The Implications of Early Marital Conflict for Children’s Development

Abstract

Although negative associations between the frequency and intensity of marital conflict and children’s adjustment are well documented, less is known about how parents’ conflict styles are related to children’s developmental outcomes. The present study examines whether exposure to different types of parents’ conflict styles, during a child’s first year of life, is related to children’s behavioral outcomes in the first grade. Parents’ conflict resolution styles (CRSs) and child outcomes were examined in a sample of 150 working-class, first-time parents and their children. It was hypothesized that infants’ exposure to more conflictual conflict resolution styles would predict poorer child outcomes over time. Results revealed that parents’ unique conflict styles mattered in unique ways for children’s development, but also that the interaction of parents’ styles, their dyadic conflict patterns, was also related to child outcomes. Results revealed that higher levels of parents’ depressive or angry CRSs in the first year predicted more internalizing problems for children, while constructive CRS was related to fewer externalizing problems. However, gender effects showed that higher rates of parental compliance during conflict were related to more internalizing problems in girls. Furthermore, dyadic results revealed that having one parent angrily engage in conflict and the other parent - withdraw, comply or angrily engage – was related to more externalizing problems for boys. Overall, results showed that parents’ different conflict resolution styles, during a child’s first year of life, are related to their children’s developmental outcomes 6 years later. These results emphasize children’s early vulnerability to parental conflict and hold implications for clinicians and practioners.

Exposure to parents’ conflict is one of the most robust predictors of poor social and emotional development in children (Cummings & Davies 2002; Rhoades, 2008; Zemp, Bodenmann, & Cummings, 2016). Despite parents’ efforts to protect their children from their arguments, most children are still exposed to interparental conflict. Although exposure to mild disagreements may help a child to develop effective strategies for coping with conflict, such as cooperation and positive social functioning (McCoy, Cummings, & Davies, 2009); poorly managed interparental conflict places children at risk for emotional and behavioral problems (Sturge-Apple, Davies, & Cummings, 2006). Research has shown that children from high-conflict families have more adjustment problems compared to children from low-conflict homes (Harold, Shelton, Goeke-Morey, & Cummings, 2004).

Some of the most consistent findings point to the negative effects of exposure to high interparental conflict and children’s internalizing and externalizing problems (Brock & Kochanska, 2016; Schoppe, Mangelsdorf, & Frosch, 2001; Tavassolie, Dudding, Madigan, Thorvardarson, & Winsler, 2016). Cui, Conger and Lorenz (2005) found that increases in marital problems, over a 5–year period, were associated with increases in adolescent difficulties, including the development of anxiety and depression. Extending this line of research, Kouros, Cummings, and Davies (2010) explored similar issues in younger children, ages 6-9 years old, and found that as interparental conflict increased, so too did children’s problem behaviors. These longitudinal findings highlight the significance of early childhood exposure to interparental conflict in predicting children’s later functioning.

While longitudinal studies have linked interparental conflict to the maladjustment of young children and adolescents, few studies have focused on how infants’ exposure to interparental conflict might impact their later development (Dorn & Du Rocher Schudlich, 2020; Du Rocher Schudlich, Jessica, Erwin, & Rishor, 2019; Du Rocher Schudlich, White, Fleischhauer & Fitzgerald, 2011). Assessing couples’ conflict during infancy is important not only because research indicates that interparental conflict increases across the transition to parenthood (Belsky & Rovine, 1990; Christopher, Umemura, Mann, Jacobvitz, & Hazen, 2015), but because the infant brain has great plasticity during this sensitive period of development (Shonkoff, Boyce, McEwen, 2009). Specifically, infants’ exposure to stressful experiences in the first year of life alters the size and neuronal architecture of the amygdala, hippocampus, and prefrontal cortex, and leads to functional differences in learning, memory, and executive functioning (National Scientific Council on the Developing Child, 2014). Graham, Fisher and Pfeifer (2013) found that 6-12 month old infants exposed to interparental conflict, what the authors referred to as “moderate stress” (e.g., angry speech of parents), was associated with increased neural activity in brain regions associated with emotion and stress reactivity. Furthermore, research has shown that as early as six months of age, infants’ exposure to interparental conflict is related to more disrupted physiological and behavioral indices of emotional regulation (Crockenberg, Leerkes, & Lekka, 2007; Moore 2010). Consequently, interparental conflict may be one critical source of stress during infancy that holds implications for children’s later socio-emotional development.

Conflict Resolution Style

Beyond focusing simply on the frequency or intensity of interparental conflict, research has shifted to understanding how parents fight or resolve disagreements and the implications for children. Parents’ conflict resolution styles are typically defined as constructive or destructive. Constructive conflict resolution styles include behaviors aimed at a positive and productive resolution of the conflict (e.g., problem solving, support, respect, active listening, and affection). In contrast, destructive conflict resolution styles include negative behaviors such as criticism, stonewalling, violence, threat, personal insult, hostility, marital withdrawal, and defensiveness (Cummings & Davies, 2002; McCoy et al., 2009, Overall & McNulty, 2016).

Destructive conflict resolution styles disrupt children’s sense of emotional well-being resulting in anger, worry, and sadness (Goeke-Morey, Cummings, & Papp, 2007; McCoy et al., 2009). Researchers suggest that destructive conflict resolution styles put children at risk for developing adjustment problems (i.e., internalizing and externalizing disorders) due, in part, to their lack of coping skills and/or their learned use of maladaptive skills (i.e., aggressive behavior) to cope with conflict (Buehler, Lange, & Franck, 2007; Madigan, Plamondon, & Jenkins, 2017). Conversely, children’s exposure to constructive conflict can ameliorate the effects of interparental conflict (Cummings & Davies, 2002). Research has found that parents’ displays of validation and affection to one another during conflict elicited positive emotional responses from children, such as happiness, safety, and security (Cummings, Goeke-Morey, & Papp, 2004; Lopez Larrosa, Escudero, & Cummings, 2009; McCoy et al., 2009). The use of constructive tactics has also been associated with a lower probability of children developing aggressive tendencies and adjustment problems (Cummings and Davies, 2010; Cummings et al., 2004).

Beyond broad categorizations of conflict as either destructive or constructive, research points to the utility of identifying different subtypes of conflict resolution styles (CRSs; Cummings & Davies 2002, McCoy et al. 2009). Specifically, Du Rocher Schudlich, Papp, and Cummings (2011a) conceptualized three conflict resolution styles (CRS): 1) Depressive CRS, 2) Angry CRS, and 3) Constructive CRS. A Depressive CRS includes behaviors such as avoidance, emotional distress, and withdrawal from conflict. Angry CRS includes behaviors that are physically or verbally aggressive, defensive, or hostile in nature, and are typically accompanied by high levels of conflict. Finally, Constructive CRS is characterized by respect, active listening, regulated communication and compromise.

