. Thus, school-aged children do appear to differentiate different types of marital conict styles in the home, in a similar vein to outside observers. The next set of questions in this study were related to the impact that different types of conict have on childrens appraisals of the conict in terms of both feeling threatened and blaming themselves for the conict. The second noteworthy set of ndings from this study is that childrens report of feeling threatened by the marital conict differed across the MCS groups, though this relationship was modied by family cohesion. When families are highly cohesive, children from all groups reported low threat. Feeling loved, cared for, and part of a unied team may ameliorate threat by allaying anxiety that the marital problems are serious or that they will disrupt the larger family system. Thus, even when couples have hostile, combative MCSs, if they can compartmentalize this and not let the conict seep into family interactions such that at a triadic level, family subsystem relationships are warm and a sense of togetherness or we-ness is maintained, then fear or anxiety about marital conict is low and no different for children exposed to hostile marital conict than for children with parents with harmonious marriages. Perhaps most interesting in terms of the moderating effect of cohesion on the relationship between MCS and threat is the contrast between the two conict groups at low levels of cohesion, as it suggests most clearly that different types of conict expression have very different meanings for children, depending on the larger context in which the conict occurs. Children from the Conictual-Hostile group reported higher level of threat than children from the Conictual-Expressive group when families lacked warmth and a sense of interconnectedness. Thus, in families where negative emotional expression is normative but effective at helping solving disagreements and the conict is not hostile, a lack of cohesion in the larger family context is irrelevant to perceived threat and threat is low. In contrast, in families with acrimonious marriages, childrens level of threat appears to be more affected by the nature of the larger emotional family context, and the results suggest that familylevel disengagement and distance is related to higher levels

of anxiety. A lack of togetherness at a family level may make the overt marital problems more ominous and more of a threat to the familys foundation. If cohesion is the glue that keeps the family together and helps them weather challenging life events, then when this bonding is lacking, childrens sense of security might be more profoundly threatened when the pillar of the family, the marital system, also gives off obvious signs of vulnerability. By examining specic types of marital conict, a clearer picture emerges, indicating that emotional fault lines at the family level can mean an important safety net for children is missing when they are in families where conict may be intense, frequent, and ongoing. At low levels of family cohesion, threat also was higher in the Harmonious group, as compared to the Conictual-Expressive group, suggesting that even in families where conict is an unpredictable occurrence in an otherwise calm and positive interparental relationship, children experience more anxiety when family togetherness is lacking. The present ndings depart from Fosco and Grych (2007), who did not nd positivity at the family level to modify the relationship between marital conict and appraisals of threat. In their study, however, marital conict was measured by childrens perceptions and was unidimensional. It is likely that in the present study, by tapping into elements of marital conict management that were either not assessed or not differentiated in the Fosco and Grych (2007) study, that information important to childrens appraisals was obtained. A third important nding from the study is that selfblame differed across the four MCSs, and this relationship also was moderated by family cohesion. As was the case for perceived threat, when family cohesiveness was high, no group differences were found. When family cohesiveness was low, however, children from homes with high conict marriages (i.e., expressive as well as hostile) reported more self-blame than children from homes with disengaged and avoidant marital interaction styles. As suggested elsewhere (Fosco & Grych, 2007; Sturge-Apple, Davies, & Cummings, 2006b), expressions of warmth and understanding in the family may convey acceptance and respect, whereas emotional distance or unavailability may convey rejection or indifference, with the former reducing self-attributions for problems in the family and the latter increasing a sense of personal responsibility. In support of this hypothesis, for both the Conictual-Expressive as well as the ConictualHostile groups, report of self-blame was reduced by half when family cohesiveness was high, as compared to when it was low. In the case of the Disengaged group, it may be that when there is emotional distance and disconnect in both the marital and the family subsystems, children perceive little reason to assume responsibility for problems in the marriage. Limitations and Future Directions The limitations of this study merit consideration. Although interesting associations between different styles of marital conict management, family cohesion, and chil-

