Fathers’ Parenting and Coparenting Behavior in Dual-Earner Families: Contributions of Traditional Masculinity, Father Nurturing Role Beliefs, and Maternal Gate Closing

Abstract

We investigated whether dual-earner fathers’ adherence to traditional masculine norms, father nurturing role beliefs, and maternal gate closing behavior predicted the quality of new fathers’ observed parenting and coparenting behavior. Data were drawn from a longitudinal study of the transition to parenthood among 182 dual-earner different-sex couples. Expectant fathers reported their masculine agency, hostile sexism, gendered provider beliefs, and father nurturing role beliefs in the third trimester of pregnancy. Maternal gate closing behavior was coded from observations of mother-father-infant interaction at 3 months postpartum. At 9 months postpartum the quality of fathers’ parenting behavior was coded from observations of father-infant interaction, and the quality of fathers’ coparenting behavior was coded from observations of mother-father-infant interaction. SEM analyses indicated that fathers who held stronger father nurturing role beliefs showed more positive parenting behavior and less undermining coparenting behavior. Fathers higher in masculine agency also showed more positive parenting behavior. Mothers’ greater gate closing behavior was linked to less positive parenting and less supportive coparenting behavior by fathers. More positive couple behavior observed prenatally was also associated with better parenting and coparenting by fathers. These results highlight the complexity of relations of traditional masculinity, father role beliefs, and maternal gate closing with the quality of new fathers’ behaviors with children and partners in dual-earner families.

During the past 50 years, American fatherhood has changed due to shifting expectations, norms, and behaviors (Bianchi et al., 2006; Galinsky et al., 2011). Contemporary fathers are more involved with children, do a greater share of housework, and report that emotional availability, family time, and father-child bonding are more important to them than ever before (Gerson, 2010; McGill, 2014). Despite trends toward more engaged and equitable parenting by men, however, substantial gender inequalities in family life persist (Parker & Livingston, 2017). The persistence of gaps between mothers and fathers in domestic labor may be rooted in paternal expectations reified by social pressures, economic and institutional barriers, gender norms, and gendered identities (Connell & Messerschmidt, 2005; Petts et al., 2018; Risman, 2004). From the colonial period in the U.S. through the 1970s, these traditional expectations stipulated that fathers, and particularly European American fathers, emphasize stern patriarchal household control and breadwinning (Pleck & Pleck, 1997), although more contemporary expectations also underscore the importance of caregiving and emotional labor (Pleck, 2010).

Because variability in paternal childrearing practices and traditional fathering expectations persist, scholars have started to consider how traditional masculine norms influence men’s parenting. Fathers who endorse a traditional view of masculinity which eschews caregiving and emphasizes male dominance, emotional stoicism, and the prioritization of work tend to be less involved in their children’s lives (Gaunt, 2006; Petts et al., 2018) than fathers who believe that nurturance is central to the paternal role (Rane & McBride, 2000). Similarly, fathers—and mothers—who endorse traditional gender roles report more conflictual coparenting behaviors in their partners (Kuo et al., 2017). Although the initial evidence on the relation between masculinity and fathering provides significant insights into contemporary fatherhood, the current literature uses survey reports of father involvement and coparenting. The use of observational data is largely absent from the literature, even though the quality of fathering behavior may be more central to child outcomes than the quantity of involvement (Cabrera et al., 2018). Guided by the resource theory of fathering (Palkovitz & Hull, 2018), this paper helps address these limitations by considering the influence of masculine norm adherence, father role beliefs, and maternal gate closing on observed parenting and coparenting among new fathers in dual-earner families, in which demands for involved fathering may be especially strong.

Resource Theory of Fathering

The resource theory of fathering (Palkovitz & Hull, 2018) contextualizes fathering within physical and social locations to better understand factors associated with various aspects of fathering, including fathers’ involvement with children and as coparents. Resources are defined broadly to include personal characteristics (e.g., demographic factors, role beliefs, personality traits), interpersonal relationships (e.g., the coparenting relationship and social support networks), and contextual characteristics (e.g., family structure, societal values, and social class) associated with how fathers parent their children. These resources shape the health of the fathering axis, which comprises fathers’ ability to manage these resources, fathers’ lived experience (i.e., affect, behavior, and cognitions experienced during interactions), and the father-child relationship. By conceptualizing fathers’ experiences, behaviors, relationships with children within a network of supports, resource theory is broad and includes a diverse range of factors that can help explain both the quality and quantity of paternal engagement with children and how fathers work in concert with coparents to provide for their child’s development. As a result, it may be best to think of it as a heuristic, highly adaptable framework that considers how personal and interpersonal factors help shape men’s parenting and coparenting behaviors (Cabrera et al., 2014). The current study investigates the quality of observed fathering behavior in dual-earner couples (a part of fathers’ contextual resources) as shaped by gendered role beliefs (personal resources) and coparenting quality (interpersonal resources).