In a study of 262 children between the ages of 8 and 16, Du Rocher Schudlich and Cummings (2007) found that parents’ constructive CRSs were related to fewer internalizing problems in children, while angry, and depressive CRSs were related to more internalizing problems in children. In general, previous literature indicates that angry and depressive conflict negatively impacts children’s emotional development and are related to more internalizing and externalizing problems (Du Rocher Schudlich and Cummings 2003; Koss et al. 2011). Whereas constructive conflict is positively linked to children’s healthy emotional development and fewer internalizing and externalizing problems (Cummings and Davies 2010; McCoy et al. 2009).

Whilst the relationship between mothers’ and fathers’ conflict resolution styles (CRS) and their children’s developmental outcomes has received some attention, less is known about the nature of this relationship for infant development. Although research indicates that interparental conflict increases across the transition to parenthood (Belsky & Rovine, 1990; Christopher et al., 2015), it is unclear if infants’ exposure to parents’ conflict styles in the first year holds long-term implications for their development. Based on the Emotional Security Theory (EST), children’s adjustment is influenced by their feelings of security in the family system. Specifically, children’s emotional security may be directly affected by their exposure to interparental conflict, and indirectly through the effects of interparental conflict on parenting, attachment relationships, and other family contexts (Cummings & Davies, 2011). Cummings and Davies’s (2011) EST theory provides a framework for understanding how parental conflict can affect infants. Infants as young as 3-6 months old have been shown to have different emotional responses to their parents’ behavior (Graham et al, 2013; Izard et al., 1995); specifically, infants are sensitive to angry parental exchanges that, in turn, may impact infants’ sense of security and safety.

In fact, Du Rocher Schudlich, White, Fleischhauer and Fitzgerald (2011b) found that infants have differential responses to parents’ conflict resolution styles. In a behavioral observation study of mothers, fathers and their 10-month-old infants, researchers found that angry and depressive conflict resolution styles were associated with infants attending more to parental discussions and expressing more negative reactions, whereas constructive conflict was associated with more positive reactions, and more infant play (Du Rocher Schudlich et al., 2011b). Similarly, in a study of 74 families with infants 6-14 months old, Researchers found that mothers’ and fathers’ greater use of constructive CRS were related to infants heightened emotional security through parental attachment (Du Rocher Schudlich et al., 2019). Additionally, fathers’ use of angry CRSs were related to greater infant emotional insecurity through insensitive parenting (Du Rocher Schudlich et al., 2019).

The present study expands upon the work of Du Rocher Schudlich and colleagues (2011b; 2019). The proposed research will evaluate 4 different conflict resolution styles (CRS; Kurdek 1994): (1) Constructive CRS, (2) Angry CRS, (3) Depressive CRS, and (4) Compliance CRS. Compliance CRS, an additional and distinct type of CRS proposed by Kurdek (1994) is defined as giving in with little attempt to present one’s own side of the issue. While compliance is a more covert CRS, similar to depressive CRS, it has not been studied in relation to children’s outcomes. Unlike depressive CRS, where infants might experience a lack of contact with both their parents due to parental withdrawal from negative interactions; a compliance CRS, which includes “giving in” during conflicts may potentially negatively impact children through parents’ suppression of their emotions. Thus, the current study aims to extend Du Rocher Schudlich and colleagues’ (2011b; 2019) by adding in a fourth type of CRS, compliance, and by examining how infant exposure to distinct parental conflict strategies is related to children’s developmental outcomes in the first grade. As a second step, beyond examining connections between mothers’ and fathers’ conflict styles and child outcomes, research also points to the interaction of parents’ conflict styles, their dyadic conflict patterns (e.g., demand-withdraw) as significant in predicting child outcomes.

Interaction of Parents’ Conflict Resolution Styles

Currently, the majority of research on couple conflict examines how the frequency or intensity of couple conflict is related to child outcomes, or how individual parental conflict styles affect children (Du Rocher Schudlich et al., 2011b). A large literature shows that the ways in which parents’ conflict styles intersect, such as one parent attacks and the other withdraws or both use attacking styles, has differential effects on relationship satisfaction and stability (Gottman & Notarius, 2000). In addition, there is a growing research base that examines how the interaction of mothers’ and fathers’ parental conflict styles are related to children’s socioemotional outcomes. Research on dyadic conflict styles and child outcomes reveals that one particularly harmful dyadic pattern of interparental conflict is referred to as the “demand-withdraw” pattern (Katz & Gottman, 1993; Katz & Woodin, 2002). This interaction style is characterized by one parent demanding or exhibiting an angry CRS, often resorting to pressure, demands, hostility and criticism, and the other person withdrawing or trying to avoid the discussion (depressive CRS). In a longitudinal study, Katz and Gottman (1993) found, through observational assessments of marital interaction, that parents’ demand-withdraw conflict resolution style, assessed when children were 5-years old, predicted higher teacher ratings of internalizing problems when children were 8-years old. Similarly, Katz and Woodin (2002) found that 4-5-year-old children exhibited significantly higher levels of psychological problems when at least one parent was withdrawn (depressive CRS) and one parent displayed an angry CRS during conflicts, than when both parents exhibited the same CRS (i.e., only angry or depressive CRS). Specifically, the demand-withdraw pattern predicted more externalizing problems and negative affect in children as well as noncompliance with peers compared to all other patterns of interaction.

Another negative dyadic type of interparental conflict is the “mutually hostile” pattern (Katz & Gottman, 1993). This type of interaction is characterized by both parents “demanding” or being contemptuous and belligerent with one another, often using sarcasm, mockery, and/or malicious attacks on a partner’s character to provoke or inflame the conflict. Katz and Gottman (1993) found this pattern of interaction predicted externalizing behavior patterns when children were 8 years old.

The present study seeks to build upon this work (Katz & Gottman, 1993; Katz & Woodin, 2002) by extending it to examine infants’ exposure to dyadic conflict patterns. Specifically, we examine how parents’ dyadic patterns of CRS, that infants experience during their first year of life, are related to children’s internalizing and externalizing behaviors in the first grade. Additionally, this study expands on the type of dyadic CRS patterns examined by Katz and colleagues. Specifically, past studies limited definitions of the demand-withdraw conflict pattern as only mothers high in demands and fathers high in withdrawal. The present study will consider this pattern in both directions (i.e. mother demand- father withdraw and father demand-mother withdraw). Furthermore, Katz and Woodin (2002) examined parents’ conflict resolution styles in a sample of children high in conduct problems. This may have magnified the relationship between patterns of parents’ conflict resolution styles and children’s externalizing problems, while overlooking implications for children’s internalizing symptoms, as well as limiting our understanding of these processes for children’s behavior in nonclinical community samples.