200

LINDAHL AND MALIK adjustment: An emotional security hypothesis. Psychological Bulletin, 116, 387 411. Davies, P. T., Cummings, E. M., & Winter, M. A. (2004). Pathways between proles of family functioning, child security in the interparental subsystem, and child psychological problems. Development and Psychopathology, 16, 525550. Davies, P. T., Harold, G. T., Goeke-Morey, M. C., & Cummings, E. M. (2002). Child emotional security and interparental conict. Monographs of the Society for Research in Child Development, 67, 27 62. DeBoard-Lucas, R. L., Fosco, G. M., Raynor, S. R., & Grych, J. H. (2010). Interparental conict in context: Exploring relations between parenting processes and childrens conict appraisals. Journal of Clinical Child and Adolescent Psychology, 39(2), 163-175. Fincham, F. D. (2003). Marital conict: Correlates, structure and context. Current Directions in Psychological Science, 12, 2327. Fosco, G. M., & Grych, J. H. (2007). Emotional expression in the family as a context for childrens appraisals of interparental conict. Journal of Family Psychology, 21, 248 258. Gerard, J. M., Buehler, C., Franck, K., & Anderson, O. (2005). In the eyes of the beholder: Cognitive appraisals as mediators of the association between interparental conict and youth maladjustment. Journal of Family Psychology, 19, 376 384. Gottman, J. M. (1993). The roles of conict engagement, escalation, and avoidance in marital interactions: A longitudinal view of ve types of couples. Journal of Consulting and Clinical Psychology, 61, 6 15. Grych, J. H., & Fincham, F. D. (1990). Marital conict and childrens adjustment: A cognitive-contextual framework. Psychological Bulletin, 108, 267290. Grych, J. H., & Fincham, F. D. (2001). Interparental conict and child development: Theory, research, and applications. New York: Cambridge University Press. Grych, J. H., Fincham, F. D., Jouriles, E. N., & McDonald, R. (2000). Interparental conict and child adjustment: Testing the meditational role of appraisals in the cognitive-contextual framework. Child Development, 71, 1648 1661. Grych, J. H., Harold, G. T., & Miles, C. J. (2003). A prospective investigation of appraisals as mediators of the link between interparental conict and child adjustment. Child Development, 74, 1176 1193. Grych, J. H., Seid, M., & Fincham, F. D. (1992). Assessing marital conict from the childs perspective: The Childrens Perception of Interparental Conict Scale. Child Development, 63, 558 572. Katz, L. F., & Gottman, J. M. (1995). Marital interaction and child outcomes: A longitudinal study of mediating and moderating processes. In D. Cicchetti & S. L. Toth (Eds.), Emotion, cognition, and representation. Rochester, NY: University of Rochester Press. Keppel, G., & Wickens, T. (2007). Design and analysis (5th ed.). Upper Saddle River, NJ: Prentice Hall. Lindahl, K. M., Clements, M., & Markman, H. J. (1997). Predicting marital and parent functioning in dyads and triads: A longitudinal investigation of marital processes. Journal of Family Psychology, 11(2), 139 151. Lindahl, K. M., & Malik, N. (2000). The System for Coding Interactions and Family Functioning. In P. K. Kerig & K. M. Lindahl (Eds.). Family observational coding systems: Resources for systemic research (pp. 7792). Philadelphia, PA: Brunner/ Mazel. Malik, N. M., & Lindahl, K. M. (2004). System for Coding Interactions in Dyads. In P. Kerig & D. Baucom (Eds.), Couple