The central outcome of resource theory is father-child relationship quality, which comprises both the quality of father-child interactions and the relationship that is built on such interactions (Palkovitz & Hull, 2018). This perspective contrasts with early research on fathering that tended to operationalize fathering as the frequency of paternal involvement (Lamb et al., 1987). The emerging empirical literature clearly supports the shift in focus from the quantity of father involvement to fathering quality; developmentalists such as Pleck (2010) and Cabrera et al. (2018) have demonstrated that the quality of fathers’ parenting behaviors and relationships with children are more strongly associated with child outcomes than the time fathers spend with children. Resource theory and its emphasis on fathering quality is thus an appropriate theory with which to conceptualize the factors that explain variability in father-child relationships.

Considering the increasing emphasis on understanding the quality of fathering behaviors and father-child relationships, Volling and Cabrera (2019) urged researchers to make greater use of observational data. Measures of fathering quality that are derived from observations of father-child interactions, in contrast with survey-based measures, better capture how fathers relate to their children, the quality of these relationships, and how fathers cooperate with mothers to positively impact child wellbeing. Despite the benefits of observational measures, studies employing these methods are still relatively rare when compared to survey-based research on father involvement in childrearing (Palkovitz & Hull, 2018). Fortunately, the number of studies using observational methods is increasing (see Volling & Cabrera, 2019). Much of the research on fathers’ parenting behaviors has been motivated by attachment theory, which focuses on parents’ positive engagement and their sensitivity to their child’s needs. Despite some criticism arguing that applying this “maternal template” to fathers may miss unique aspects of fathers’ optimal parenting (e.g., Paquette, 2004), fathers’ sensitive, engaged, and warm parenting behaviors are consistently associated with higher-quality father-child relationships and positive child social-emotional development (e.g., Malmberg et al., 2016; Mills-Koonce et al., 2015). Following these empirical findings, as well as resource theory’s emphasis on the quality of father-child interactions, we focused on these parenting behaviors in the current study.

Coparenting

Interpersonal resources, including the family system, are important contributors to father-child relationships (Palkovitz & Hull, 2018). The family system is governed by coparents, who operate as the family’s executive subsystem (Van Egeren & Hawkins, 2004). Within the executive subsystem, parents’ interpersonal relationship is distinct from the coparenting relationship (Feinberg, 2003). The interpersonal relationship comprises the romantic and companionate aspects of the relationship, whereas the coparenting relationship comprises solidarity in parental goals and values, equitable sharing of parenting labor, support for the partner’s efforts to accomplish parenting goals, and efforts to undermine the partner’s parenting efforts (Van Egeren & Hawkins, 2004). Notably, resource theory argues that coparenting is both an important interpersonal resource for developing high-quality father-child relationships and an outcome associated with fathers’ contextual and personal resources.

As both an important outcome and a resource that influences fathering behavior, the quality of the coparenting relationship affects children’s social-emotional development over and above parents’ interpersonal relationship (Feinberg, 2003; Teubert & Pinquart, 2010). Consistent with a “father vulnerability” perspective, some have suggested that interparental relationships may shape fathers’ behavior in more powerful ways than mothers’ behavior (Cummings et al., 2010). This has sparked a focus on mothers’ coparenting behaviors towards fathers (Schoppe-Sullivan & Altenburger, 2019). However, fathers’ coparenting behaviors toward mothers also matter. Fathers’ coparenting support is associated with stronger maternal mental health (Mallette et al., 2019), whereas fathers’ triangulation of the child to undermine maternal authority is associated with increases in mothers’ parenting stress (Lau & Power, 2019). Thus, in the current study we included an observational assessment of fathers’ coparenting behavior in addition to fathers’ parenting behavior to capture the extent to which adherence to masculine norms and father nurturing role beliefs affect fathers’ behavior across family relationships. Moreover, our focus on fathers within a sample of dual-earner couples is consistent with Palkovitz and Hull’s (2018, p. 186) call to study the impact of resources by “carefully examin[ing] within-class diversity to nuance our understanding of differences in father-child relationship quality.”

Traditional Masculine Norms

Resource theory argues that paternal identity is a personal resource that shapes how fathers see themselves and how they undertake the various roles and responsibilities associated with father involvement, coparenting relationships, and father-child relationships. As a result, how men view the paternal role represents an important resource that affects their level of involvement, relationship with their child, and their lived experience as a father. Given that resource theory has been created, in part, to generate an organizing framework that integrates other theoretical perspectives on fatherhood, identity theories can help explain how personal resources impact fathers (see Palkovitz & Hull, 2018). Identities shape attitudes, values, and beliefs about how individuals act within various domains of their lives and, as a result, constitute an important personal resource associated with fathering behavior and coparenting quality. Identities are formed when people attach meaning to their statuses and their associated roles. For example, father is a social status with many potential roles associated with it, including provider, protector, caregiver, and nurturer (Palkovitz & Hull, 2018; Rane & McBride, 2000).