The Social-Contextual Context of Parents’ Conflict

The current study examines parental conflict and child outcomes in a unique social context, low-income working families, and at critical time in the family life cycle, the birth of a child. A large literature documents that interparental conflict increases during the transition to parenthood (Belsky & Rovine, 1990; Christopher et al., 2015); thus, we are studying interparental conflict during the especially challenging period of new parenthood. In addition, we examine these processes among a sample of low-income, first-time parents, all of whom are returning to low-wage jobs soon after the birth. These working parents have fewer financial resources and fewer family-friendly benefits to manage new parenthood compared to their middle-class counterparts, and are at higher risk for conflict and stress (Perry-Jenkins, Laws, Sayer & Newkirk, 2019). Finally, research indicates that brain plasticity during the first year of life confers great opportunity, as well as vulnerability, for children’s neurobehavioral development. Thus, infant exposure to stress, in this case exposure to parental conflict (Graham et al, 2013), may be especially salient during the first year of life and hold consequences for later development.

Current Study

Although the negative association between interparental conflict and children’s adjustment is well documented, less is known about how interparental conflict resolution styles (CRSs) predict children’s socio-emotional development, especially when the exposure to conflict occurs early in life. Moreover, we know little about how the interaction of parents’ conflict styles, as opposed to the independent effect of mothers’ and fathers’ styles, may have unique effects on children’s socioemotional development. Although we could not find studies linking dyadic conflict patterns to infant outcomes, the EST has shown that early infant exposure to conflict may be related to children’s wellbeing indirectly by threatening their sense of security (Du Rocher Schudlich et al., 2011b). The current study examines: 1) the ways in which parental conflict styles in a child’s first year of life predict children’s externalizing and internalizing behavior 6-years later, and 2) the interaction of mothers’ and fathers’ conflict styles as they relate to children’s externalizing and internalizing behaviors. These questions are examined in an understudied population of low-income parents experiencing the transition to parenthood.

Four different conflict resolution styles (CRSs) are examined, (a) Angry CRS, (b) Depressive CRS, (c) Compliance CRS, and (d) Constructive CRS. It is hypothesized that:

1. mothers’ and fathers’ angry conflict styles (e.g., yelling, attacking) in the first year of life will be associated with more negative outcomes, including more internalizing and externalizing problems, in children at age 6 (Cummings et al., 2004), however, we expect there to be a stronger relationship between parents’ angry CRSs and children’s externalizing problems compared to children’s internalizing problems;

2. parents’ depressive CRSs will be associated with greater internalizing and externalizing behavior problems in children at age 6 (Du Rocher Schudlich & Cummings, 2007; Sturge-Apple et al., 2006). Given that depressive conflict resolution strategies include more avoidant and withdrawn behaviors, we expect this strategy to be more strongly related to children’s internalizing problems compared to externalizing problems;

3. parents’ more compliant conflict resolution styles will be associated with more internalizing and externalizing behaviors in children at age 6. Although little prior research examines compliance, as a negative CRS, we expect that parental compliance, resulting in the suppression of emotion and/or more flat affect, will be related to worse outcomes in children (i.e. more internalizing and externalizing problems);

4. parents’ constructive CRSs will be associated with fewer behavior problems (internalizing and externalizing) in children at age 6 (Cummings et al., 2004).

The second set of research questions address the inherently dyadic nature of couple conflict. The dyadic patterns examined are based on past research and include: 1) mutual hostile or mutual angry CRSs– when both parents utilize angry CRSs in conflict, 2) demanding father-withdrawing mother – when fathers angrily engage in conflict and mothers withdraw (depressive CRS), and 3) demanding mother-withdrawing father – when mothers angrily engage in conflict and fathers withdraw. It is hypothesized that mutual angry CRS patterns will be related to more externalizing problems; demand-withdraw patterns will be related to more internalizing and externalizing problems. Exploratory analysis will also examine other possible patterns dyadic interaction including demand-comply -- when one partner angrily engages in conflict and the other partner complies to ending the conflict without defending their position. Although not previously examined, we believe that this pattern of engagement may breed resentment between parents as suppress their opinions and attempt to stifle their disagreement. In this case, because conflict is not being resolved, but instead building and exacerbating the issue, we believe that demand-comply patterns will be related to more negative child outcomes. Lastly, we expect that passive CRS patterns, like mutually depressive or mutually compliant in which both parents are averse to confrontations, will be generally related to more internalizing and externalizing problems in children.

In all analyses, child gender will be tested as a moderator of these processes. Past research is inconclusive regarding the role of parent or child gender in shaping the relationship between marital conflict and child outcomes. Some results indicate that boys are more vulnerable (consistent with the ‘male vulnerability model’; Kerig, 1996); others suggest that girls are more vulnerable (Baviskar, 2011; Cummings and Davies, 1994); and still others find that boys and girls are equally vulnerable, but react in different ways (the ‘differential reactivity model’; Davies and Lindsay 2001). Others report no gender differences (Buehler, Anthony, Krishnakumar, & Stone, 1997); thus, gender analyses will be exploratory.

Research Design and Method

Participants

Participants were part of a larger, longitudinal project examining the transition to parenthood among 153 dual-earner, working-class, heterosexual couples who were all going to be first-time parents. Participants were recruited from a metropolitan area in Western Massachusetts. Refer to Perry-Jenkins Smith, Goldberg and Logan (2011) for more information on recruitment criteria.

Men’s average age at the time of their partner’s pregnancy was 28.9 years (SD = 5.02). Women’s average age was 27.0 (SD = 4.80). The median family income was $46,000 with both parents being employed. All babies were full term and healthy at birth. At the time of recruitment, almost 80% of the couples (n = 119) were married, and the remaining were cohabiting. By the end of the first year, however, 84% of the families were married, nine of the cohabiting couples had married in the first year (n = 128), 14% were cohabitating (n = 22) and 2% were divorced or separated (n = 3). By year 6, 120 families remained in the study: 79% were married, while 4% were cohabiting, and 17% were divorced or separated. Given that we were interested in how the first year of marital conflict was related to children’s outcomes at age 6, we controlled for family structure at year 1. The three couples who separated over the course of the first year were excluded from the analyses because exposure to conflict would be different in intact versus separated households.

Since the original study was focused on working-class families, the sample was restricted to participants who had an associate’s degree or less. The majority of the sample (50.3% of women and 52.3% of men) had some type of trade school or vocational training (e.g., cosmetology school, refrigeration mechanic training). The remaining sample held an associate’s degree (27.5% of women and 15% of men), a high school diploma or GED (19% of women and 31.4% of men), or less than a high school degree (3.2% of women and 1.3% of men). In terms of racial and ethnic diversity the majority of participants were White (94.8% of women, 90.2% of men), followed by Latino (1.3% of women, 3.3% of men), Black (1.3% of men and women), Other (2.6% of women, 2.7% of men) or multiracial (2.6% of men).