drens appraisals were found, as a cross-sectional study, determination regarding causality cannot be made. Longitudinal studies have shown marital distress to predict later family and child maladjustment (e.g., Katz & Gottman, 1995; Lindahl, Clements & Markman, 1997; Sturge-Apple, Davies, and Cummings, 2006a), but it also may be the case that some children are naturally more predisposed to appraisals of threat or self-blame due to temperamental characteristics or other biologically based factors (e.g., genetic risk for an anxiety disorder). Systemic theories of family functioning predict bidirectional and reciprocal effects among family subsystems over time, and only longitudinal data will be able to help us understand exactly how these transactional processes unfold. In addition, the marital typology used in this study focused on couples conict management styles, and it is unknown whether similar results would be obtained from organizing couple functioning along other dimensions, such as attachment style or satisfaction. Several directions for future research are suggested. As this is one of the rst investigations to apply a typological approach to examining marital conict and its impact on child functioning, replication of the present ndings is needed and, in particular, a better understanding of the lives of children from home, characterized by avoidant and distant marriages, is needed. Little is known about factors related to the capacity to compartmentalize marital problems and keep conict and tension restricted to interparental interactions. Developing a better understanding of how couples can minimize the impact of their disagreements on family interactions seems like an important aim for research, with implications for clinical practice. In addition, although the present study was one of the rst to examine associations between different types of marital conict management strategies and appraisals of threat and self-blame and how these associations are modied by the larger family context, it did not differentiate parent-child relationship quality from the triadic-level construct of cohesion, and research indicates that the unique relationships children have with each of their parents also affect the nature of their appraisals (DeBoard-Lucas et al., 2010). References
Bradbury, T. N., Fincham, F. D., & Beach, S. R. H. (2000). Research on the nature and determinants of marital satisfaction: A decade in review. Journal of Marriage and the Family, 62, 964 980. Canary, D. J., & Cupach, W. R. (1988). Relational and episodic characteristics associated with conict tactics. Journal of Social and Personal Relationships, 5, 305325. Cummings, E. M., & Davies, P. T. (1994). Children and marital conict: The impact of family dispute and resolution. New York: Guilford. Dadds, M. R., Atkinson, E., Turner, C., Blums, G. J., & Lendrich, B. (1999). Family conict and child adjustment: Evidence for a cognitive-contextual model of intergenerational transmission. Journal of Family Psychology, 13, 194 208. Davies, P. T., Cummings, E. M. (1994). Marital conict and child

MARITAL TYPOLOGY AND CHILDRENS APPRAISALS observational coding systems (pp. 173188). Hillsdale, NJ: Erlbaum. McDonald, R., & Grych, J. H. (2006). Young childrens appraisals of interparental conict: Measurement and links with adjustment problems. Journal of Family Psychology, 20, 88 99. Minuchin, S. (1974). Families and family therapy. Cambridge, MA: Harvard University Press. Olson, D. H., Russell, C. S., & Sprenkle, D. H. (1982). The circumplex model of marital and family systems, VI: Theoretical update. Family Process, 22, 69 83. Richmond, M. K., & Stocker, C. M. (2007). Changes in childrens appraisals of marital discord from childhood through adolescence. Journal of Family Psychology, 21, 416 425. Roberts, L. J. (2000). Fire and ice in marital communication: Hostile and distancing behaviors as predictors of marital distress. Journal of Marriage and the Family, 62, 693707. Stocker, C., Richmond, M. K., Low, S. M., Alexander, E. K., &

201

Elias, N. M. (2003). Marital conict and childrens adjustment: Parental hostility and childrens interpretations as mediators. Social Development, 12, 149 161. Sturge-Apple, M. L., Davies, P. T., & Cummings, E. M. (2006a). Impact of hostility and withdrawal in interparental conict on parental emotional unavailability and childrens adjustment difculties. Child Development, 77, 16231641. Sturge-Apple, M. L., Davies, P. T., & Cummings, E. M. (2006b). Hostility and withdrawal in marital conict: Effects on parental emotional unavailability and inconsistent discipline. Journal of Family Psychology, 20, 227238. Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press.

Received August 16, 2009 Revision received January 6, 2011 Accepted January 7, 2011

Members of Underrepresented Groups: Reviewers for Journal Manuscripts Wanted

If you are interested in reviewing manuscripts for APA journals, the APA Publications and Communications Board would like to invite your participation. Manuscript reviewers are vital to the publications process. As a reviewer, you would gain valuable experience in publishing. The P&C Board is particularly interested in encouraging members of underrepresented groups to participate more in this process. If you are interested in reviewing manuscripts, please write APA Journals at Reviewers@apa.org. Please note the following important points: To be selected as a reviewer, you must have published articles in peer-reviewed journals. The experience of publishing provides a reviewer with the basis for preparing a thorough, objective review. To be selected, it is critical to be a regular reader of the ve to six empirical journals that are most central to the area or journal for which you would like to review. Current knowledge of recently published research provides a reviewer with the knowledge base to evaluate a new submission within the context of existing research. To select the appropriate reviewers for each manuscript, the editor needs detailed information. Please include with your letter your vita. In the letter, please identify which APA journal(s) you are interested in, and describe your area of expertise. Be as specic as possible. For example, social psychology is not sufcientyou would need to specify social cognition or attitude change as well. Reviewing a manuscript takes time (1 4 hours per manuscript reviewed). If you are selected to review a manuscript, be prepared to invest the necessary time to evaluate the manuscript thoroughly.