Those identities on which individuals place greater importance are more likely to influence attitudes and behaviors than less central identities (Pasley et al., 2014). Even within the same identity, differences in role salience can shape behavior in important ways. For example, fathers who strongly emphasize their role as a provider may spend more time at work, emphasize career advancement, and focus on their family’s economic outlook at the expense of spending time with their partner or children. In contrast, fathers who see themselves as nurturers may be more likely to engage in caregiving, equal coparenting, and acting as an emotional support person—sometimes at the cost of other aspects of the paternal role, like economic provision.

Traditional masculinity represents one identity and, by extension, a personal resource, that is associated with men’s behavior as fathers and coparents (Adamsons & Pasley, 2016; Pasley et al., 2014). These masculine norms and their associated identity are rooted in social structures, socialization, and cultural scripts (Risman, 2004). Strong masculine identity is associated with a variety of attitudes and behaviors, including emotional stoicism and non-egalitarian attitudes, that influence how one acts in various roles (Connell & Messerschmidt, 2005; Levant et al., 1992; Mahalik et al., 2003). Adherence to these norms is associated with diverse outcomes, including physical health (Courtenay, 2011), mental health (Emslie et al., 2006), body image (Gattario et al., 2015), and family violence (Hearn & Morrell, 2012). Within the resource theory framework, masculine norm adherence operates in concert with fathers’ individual and interpersonal resources to influence how they think about the paternal role, behave toward their children, and interact with coparents. Pleck (2010), for example, argued that low identification with traditional masculinity represented an important cognitive resource in men that allowed them to undertake behaviors that are commonly gender-typed as feminine, such as caregiving, emotional support, warmth, and other actions not traditionally aligned with the paternal role. Empirical evidence underscores this point, finding that adherence to traditional masculine norms is negatively associated with contemporary fathering norms such as coequal parenting, caregiving, and emotional availability (Petts et al., 2018).

While overall adherence to a broad set of masculine norms has been linked to fathering behavior (Petts et al., 2018; Shafer et al., 2020), several aspects of masculinity need additional attention because they are associated with both personal and interpersonal resources that influence how mothers and fathers work together as coparents and how fathers engage with their children. Traditional masculine adherence is strongly associated with endorsement of power over women, which can manifest in many ways, including a preference for a highly gendered division of domestic labor, negative feelings about working women and mothers, hostile sexism, and misogynistic attitudes (Glick et al., 2015; Levant et al., 1992; Mahalik et al. 2003). Such attitudes may be associated with decreased interpersonal resources, such as coparenting quality, because men feel little need to work alongside women on tasks that fall outside their narrow beliefs about paternal responsibilities (Bulanda, 2004; Stykes, 2015). Other masculine attitudes and norms, including a strong emphasis on breadwinning, may also elucidate how masculinity represents a resource that shapes fathering behavior. Traditional masculine norms are strongly associated with an emphasis on the provider role and many highly masculine men see their economic contributions to the family as the primary means by which men engage in high-quality fathering (Pleck, 2010). Consistent with resource theory’s conceptualization of personal resources such as identity and attitudes impacting father-child relationships, fathers who emphasize the provider role tend to put their efforts into financial provision, not childcare, emotional support, or the relationship with their child (Macon et al., 2017). While these findings are consistent with a bivariate relationship between fathers’ gendered provider beliefs and involvement in caregiving, it is unclear whether men who emphasize their economic contributions engage in lower quality parenting behavior or are comparatively poor coparents.

Father Nurturing Role Beliefs

In contrast to traditional masculine identities and beliefs, some fathers have integrated progressive and nurturant beliefs about the paternal role into their identities, creating an important personal resource that may be associated with paternal engagement and coequal coparenting. The endorsement of progressive beliefs about the paternal role is positively associated with nurturance and direct care (Gaunt, 2006; Rane & McBride, 2000). These findings are robust across maternal and paternal reports of fathers’ involvement (Beitel & Parke, 1998), over time (Adamsons & Pasley, 2016), and are predictive of the division of childcare among dual-earner families (Jacobs & Kelley, 2006). Observational studies, while less common, corroborate this finding; fathers who hold nurturing role beliefs display greater sensitivity during dyadic interactions with children (NICHD, 2000; Wong et al., 2009). Taken together, these findings are consistent with theoretical models of fathering (e.g., Cabrera et al., 2014; Palkovitz & Hull, 2018; Pleck, 2010), including resource theory, that posit fathering behaviors are shaped, in part, by fathers’ attitudes and beliefs about the paternal role.

Less research has considered whether fathers’ nurturing role beliefs are also associated with better coparenting relationship quality. The limited available evidence suggests that fathers’ expectations about the gendered division of parenting responsibilities are not associated with either reported (Isacco et al., 2010) or observed (Buckley & Schoppe-Sullivan, 2010) coparenting support. However, these studies were limited to perceptions of parental roles and did not investigate adherence to a broader set of norms concerning masculinity, which may be more important to coparenting behavior. In fact, Kuo et al. (2017) reported that fathers in dual-earner couples who endorsed more traditional views on a general measure of gender role attitudes were part of more competitive coparenting relationships.