Procedure

Data were collected at five time points across the first year of pregnancy, and at a sixth time point when the child was in the first grade. The specific time points were: 1) third trimester of pregnancy, 2) one month after the baby’s birth, but before the mother had returned to work, 3) one month after mothers returned to work fulltime, babies were 3-4 months old, on average, 4) six-months postpartum, 5) one year postpartum, and 6) when the children were six-seven years old. Parents were interviewed separately in their homes. Teachers completed questionnaires regarding children’s behaviors in the first grade at year 6.

Measures

Conflict resolution styles (CRSs).

Conflict resolution styles were measured using Kurdek’s 16-item Conflict Resolution Style Inventory (CRSI; Kurdek 1994). Mothers and fathers rated the extent to which they used four conflict resolution styles (CRSs) when fighting with their spouse. The four subscales in this measure are: positive problem solving, conflict engagement, withdrawal and compliance. In an effort to maintain consistent terminology with past research, we modified Kurdek’s terms to match Du Rocher Schudlich and colleagues’ constructs (2011b; 2019). The problem solving scale was defined as focusing on the problem at hand, sitting down and discussing differences constructively, negotiating, and compromising Hence forth, we refer to this as Constructive CRS. Conflict Engagement, or Angry CRS, was defined as launching personal attacks, exploding and getting out of control, getting carried away and saying things that aren’t meant, and throwing insults and digs. Withdrawal, or Depressive CRS, included remaining silent for long periods of time, “shutting down” and refusing to talk, tuning one’s partner out, and withdrawing, acting distant and not interested. Lastly, Compliance CRS was defined as not being willing to stick up for oneself, not defending one’s position, and giving in with little attempt to present their side of the issue. Spouses indicated how frequently (1 = never, 5 = always) they used each of the four styles to manage spousal disagreements. Subscales scores were calculated by averaging scores across the four items in each subscale. Conflict resolution styles were assessed at only four, of the five time points across year 1 (times 1, 3, 4 and 5). Cronbach’s α were calculated at time point for mothers and father separately. For angry CRS, the Cronbach’s α ranged from .72 to .83 and .75 to .84, for mothers and fathers respectively. There were similar Cronbach’s α ranges for each subscale: Constructive CRS: .66 to .76 (mothers), .71 to .80 (fathers); Depressive CRS: .79 to .81 (mothers), .80 to .85 (fathers); Compliance CRS: .77 to .83 (mothers), .64 to .75 (fathers).

To address the first research question, four latent variables were created for each CRS subscale (Constructive, Angry, Depressive and Compliant) using the measurements from the 4 time points. Because research has shown conflict resolution style to be a stable construct (Kurdek, 1994), latent variables represent each parents’ self-reported, trait-level conflict resolution behavior, or average use of each CRS.

In order to address the second research question, focused on the dyadic nature of conflict, we linked the conflict style used most often by each parent. We constructed a variable based on the match of each parent’s most commonly used CRS. Possible patterns included: (a) mutual angry (angry CRS x angry CRS), (b) demand – withdraw (Angry CRS x Depressive CRS), (c) demand – comply (Angry CRS x Compliance CRS), (d) mutual withdrawal (Depressive CRS x Depressive CRS), (e) mutual compliance (Compliance CRS x Compliance CRS), and (f) withdraw – comply (Depressive CRS x Compliance CRS).

Child Behavior.

The Behavior Assessment System for Children-Teacher Rating Scale (BASC-TRS) was used to assess behavioral and emotional problems of the children at age 6. Teachers were asked to rate, on a 4-point scale, the degree to which each item described the child. For the purpose of this study only the externalizing and internalizing composites were used. The externalizing composite (37 items) includes subscales that measure aggression, hyperactivity, and conduct problems. The internalizing composite (26 items) includes subscales that measure depression, anxiety, and somatization. The BASC-TRS demonstrates good reliability and validity with school-age children, with Cronbach’s alphas ranging from .89 to .97 for girls and .90 to .96 for boys (Reynolds & Kamphaus, 1992). The range of T scores for internalizing and externalizing was 39 to 81 and 40 to 69 respectively.

Control Variables

Relationship Conflict.

Relationship conflict was assessed using the conflict subscale of the Relationship Questionnaire (Braiker & Kelley, 1979). The conflict subscale (5 items) assesses how often parents argue and/or have negative interactions using a 9-point scale ranging from 1 (not at all/never) to 9 (very much/very often). Psychometric properties of the Relationship Questionnaire have been tested elsewhere (Belsky, Lang, & Rovine, 1985) and been shown to be consistent. Conflict was measured five times during year 1 (early conflict) and during year 6 (concurrent conflict). An average measure of early family conflict was created by averaging mothers’ and fathers’ scores across year 1. Early conflict and concurrent conflict at Time 6 were highly correlated (see Table 2). Initial models tested both early and concurrent conflict as control variables because research has linked parents’ conflict to internalizing and externalizing symptoms in children. To provide the most stringent test of the relationship between early parental CRS and children’s development, the analyses address whether, above and beyond concurrent (year 6) levels of conflict, early conflict resolution styles predict child outcomes. Year 6 conflict Cronbach’s α was .77 and .73 for mothers and fathers respectively. Year 1 conflict Cronbach’s α was .92 for mothers and fathers.

Table 2.

Intercorrelations

	1	2	3	4	5	6	7	8	9	10	11	12	13
1. Mother Conflict (year 1)
2. Father Conflict (year 1)	.54**
3. Mother Conflict (year 6)	.38**	.33**
4. Father Conflict (year 6)	.13	.45**	.48**
5. Mother Constructive CRS	−.49**	.34**	−.21*	−.06
6. Father Constructive CRS	−.14+	−.33**	−.26**	−.19+	.08
7. Mother Angry CRS	.59**	.42**	.36**	.13	−.56	−.04
8. Father Angry CRS	.31**	.57**	.26**	.33**	−.22**	−.45**	.28**
9. Mother Compliance CRS	.18*	.18*	.03	.08	−.31**	.04	.08	.12
10. Father Compliance CRS	.17*	.28**	.07	.20*	−.08	−.39**	.15+	.23**	−.00
11. Mother Depressive CRS	.26**	.21**	−.09	.01	−.35**	.10	.30**	.20*	.39**	.04
12. Father Depressive CRS	.19*	.36**	.21*	.28**	−.02	−.52**	.20*	.49**	.04	.42**	−.13
13. Externalizing Problems	−.04	.07	.09	.15	−.04	−.16+	.00	.07	−.01	.09	−.04	.08
14. Internalizing problems	.11	.27**	.16	.24*	−.08	−.07	.15	.24*	.13	−.03	−.01	.13	.25**

Note: Variables are an average measure of each CRS across a child’s first year of life unless specified.