Maternal Gate Closing

Masculine norms and father role beliefs are not the only resources that matter for paternal behavior and coparenting. Mothers’ behaviors toward fathers in the context of coparenting relationships, often described as “maternal gatekeeping,” are an important interpersonal resource that can influence father-child relationships (Schoppe-Sullivan & Altenburger, 2019). Maternal gatekeeping is rooted in traditional gendered roles in families that emphasize gender-specific parenting responsibilities associated with men’s and women’s parental identities. As an interpersonal resource, mothers can either open (i.e., encourage) or close (i.e., discourage) the gate to fathers’ involvement in parenting and coparenting (Schoppe-Sullivan & Altenburger, 2019). Notably, most research has focused on maternal gate closing behavior and its negative association with men’s engagement with children. Mothers engage in gate closing through mothers’ criticism of the father’s parenting, monopolizing time with the child, and other attitudes and behaviors that actively discourage men from being involved in instrumental and expressive parenting (Cannon et al., 2008). In addition, maternal gate closing behavior seems closely tied to mothers’ attitudes about gender roles (Kulik & Tsoref, 2010) and to fathers’ parenting quality (Altenburger et al., 2018). Thus, in the current study, we focused on the role of maternal gate closing behavior in the quality of fathers’ parenting and coparenting behavior.

The Current Study

The current study investigated associations of fathers’ personal resources—adherence to masculine norms and father nurturing role beliefs—and interpersonal resources—mothers’ gate closing behavior—with the quality of fathers’ observed parenting and coparenting behavior. These associations were investigated within a sample of dual-earner couples, an important contextual resource that shapes parenting according to resource theory. Adherence to masculine norms was assessed based on prenatal measurements of fathers’ masculine agency, hostile sexism, and gendered provider beliefs (Connell & Messerschmidt, 2005; Glick et al., 2015). Maternal gate closing behavior was observed at three months postpartum during mother-father-infant interactions. At 9 months postpartum, fathers’ parenting and coparenting behavior was observed during separate semi-structured interactions (dyadic and triadic, respectively).

Given associations between fathers’ parenting quality and education (Tamis-LeMonda et al., 2004) and between adherence to masculine norms and religiosity (Shafer et al., 2019), fathers’ education and religiosity were controlled in analyses. Also, given that interpersonal relationship quality predicts coparenting quality (Schoppe-Sullivan & Mangelsdorf, 2013), analyses controlled for the prenatal quality of the couple relationship. Net of these controls, we hypothesized that: (1) Higher levels of fathers’ masculine agency, hostile sexism, and gendered provider beliefs would be associated with fathers’ poorer-quality parenting and coparenting behavior; (2) Fathers’ greater endorsement of nurturing father role beliefs would be associated with fathers’ higher-quality parenting and coparenting; (3) Higher levels of maternal gate closing behavior would be associated with fathers’ poorer-quality parenting and coparenting behavior.

Method

Participants and Procedure

Data came from a study of 182 different-sex dual-earner couples assessed during the third trimester of pregnancy and when their first-born child was 3, 6, and 9 months old. This study was approved by Ohio State University’s Institutional Review Board (Project title: New Parents Project; Protocol number: 2007B0228). Couples were recruited via childbirth education classes, newspaper ads, snowball sampling, and word-of-mouth. Eligibility criteria included: married or cohabiting, expecting first biological child, both partners working full time and planning to return to work post-birth, able to read/speak English, and at least 18 years old. Of expectant fathers, 85% identified as White/European American, 7% as Black/African American, 4% as Asian American or Pacific Islander, 3% as other race, and 1% as mixed race; 2% identified as Hispanic/Latin American. Expectant fathers’ ages ranged from 18 to 50 years (M = 30.20, SD = 4.81) and 65% had earned at least a bachelor’s degree. Of expectant mothers, 85% identified as White/European American, 6% as Black/African American, 3% as Asian American, 2% as other race, and 4% as mixed race; 4% identified as Hispanic/Latin American. Expectant mothers’ ages ranged from 18 to 42 years (M = 28.24, SD = 4.02) and 75% had earned at least a bachelor’s degree. Households had a median income of 78,217 USD and 86% of the couples were married.