⁺p < .10,

^*p < .05,

^**p < .001

Family Structure.

Family structure, married versus cohabiting, was also tested as a control variable. Research has shown that children born to married parents tend to achieve better cognitive and social outcomes, on average, than children born into other family structures, including cohabiting parents (Crawford, Goodman, & Greaves, 2013). Family structure was defined as 0 = cohabitating or 1 = married.

Child Gender.

Previous research on gender and interparental conflict has been inconsistent. Exploratory analysis will look at interactions between CRS and child gender to test gender differences. Gender was a dichotomous variable (0 for male and 1 for female).

Analytic Plan

Latent-variable structural equation models were used to test the hypothesized relationships in Mplus v.7 (Muthén & Muthén, 1998-2015). Using Kline’s (2011) “two-step” process of structural regression modeling, we first established fit of the measurement model before allowing any structural relations. For each parent, the measurement model created 4 latent variables for each CRS (angry, depressive, compliance, constructive) using measures taken at 4 time points across the child’s first year of life to ascertain an average self-report trait level measurement of each CRS across the child’s first year of life, for each parent.

To evaluate the absolute model fit for the measurement model and the baseline model, we used recommended fit indices and guidelines suggested by Kline (2016). Models with RMSEA values < .08, SRMR <.10, and CFI >.90 were considered a close fit to the data, and non-significant model X² statistics are considered a perfect fit to the data (Kline, 2016). Models used Full Information Maximum Likelihood to account for missing data.

Question 1: To examine the direct relationship between parents’ conflict resolution styles (CRS) and children’s socio-emotional development, 2 structural regressions were tested. The baseline model allowed each latent variable (mothers’ and fathers’ CRS) to be related to children’s problems (internalizing and externalizing) controlling for child gender, marital status and parental conflict (see Figure 1). The baseline model was then constrained; estimates for mother and father predictors were set to be equal. This approach compares a full model allowing for parent gender differences with a simpler nested model constraining parent gender effects to be equal across equations. A nonsignificant χ² was found for each outcome (Externalizing: χ²= 5.82, df = 4, p = .21; Internalizing: χ²= 4.98, df = 4, p = .29); thus, all final models were estimated constraining the parent gender effects to be equal. Although both mother and father CRS predictors were entered into each model, because the values were equal, tables only present one value for each CRS. Next, the baseline constrained model was modified to add latent variable interaction paths (parent CRS by child gender). This enabled us to test whether the relationship between parents’ CRS and child outcome (internalizing and externalizing), were moderated by child gender.

Figure 1.

Conceptual Model: The direct relationship between parents’ early conflict resolutions styles and children’s later socioemotional outcomes

Note: T1, T3, T4 and T5 represent different time points of measurement across the child’s first year of life. CRS is conflict resolution style. This model will include each CRS: Angry, Depressive, Compliance, and Constructive. Child Outcome: models tested two different child outcomes: internalizing problems and externalizing problems. Concurrent conflict is conflict level measured when the child is 6 years old.

To address question 2, the dyadic relationship between parents’ CRS and children’s outcomes, couple typologies were created by first identifying each partners’ most used conflict style (angry, depressive or compliant). While couples tended to report, on average, high levels of constructive CRS, which are related to positive child outcomes, our analysis focuses on when conflict breaks down; what parents reported as their most common negative patterns of interaction. Thus, focusing on the other CRSs, the possible typologies included (angry-angry, angry-depressive, angry-comply, depressive-depressive, depressive-comply, and comply-comply). Realigning these dyadic CRS typologies with the existent literature, the typologies examined are as follows: (a) mutual angry CSR (angry-angry), (b) demand – withdraw (angry-depressive), (c) demand – comply (angry-comply), (d) mutual depressive CRS (depressive-depressive), (e) mutual compliance CRS (comply-comply), and (f) withdraw – comply (depressive-comply). ANOVAs were run to test for group differences in child outcomes.

We accounted for multiple hypothesis testing using the Benjamini–Hochberg procedure (Benjamini and Hochberg 1995) for each set of analysis (child internalizing or externalizing outcomes). In this procedure, the original p-values for the individual estimates were ranked from 1 to m, where m was the total number of tests per model or outcome variable. If a p-value was smaller than a calculated value equal to its rank position in the ordering multiplied by 0.1 and divided by m, the estimate was declared to pass the Benjamini–Hochberg test with a 5% false discovery rate. All p values presented in the analysis are the Benjamini-Hochberg corrected P-values.

Overall, out of 150 families, there was complete data across study variables for 149 families at time 1, 133 families at time 3, 122 families at time 4, 115 families at time 5 and 110 families at year 6. Missing data ranged from less than 5% on CRS indicators to 27% on teacher reported child outcomes. The Little MCAR test showed that the data were missing completely at random (MCAR), X² (677) = 733.56, p = .065. Missing data in all ANOVAs and correlations were addressed through list-wise deletion while, as mentioned, Full Information Maximum Likelihood (FIML) was used to address missing data in all structural equation models.

Results

Descriptive Data

The means and standard deviations for the main study variables are presented in Table 1. Repeated measure ANOVAs were used to test for differences in conflict and conflict resolution style (CRS) over time and between parents. A repeated measures ANOVA revealed that parents reported significantly higher levels of conflict over time (F (1, 212) = 23.36, p < .001). Mothers also reported significantly higher levels of conflict compared to fathers’ (F (1, 212) = 9.23, p = .003).

Table 1.

Conflict Resolution Style (CRS) Descriptives

	Mother		Father
	M	SD	M	SD
Conflict
Year 1	3.84	1.09	3.35	1.06	Partner: F (1, 212) = 9.27* Time: F (1, 212) = 23.36**
Year 6	4.21	1.42	3.82	1.34
Angry CRS
T1	2.28	0.65	1.93	0.62	Partner: F (1, 206) = 12.22* Time: F (2.93, 602.95) = 0.54 a
T3	2.28	0.75	1.97	0.66
T4	2.26	0.74	1.93	0.71
T5	2.29	0.73	2.01	0.73
Depressive CRS
T1	2.40	0.76	2.34	0.81	Partner: F (1, 206) = 2.16 Time: F (3, 618) = 3.54* Quadratic: F (1, 206) = 7.60*
T3	2.32	0.79	2.22	0.80
T4	2.27	0.78	2.13	0.72
T5	2.34	0.72	2.23	0.78
Compliance CRS
T1	2.00	0.68	2.13	0.73	Partner: F (1, 206) = 0.05 Time: F (2.93, 602.62) = 0.23 a
T3	2.07	0.76	2.05	0.64
T4	2.01	0.74	2.05	0.69
T5	2.05	0.79	2.03	0.70
Constructive CRS
T1	3.56	0.53	3.75	0.59	Partner: (F (1, 206) = 3.36+ Time: F (2.93, 603.97) = 5.90*a Linear: F (1, 206) = 13.71**
T3	3.51	0.50	3.65	0.57
T4	3.57	0.56	3.68	0.65
T5	3.47	0.58	3.60	0.61

Note: CRS is conflict resolution style. T# represents the time the variable was measured at. T1 occurs during the 3^rd trimester, T3 occurs when the mother returns back to work ~ 4 to 6 weeks post birth, T4 occurs about 6 months post birth and T5 is at 1-year post birth.