We focused on surveys completed by expectant fathers and observations of couple behavior during the third trimester, observations of maternal gatekeeping behavior at 3 months postpartum, and observations of fathers’ parenting and coparenting behavior at 9 months postpartum. Expectant fathers completed surveys online or via paper-and-pencil during the third trimester. Two weeks later, couples participated in a home visit that included a 10-min video recorded observational assessment of couple behavior in which partners discussed a relationship problem and tried to reach a solution. At 3 months postpartum, couples participated in a home visit together with their infants, which included a 5-min video recorded mother-father-infant interaction in which they were asked to play together with their baby using an infant jungle gym. At 9 months postpartum, parents completed an observational assessment at a local science center (67.3%) or at home. These assessments included 5-min video recorded parent-infant interactions in which each parent was asked to teach their child how to use a new toy (shape sorter or ring stacker). The order of mother- and father-child interactions and toy selection were randomized. Parents also completed two 5-min video recorded mother-father-infant interactions in which they were asked to introduce two toys (jack-in-the-box and pop-up toy) to their infants together.

Missing data at the third trimester and three months postpartum was minimal, ranging from 0–8% on variables of interest. At 9 months, 153 of the 182 families (84%) participated in the observational assessment. Comparisons of families who participated at 9 months to those who did not on demographic, predictor, and control variables yielded no significant differences.

Measures

Masculine Agency.

Expectant fathers were asked to rate themselves (1 = not at all like me; 4 = very much like me) on a list of seven masculine stereotypical agentic characteristics derived from Diekman and Eagly (2000): competitive, daring, adventurous, dominant, aggressive, courageous, stands up under pressure. Cronbach’s alpha was .74.

Hostile Sexism.

Expectant fathers completed the Hostile Sexism subscale of the Ambivalent Sexism Inventory (Glick & Fiske, 1996; α = .90). This required expectant fathers to respond to 11 items using a 6-point scale (0 = disagree strongly; 5 = agree strongly). Example items included “Most women interpret innocent remarks or acts as being sexist” and “Feminists are making unreasonable demands of men.” Higher scores reflected greater hostile sexism.

Gendered Provider Beliefs.

Two items from the adapted Roles Questionnaire (Perry-Jenkins et al., 1992) were used to assess fathers’ gendered provider beliefs. The first item asked expectant fathers to answer this question: With reference to your own family, who do you feel should provide the income?” Coded response choices included: 4 = man entirely, 3 = man more than woman, 2 = man and woman about the same, 1 = woman more than man. There was an option for “woman entirely,” but no respondents selected this option. There was also an “other” option; those responses were coded as missing. The second item asked expectant fathers, “In general, do you think a woman should work if her partner makes an income equal to what the man in your relationship makes?” Coded response choices included: 4 = her primary responsibility is the care of the family and the home, 3 = only if the woman really wants to work, 2 = it would be better in most circumstances for her to work, 1 = it is her duty to work. Expectant fathers’ responses to these two items were correlated at r = .29, p < .01, and were averaged to create an indicator for which higher scores reflected more gendered provider beliefs.

Nurturing Father Role Beliefs.

Expectant fathers’ responses to 9 items from the Beliefs Concerning the Parental Role scale (Bonney & Kelley, 1996) were selected to represent nurturing father beliefs (α = .70). Items were rated on a 5-point scale (1 = disagree strongly; 5 = agree strongly). Item examples included “Men should share with child care such as bathing, feeding, and dressing the child” and “A father should be emotionally involved with his children.” Items were averaged and higher scores indicated stronger nurturing father beliefs.

Mothers’ Gate Closing Behavior.

The mother-father-infant interactions at 3 months postpartum were coded by a team of trained raters for mothers’ gate closing behavior using three 5-point rating scales (see Cannon et al., 2008, and Altenburger et al., 2014): maternal negative control, displeasure, and coldness. Maternal negative control reflected the extent to which the mother made verbal and nonverbal attempts to limit the father’s interaction with the infant, displeasure reflected the degree to which mothers displayed dislike for the father’s style of interacting with or relationship with the infant, and coldness reflected the extent to which the mother exhibited distance, snubbing, or disdain from/for the father. The coding team double-coded 50.6% of the episodes. Gamma statistics were .80 for maternal negative control, .81 for displeasure, and .81 for coldness, reflecting acceptable interrater reliability. Scores on these three scales were used as indicators of a “maternal gate closing” latent variable.

Fathers’ Positive Parenting Behavior.

The observations of father-infant interaction at 9 months were coded by a team of two trained raters who used a series of 5-point scales including sensitivity, positive regard, and detachment (Cox & Crnic, 2002). All episodes were double-coded, and gammas were .77 for fathers’ sensitivity, .83 for fathers’ positive regard, and .70 for fathers’ detachment. Coders’ scores were averaged to yield a single score for each dimension, and these scores served as indicators of a “fathers’ observed positive parenting” latent variable.

Fathers’ Coparenting Behavior.