^a= Mauchly’s test of assumption of sphericity was violated; degrees of freedom corrected using Huynh-Feldt (ε > .75).

⁺p < .10,

^*p < .05,

^**p < .001

Repeated measures ANOVAs were also used to test for differences in the use of each CRS style across time and across partners (mother/father). Overall, no effects emerged for compliance CRS or any interaction effects (parent by time effects). Mothers reported significantly higher use of angry CRS compared to fathers (F (1, 206) = 12.22, p = .001. No time effects emerged for Angry CRS. Results also showed that there was a significant negative linear effect of time in the use constructive CRS; parents reported using less constructive conflict resolution over time (F (1, 206) = 13.71, p < .001). There was also a trend level effect for parent such that fathers reported higher use of constructive CRS compared to mothers (F (1, 206) = 3.36, p = .055). For depressive CRS, results revealed a significant quadratic effect of time (F (1, 206) = 7.60, p = .006). The quadratic trend suggests that parents use of depressive CRS is highest at time 1 (m = 2.43) and 5 (m = 2.35), while parents use of depressive CRS declines over the intermedial timepoints (time 3 m = 2.35 and time 4 m = 2.30). No parent effects emerged for depressive CRS. Correlations between conflict, conflict resolution styles (CRS) and child outcomes were in the expected directions (Table 2); conflict level was weakly correlated with CRS variables.

In terms of child outcomes on the BASC, externalizing behavior scores ranged from 40 to 69 with an average of 46.72 (SD = 7.07). Internalizing scores ranged from 39 to 81 with an average of 48.96 (SD = 9.06). ANOVAs revealed that, on average, teachers reported that boys had significantly higher levels of externalizing problems (boys = 52.13; girls = 46.65; F (1, 108) = 12.62, p = .001) than girls, but virtually the same rates of internalizing problems compared to girls (boys = 47.29; girls = 47.16; F (1, 110) = 0.04, p = .847). A score of 60 to 69 is considered “at-risk” for a condition on the BASC, while a score of 70 or above is considered clinically significant. Only 8% of children were at risk for externalizing problems; no child met the clinical cut off. Similarly, 9% of children were considered at risk for internalizing behaviors and four children (4%) were above the clinical cut off. This is consistent with the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) which states that the 12-month prevalence rates of internalizing and externalizing disorders range from 2-5% of children (American Psychiatric Association, 2013). In our community sample, higher scores refer to elevated levels of symptoms but not clinically relevant levels.

Measurement Model

To evaluate the absolute model fit for each model, we used recommended fit indices and guidelines suggested by Kline (2016). Based on these metrics, the measurement model was an adequate fit to the data (χ² (525) = 874.514, p < .001; RMSEA = 0.07; CFI=0.87; SRMR=0.09).

Additionally, calculations of average variance explained indicated that at least 50% of the variance in the indicators was explained by the latent constructs (Angry: Mother 73%, Father 74%; Depressive: Mother 69%, Father 74%; Compliance: Mother 65%, Father 56%; and Constructive: Mother 54%, Father 64%). This suggests that the indicators are a valid measure of each construct. Composite reliability measures indicated that the measurement models provided a reliable measure of the latent constructs as demonstrated through their high internal consistency values (Angry: Mother CR = .91, Father CR = .92; Depressive: Mother CR = .90, Father CR = .80; Compliance: Mother CR = .88, Father CR = .83; Constructive: Mother CR = .82, Father CR = .87). Furthermore, all indicators were significantly predicted by their respective latent constructs -- all factor loadings and their error variances were significantly different from zero. Thus, these indicators were considered to be valid and reliable measures of each latent construct (see Appendix 1 for full model statistics).

Parents’ Conflict Resolution Styles (CRSs) and Children’s Development

Model 1 examined parents’ conflict resolutions styles across the first year of parenthood as predictors of children’s internalizing and externalizing problems in the first grade. The model tested family structure, concurrent conflict, child gender and income as control variables. Family structure and income were unrelated to child outcomes and thus were excluded from the final model to preserve power. Additionally, it is important to note that Model 1 did not find support for significant differences between mothers’ and fathers’ CRS and children’s outcomes (externalizing: χ²= 5.82, df = 4, p = .21; internalizing: χ²= 4.98, df = 4, p = .29). Thus, the following results represent parents’ CRSs collectively. Overall, results indicated partial support for the hypotheses that parents’ conflict resolution styles in the first year of their child’s life predicted child outcomes at age 6. See Appendix 2–4 for full model statistics.

Angry CRS.

As hypothesized, a 1-unit increase in parents’ use of Angry conflict resolution styles in the first year of parenthood was associated with a 4.46-unit increase in children’s internalizing problems in the first grade (B = 4.46, SE = 1.36, p = .004; table 3); but was unrelated to externalizing behaviors. No interactions emerged between parents’ angry CRS and child gender.

Table 3.

Unstandardized Pattern coefficients for Child Outcomes

	Internalizing Model		Externalizing Model
Structural Model	Estimate	(SE)	Estimate	(SE)
	Parameter Estimate
Child Gender ➔ Internalizing	28.44*	(9.87)	–––	–––
Concurrent Conflict ➔ Internalizing	1.00	(2.38)	–––	–––
Angry CRS ➔ Internalizing	4.46*	(1.36)	–––	–––
Compliance CRS ➔ Internalizing	5.16**	(1.07)	–––	–––
Depressive CRS ➔ Internalizing	-3.56*	(1.39)	–––	–––
Constructive CRS ➔ Internalizing	0.29	(1.97)	–––	–––
Compliance xG ➔ Internalizing	-7.28*	(2.43)	–––	–––
Child Gender ➔ Externalizing	–––	–––	4.95**	(1.32)
Concurrent Conflict ➔ Externalizing			1.92	(1.87)
Angry CRS ➔ Externalizing	–––	–––	−0.91	(1.22)
Compliance CRS ➔ Externalizing	–––	–––	−0.14	(1.40)
Depressive CRS ➔ Externalizing	–––	–––	0.29	(1.49)
Constructive CRS ➔ Externalizing	–––	–––	−3.48*	(1.58)
	Disturbance Variance
Internalizing	66.80**	(9.49)	–––	–––
Externalizing	–––	–––	41.23**	5.72

Note: Model Fit: Internalizing Model: AIC = 7310.46; Externalizing Model: AIC = 7108.44; xG = interaction with child gender, mothers and fathers CRS coefficients were constrained to be equal

⁺p < .10,

^*p < .05,

^**p < .001 after Benjamini-Hochberg correction.