The observations of mother-father-infant interaction at 9 months postpartum were coded by a different team of trained raters who used a series of 5-point scales to assess coparenting behavior (see Altenburger et al., 2014 and Schoppe-Sullivan et al., 2004). The pleasure (degree to which father enjoyed collaborating in parental role and watching mother parent the infant) and warmth (extent to which father demonstrated affection and emotional support to mother) scales reflected fathers’ supportive coparenting behavior, whereas the displeasure (degree to which fathers displayed dislike for mother’s style of parenting or relationship with the infant) and coldness (extent to which father exhibited distance, snubbing, or disdain from/for mother) scales reflected fathers’ undermining coparenting behavior. The coders double-coded 56.2% of the episodes, and gammas ranged from .63 to .91. The separate ratings coders made of fathers’ pleasure and warmth in the jack-in-the-box (Episode 1) and pop-up toy (Episode 2) tasks indicated a latent variable representing fathers’ supportive coparenting behavior, whereas the ratings of fathers’ displeasure and coldness in Episodes 1 and 2 indicated a latent variable representing fathers’ undermining coparenting behavior.

Covariates.

Expectant fathers’ education was an 8-point ordinal scale from 1 = less than high school to 8 = Doctorate or equivalent, and was dichotomized as 1 = bachelor’s degree or greater (65%) and 0 = less than a bachelor’s degree (35%). Expectant fathers’ religiosity was measured using the following item: How important is religion to you? (1 = very important; 4 = not important at all). We recoded this item so higher scores indicated greater religiosity. The observational assessments of couple behavior from the third trimester were coded by another team of trained research assistants (see Altenburger et al., 2014, for details) using 5-point scales reflecting negative escalation, cohesiveness, and individual problem solving communication (separately rated for expectant mothers and fathers), and a 7-point scale of global interaction quality, which were used as indicators of an “observed couple behavior” latent variable in subsequent analyses. Both coders rated each episode and ICCs ranged from .79 to .89.

Analysis Plan

We first examined descriptive statistics and intercorrelations. Next, we employed structural equation modeling in IBM SPSS AMOS 25.0 to test a measurement model for the latent variables and subsequently a structural model to examine our hypotheses. FIML was used to estimate missing data. Criteria for good fitting models included: non-significant chi-square tests, RMSEA values < .06, and Comparative Fit Index values > .95 (Hu & Bentler, 1999).

Results

Preliminary Analyses

Descriptive statistics and intercorrelations are presented in Tables 1 and 2. Overall, these dual-earner fathers scored relatively high on nurturing father beliefs and on masculine agency, and at about the midpoint on gendered provider beliefs and hostile sexism. These fathers and their partners together demonstrated positive couple behavior, characterized by high global quality and cohesiveness, low negative escalation, and moderate levels of problem solving. Mothers demonstrated low levels of observed gate closing behavior. Regarding fathers’ parenting behavior, fathers were on average moderately sensitive and somewhat lower in positive regard and detachment. Fathers showed moderate to high levels of supportive coparenting and low levels of undermining coparenting behavior.

Table 1.

Descriptive Statistics and Intercorrelations among Predictor Variables.

Predictor	M	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. F Agency	2.75	0.53
2. F Hostile Sexism	2.68	0.94	.19*
3. F Gend Provider	2.47	0.51	−.03	.08
4. F Nurt Father	4.67	0.37	.01	−.21**	−.07
5. M Neg Control	1.58	0.82	.06	.03	.07	.00
6. M Displeasure	1.26	0.58	.04	.07	−.02	−.02	.33**
7. M Coldness	1.14	0.40	−.02	−.02	−.11	.08	.22**	.30**
8. F Religiosity	2.86	1.06	−.02	−.02	.22**	−.06	−.03	−.03	.11
9. F Education	0.65	0.48	−.10	−.09	.09	−.10	.06	.06	−.20**	.11
10. Global	5.04	1.36	−.11	−.09	.15	−.09	−.05	−.06	−.01	.16*	.21**
11. Cohesiveness	3.59	0.88	−.09	−.09	.10	−.09	−.03	−.08	.01	.15*	.19*	.87**
12. Neg Escalation	1.47	0.87	−.01	.01	−.11	.01	−.02	.11	−.02	−.12	.02	−.60**	−.48**
13. M Prob Solving	3.22	0.91	−.13	−.19*	.07	.00	.15	−.06	.11	.22**	.21**	.60**	.58**	−.34**
14. F Prob Solving	3.03	1.09	−.12	−.17*	.10	.04	.00	−.04	−.01	.21**	.23**	.70**	.67**	−.38**	.61**

Note. F = Father; Gend = Gendered; Nurt = Nurturing; M = Mother; Neg = Negative; Prob = Problem.

Ns ranged from 167 to 182.

^*p < .05

^**p < .01

Table 2.

Descriptive Statistics and Intercorrelations among Outcome Variables.