Depressive.

Inconsistent with our hypothesis, a 1-unit increase in parents’ use of depressive CRS was associated with a 3.38-unit decrease in children’s internalizing problems (B = −3.56, SE = 1.39, p = .026; table 3). Parents’ depressive CRS was unrelated to children’s externalizing behaviors. No significant interactions emerged when testing child gender effects.

Compliance.

Results revealed a significant gender by compliance interaction indicating that greater parental use of compliance predicted greater internalizing behaviors for girls than boys (B = −7.28, SE = 2.43, p = .008; see table 3, Figure 2). Parents’ compliance was unrelated to externalizing behaviors.

Figure 2.

Interaction of Parental Compliance CRS and Child Gender Predicting Children’s Internalizing Problems

Constructive.

Finally turning to results related to parental problem-solving strategies, results revealed greater parental use of constructive CRS was related to fewer externalizing problems (B = −3.48, SE = 1.58, p = .042; see table 3). Parental constructive CRS was unrelated to internalizing behaviors.

Dyadic Family Patterns of CRS

Results indicated the following breakdown of couples’ dyadic patterns of CRS: 6% Mutual Angry CRS, 39% Demand-Withdraw, 18% Demand-Comply, 8% Mutual Depressive CRS, 4% Mutual Compliance CRS, 25% Withdraw-Comply. Because of the small number of couples that fell into each dyadic pattern and to retain power, we grouped the patterns into two main categories: negative engagement patterns and passive CRS patterns, reasoning that infants’ early exposure to negative, conflictual interchanges early in life would be most detrimental. Negative engagement patterns included any family in which one partner’s main conflict resolution style was angry CRS; this included mutual angry CRS, demand-withdraw, and demand-comply couples. Passive CRS patterns included any couple that did not use a hostile engagement strategy; this included mutual depressive CRS, mutual compliance CRS, and withdraw-comply. Two-Way ANOVAs indicated that negative engagement patterns were related to higher rates of externalizing problems in boys, compared to passive CRS patterns (F (1, 108) = 6.48, p = .036). No results emerged for children’s internalizing problems (F (1, 110) = 1.00, p = .320).

Discussion

The current study examined the extent to which parents’ early conflict resolution styles (CRS) were related to children’s internalizing and externalizing problems at age 6 within a sample of low-income, employed parents experiencing the transition to parenthood. This group of parents, facing greater financial stressors and poorer work conditions compared to their more affluent counterparts, were at increased risk for early conflict and relationship discord (Perry-Jenkins & Schoppe-Sullivan, 2018); and results indicated that during that critical first year of life parents’ conflict resolution strategies were linked to children’s outcomes at age 6.

Direct Effects of Parents’ CRSs on Children’s Problems

Partially consistent with our hypothesis, higher parental angry CRS, predicted more internalizing problems in children. This suggests that for boys and girls, exposure to parents’ hostility and anger in the first year of life may negatively impact children’s socio-emotional development. This finding is consistent with extant research with older children linking destructive conflict resolution styles to negative child outcomes (Cummings et al., 2004). Additionally, results found that parents’ higher usage of constructive conflict was related to teacher’s reporting fewer externalizing symptoms in children. This is consistent with Du Rocher Schudlich and colleague’s (2011b) findings showing that constructive conflict resolution styles (CRSs) were related to positive outcomes for both male and female infants. Furthermore, higher levels of parental depressive CRS in the first year predicted more internalizing problems in children. These findings are in line with Du Rocher Schudlich and colleague’s (2011b) results showing that 6 – 14 month old infants are attuned to interparental conflict. Specifically, they found that destructive parental conflict was associated with greater infant distress, while depressive conflict was related to greater infant frustration. As noted above, the current sample is unique in its focus on low-income, employed parents, quite different from Du Rocher’s and Cummings’ samples of more well-educated parents. This is an important point because research indicates that low SES families are at greater risk for stress and conflict during early parenthood (Perry-Jenkins, Goldberg, Pierce, & Sayer, 2007), suggesting links between parental conflict and child development may be heightened in these families. Expanding Du Rocher Schudlich and colleague’s (2011b) findings, however, our data provided evidence for gender differences.

Moderating Role of Child Gender

Despite inconsistent past research, results from the present study found that child gender moderated the relationship between parents’ conflict styles and children’s outcomes. Specifically, our study evaluated parental compliance as a unique CRS and found that higher levels of parental compliance were related to daughters’, but not sons’, internalizing symptoms. Although we could find no studies linking compliance, as a CRS, to children’s outcomes, these findings are consistent with previous research suggesting that girls are more vulnerable to parent conflict than boys (Baviskar, 2011; Cummings and Davies, 1994). The question of how girls’, but not boys’, exposure to parental compliance during conflict in the first year of life is related to later internalizing outcomes is intriguing. Perhaps, it is the continual exposure over the years of parents’ suppressing emotions through compliance that affects girls more than boys because we know girls are more attuned to parental emotions. Future research on how parents’ compliant conflict resolution strategies are expressed and are related to child outcomes is needed; observational couple interaction data would be useful in this regard as well as longitudinal designs.

The question of how early patterns of interparental conflict during infancy shape child outcomes years later is in need of further attention. While, our findings found evidence for gender differences, it may be the case, with our data, that results only emerged for girls internalizing problems because of teacher bias. Teachers are more likely to endorse externalizing problems in boys opposed to internalizing problems (Molins & Clopton, 2002). To address this shortcoming, it would be useful to have more observational data of child behavior as well as direct assessments of children’s functioning from children themselves, as opposed to relying only on adult reports. This later approach is supported in a recent study finding that children are better reporters of their internalizing symptoms (Miller, Martinez, Shumka, & Baker, 2014).

Dyadic Conflict Resolution Patterns

Our study extended the work of previous conflict researchers, like Du Rocher Schudlich and colleague’s (2011b), by examining conflict at the dyadic level — between parents. Because much parenting research has over looked the role of fathers, much past research on interparental conflict has relied solely on one parent’s report of conflict, usually mothers’, or in the case of observational studies, been limited to observational time points that are unlikely to capture patterns of conflict. Given that children are exposed to couple-level conflict, we were interested in examining the interactive nature of couple conflict styles. Analyses revealed that having one parent angrily engage in conflict and the other parent - withdraw, comply or angrily engage – was related to more externalizing problems for boys. This finding is in line with previous work which have reported that having one parent negatively engage in conflict may be all it takes for negative child outcomes (Katz & Woodin, 2002)

These results further highlight the importance of studying relationship conflict at the dyadic level. If our analysis had stopped at studying mothers’ and fathers’ conflict resolution styles individually, we would have concluded that negative parental CRSs were related to children’s internalizing problems. However, conflict is complex and requires the participation of two parties. Results showed that mutual angry CRS, demand-withdraw and demand-comply CRS patterns were related to more externalizing problems, but only for boys. These findings point to the possibility that boys may be more sensitive to hostile patterns of interaction early in life while girls seem to be affected more by passive CRSs (i.e., depressive, compliance).

These results point to the importance of continuing to study dyadic patterns of conflict. The demand-comply pattern made up 18% of our sample however, this pattern has rarely shown up in past research on the effects of marital conflict on children. Given the focus on working-class couples in this study, future work should investigate whether this is a conflict pattern is unique to this population. Additionally, the shift from parents’ individual conflict styles predicting internalizing problems to dyadic conflict patterns predicting externalizing problems may help explain the inconsistency in past research studying the relationship between parental conflict and differences in children’s outcomes by gender. Overall, while more work is needed, these preliminary results point to the importance of understanding parents’ interactive conflict styles in relation to children’s development.

It is important to emphasize, that unlike most of the past literature on this topic that includes higher SES samples, the current findings emerged in a sample of low-income, new parents having their first child. Moreover, all parents planned on returning to work within 6 months of the child’s birth though, in actuality, the majority were back at work within three months. Previous publications with these data revealed that this was a challenging transition for new parents with increases in stress and depression (Perry-Jenkins et al. 2019). In short, the lack of workplace policies and family supports for low-wage workers placed them and their infants at increased risk for conflict, stress and poorer socioemotional outcomes. The transition to parenthood has been shown to be challenging for most new parents, yet most recover. It is likely that the lack of paid parental leave, little schedule flexibility, and low wages that characterize low-income jobs exacerbate couple conflict and leaves parents and children more vulnerable.

Limitations and Strengths

Given the low-income sample used in these analyses, results may not reflect experiences across other social class groups. The lack of racial and ethnic diversity in the sample is another shortcoming and studies need to examine these processes in more diverse samples. Third, this study does not address the possibility that the impact of parents’ conflict resolution style (CRS) may depend on the relationship satisfaction. In the context of a loving, supportive relationship, perhaps conflict and negative interactions have less of an impact on child outcomes. Future research should explore the role of love and marital satisfaction in understanding the relationship between parents’ CRS and children’s development.

While inclusion of both parents’ reports is a significant strength of the current investigation, it is important to acknowledge the limitations of this approach. Parents may be biased in their self-reports of conflict styles, wanting to put their best foot forward. Alternatively, parents may lack insight into what conflict resolution strategies they are using. Future research should, in addition to collecting parents’ self-reports, also have parents report on their partners’ conflict resolution strategies (see Kerig, 1996 for example) and/or include observational data of couple interaction tasks to assess conflict styles, or use daily diary assessments of conflict from both parents and children (McCormick, Hseuh, & Cummings, 2018; McCormick, Hsueh, Merrilees, Chou, & Cummings, 2017; Papp, Cummings & Goeke-Morey, 2013) . Finally, it would also be useful to have children’s perceptions of the frequency, intensity and style of their parents’ conflict if age permitting, or if not, observational data or recordings of conflict interactions in front of the child may provide more objective measurements of family conflict and conflict resolution strategies.

Another strength of this study was the use of teacher reports of child behavior, an objective measure of children’s functioning outside of the family dyad. Since the present study did not use a clinical sample, only a small number of children met the clinical cut-off for diagnosis; 12-month prevalence rates for externalizing and internalizing disorders range from 2-5% of children (American Psychiatric Association, 2013). Despite the fact that most of the children fell below conventional thresholds for diagnosis, subsyndromal individuals continue to seek treatment, experience impairment, and are at risk for negative clinical outcomes (Ruscio, 2019). Thus, the study of subclinical populations remains clinically relevant. However, while the objective report of child behavior through teachers was a notable strength, it also had limitations. Teachers are not without their own biases in identifying behavioral problems in children (Molins & Clopton, 2002). To combat this, it may be useful for future researchers to incorporate children’s reports of their behavioral functioning as research has shown that children may be better reporters of their internalizing symptoms (Kemper, Gerhardstein, Repper, & Kistner, 2003). Alternatively, researchers could provide psychoeducation to teachers about the gender bias in identifying children’s behavioral problems prior to participating.

The question of how individual differences in children may also influence how a child reacts to conflict exposure is another area worthy of greater attention. Our results revealed some gender differences in child outcomes, but it is also likely that child temperament may moderate the effects of parental conflict styles on children and future research examining this process is warranted. (Davies, Hentges, & Sturge-Apple, 2016).

Finally, a key assumption in this study is that early exposure to parental conflict affects neural processes in the infant brain, linked to emotion regulation processes, that hold long-term implications for child development; however, we were unable to test this process directly. Future work is needed that links early parental conflict resolution styles to infants’ neurophysiological reactions to stress, that may be one pathway linking exposure to early parental conflict (i.e., stress exposure) to later developmental outcomes.

Conclusions and Implications

Overall, this study points to the critical role of parents’ conflict resolution styles in children’s early development. Taken together, our results indicate that although mothers’ and fathers’ conflict resolution strategies vary, parents’ usage of different strategies have effects that extend beyond children’s current wellbeing and impact their development up to 6 years later. Nonetheless, future research is needed to explore the longevity of these relationships. Results from this study have implications for both clinical practice and intervention. Because results point to children being affected by parental conflict in their first year of life, clinicians and interventionists are urged to incorporate conflict management and resolution as a focus to their treatment plans. Early prenatal interventions designed to help reduce parents’ usage of destructive conflict resolution styles may be crucial in supporting optimal parenting practices and child outcomes in new families.

Highlights:

• Parents’ conflict resolution styles (CRSs) during children’s first year of life matter for children’s development.

• More negative CRSs (angry, depressive, compliance) were related to more internalizing problems in 6 year old children.

• More positive CRSs (constructive) were linked to less externalizing problems in 6 year old children.

• Dyadic patterns of conflict resolution (i.e. demand-withdraw, mutually angry, demand-comply) were related to children’s externalizing.

• Results support the continued exploration of dyadic patterns of parents’ CRSs to children’s outcomes.