Predictor	M	SD	1	2	3	4	5	6	7	8	9	10	11
1. F Sensitivity	3.12	0.55
2. F Pos Regard	2.78	0.77	.73**
3. F Detachment	2.68	0.62	−.71**	−.63**
4. F Pleasure 1	3.75	1.03	.37**	.36**	−.28**
5. F Warmth 1	3.63	1.03	.39**	.38**	−.30**	.88**
6. F Pleasure 2	3.38	1.05	.38**	.35**	−.35**	.69**	.69**
7. F Warmth 2	3.36	1.01	.40**	.36**	−.39**	.65**	.70**	.91**
8. F Displeasure 1	1.60	0.89	−.21*	−.10	.12	−.50**	−.46**	−.35**	−.34**
9. F Coldness 1	1.40	0.75	−.28**	−.25**	.20*	−.63**	−.62**	−.52**	−.52**	.63**
10. F Displeasure 2	1.53	0.87	−.19*	−.23**	.17*	−.38**	−.39**	−.53**	−.51**	.38**	.54**
11. F Coldness 2	1.40	0.75	−.36**	−.35**	.34**	−.51**	−.52**	−.64**	−.65**	.40**	.72**	.66**

Note. F = Father; Pos = Positive; 1 = Episode 1; 2 = Episode 2.

N = 153.

^*p < .05

^**p < .01

Preliminary correlations between the predictor and outcome variables are presented in Table 3. Significant associations included those between fathers’ greater masculine agency and lower parenting detachment, fathers’ greater hostile sexism and fathers’ lower coparenting warmth and higher displeasure, greater nurturing father role beliefs and fathers’ greater parenting sensitivity and lower coparenting coldness. Greater maternal coldness—an aspect of gate closing behavior—was associated with fathers’ lower parenting sensitivity and fathers’ less supportive and more undermining coparenting behavior, and greater maternal displeasure was positively associated with fathers’ displeasure in coparenting. There were many significant associations between indicators of observed couple behavior and fathers’ parenting and coparenting.

Table 3.

Correlations between Predictor Variables and Fathers’ Parenting and Coparenting.

Predictor	F Sens	F Pos Reg	F Detach	F Pleas 1	F Warm 1	F Pleas 2	F Warm 2	F Displ 1	F Cold 1	F Displ 2	F Cold 2
1. F Agency	.08	.16	−.18*	−.02	−.03	.10	.03	.10	−.08	−.01	−.12
2. F Hostile Sexism	−.11	.00	.04	−.09	−.11	−.15	−.17*	.19*	.14	.24**	.19*
3. F Gend Provider	−.04	−.01	.00	.09	.07	−.02	−.02	−.15	−.08	−.03	.01
4. F Nurt Father	.20*	.14	−.14	.08	.02	.04	.09	−.11	−.21*	−.13	−.25**
5. M Neg Control	−.15	−.09	.05	−.02	−.01	.01	.03	−.01	−.03	−.08	−.04
6. M Displeasure	−.13	−.09	.11	−.12	−.09	−.03	−.06	.27**	.10	.08	.09
7. M Coldness	−.18*	−.08	.10	−.16	−.20*	−.19*	−.16	.19*	.10	.04	.14
8. F Religiosity	−.09	−.08	.03	−.01	−.01	−.05	−.04	.03	.06	.04	.05
9. F Education	.16*	.12	−.12	−.06	−.01	.14	.16*	.03	.07	−.03	−.12
10. Global	.15	.19*	−.09	.30**	.36**	.18*	.24**	−.10	−.22*	−.19*	−.15
11. Cohesiveness	.14	.20*	−.10	.28**	.37**	.21*	.26**	−.10	−.26**	−.20*	−.19*
12. Neg Escalation	−.06	−.08	.04	−.20*	−.20*	−.08	−.09	.16	.10	.12	.08
13. M Prob Solving	−.05	.01	.08	.15	.18*	.07	.11	−.01	−.14	−.11	−.07
14. F Prob Solving	.28**	.29**	−.18*	.35**	.36**	.20*	.24**	−.12	−.21*	−.22**	−.21*

Note. F = Father; Gend = Gendered; Nurt = Nurturing; M = Mother; Neg = Negative; Prob = Problem; Pos Reg = Positive Regard; Detach = Detachment; Pleas = Pleasure; Warm = Warmth; Displ = Displeasure; Cold = Coldness; 1 = Epiosde 1; 2 = Episode 2.

Ns ranged from 139 to 153.

^*p < .05

^**p < .01

Intercorrelations among the observed couple behavior indicators, the mothers’ observed gate closing indicators, the fathers’ observed positive parenting indicators, and the fathers’ observed supportive and undermining coparenting indicators (Tables 1 and 2) supported the construction of the latent variables. A measurement model containing these variables was tested, including covariances between all latent variables, and covariances between the error terms for the following pairs of indicators: mothers’ and fathers’ problem solving, fathers’ pleasure and warmth coded within the same 5-min episode, and fathers’ displeasure and coldness coded within the same 5-min episode. The model fit the data adequately (chi-square = 191.21, df = 133, p = .001; CFI = .968; RMSEA = .049), and all factor loadings were significant with standardized loadings of |.47| or greater.

Structural Model Testing

Next, we tested an initial structural model in which the covariates fathers’ religiosity and education were included, and these variables together with fathers’ masculine agency, hostile sexism, gendered provider beliefs, and nurturing father role beliefs (third trimester), and latent variables representing observed couple behavior (third trimester) and mothers’ observed gate closing behavior (3 months postpartum) were modeled as predictors of fathers’ observed positive parenting and fathers’ observed supportive and undermining coparenting behavior at 9 months postpartum. Covariances between all exogenous variables were estimated, as was the covariance between the disturbance terms for observed supportive and undermining coparenting. Structural model fit was adequate according to RMSEA but did not meet criteria for CFI or chi-square test (chi-square = 334.26, df = 219, p = .000; CFI = .940; RMSEA = .054). Unstandardized and standardized coefficients, standard errors, and associated p-values for all factor loadings, structural path coefficients, and covariances for this initial model are presented in Table 4, with significant parameter estimates bolded for emphasis.

Table 4.

Coefficients and Significance Levels for Full SEM Model.

Parameter Estimate	Unstandardized (SE)	Standardized	p
Factor Loadings
Father’s Positive Parenting → Sensitivity	1.00	.92	n/a
Father’s Positive Parenting → Positive Regard	1.22 (.11)	.80	.00
Father’s Positive Parenting → Detachment	−.94 (.09)	−.77	.00
Father’s Supportive Coparenting → Pleasure 1	1.00	.83	n/a
Father’s Supportive Coparenting → Warmth 1	1.05 (.06)	.86	.00
Father’s Supportive Coparenting → Pleasure 2	1.01 (.10)	.82	.00
Father’s Supportive Coparenting → Warmth 2	.97 (.10)	.82	.00
Father’s Undermining Coparenting → Displeasure 1	1.00	.53	n/a
Father’s Undermining Coparenting → Coldness 1	1.37 (.18)	.85	.00
Father’s Undermining Coparenting → Displeasure 2	1.20 (.22)	.64	.00
Father’s Undermining Coparenting → Coldness 2	1.37 (.22)	.86	.00
Observed Couple Behavior → Global Quality	1.00	.98	n/a
Observed Couple Behavior → Cohesiveness	.60 (.03)	.90	.00
Observed Couple Behavior → Negative Escalation	−.39 (.04)	−.60	.00
Observed Couple Behavior → Mother Problem Solving	.43 (.05)	.62	.00
Observed Couple Behavior → Father Problem Solving	.60 (.05)	.73	.00
Maternal Gate Closing → Negative Control	1.00	.42	n/a
Maternal Gate Closing → Displeasure	.89 (.29)	.53	.00
Maternal Gate Closing → Coldness	.64 (.21)	.56	.00
Structural Path Coefficients
Masculine Agency → Father’s Positive Parenting	.19 (.08)	.20	.02
Masculine Agency → Father’s Supportive Coparenting	.17 (.14)	.11	.21
Masculine Agency → Father’s Undermining Coparenting	−.17 (.08)	−.19	.04
Hostile Sexism → Father’s Positive Parenting	−.02 (.05)	−.03	.74
Hostile Sexism → Father’s Supportive Coparenting	−.10 (.08)	−.11	.22
Hostile Sexism → Father’s Undermining Coparenting	.07 (.05)	.15	.11
Gendered Provider Beliefs → Father’s Positive Parenting	−.05 (.08)	−.05	.59
Gendered Provider Beliefs → Father’s Supportive Coparenting	.00 (.15)	.00	.98
Gendered Provider Beliefs → Father’s Undermining Coparenting	−.04 (.08)	−.04	.65
Father’s Nurturing Role Beliefs → Father’s Positive Parenting	.30 (.12)	.22	.01
Father’s Nurturing Role Beliefs → Father’s Supportive Coparenting	.10 (.20)	.04	.61
Father’s Nurturing Role Beliefs → Father’s Undermining Coparenting	−.33 (.12)	−.26	.01
Maternal Gate Closing → Father’s Positive Parenting	−.49 (.20)	−.34	.01
Maternal Gate Closing → Father’s Supportive Coparenting	−.73 (.33)	−.30	.03
Maternal Gate Closing → Father’s Undermining Coparenting	.30 (.17)	.22	.09
Observed Couple Behavior → Father’s Positive Parenting	.08 (.03)	.22	.01
Observed Couple Behavior → Father’s Supportive Coparenting	.25 (.06)	.38	.00
Observed Couple Behavior → Father’s Undermining Coparenting	−.09 (.03)	−.26	.01
Father’s Education → Father’s Positive Parenting	.14 (.09)	.13	.12
Father’s Education → Father’s Supportive Coparenting	−.07 (.15)	−.04	.65
Father’s Education → Father’s Undermining Coparenting	.02 (.08)	.02	.78
Father’s Religiosity → Father’s Positive Parenting	−.05 (.04)	−.10	.26
Father’s Religiosity → Father’s Supportive Coparenting	−.07 (.07)	−.09	.30
Father’s Religiosity → Father’s Undermining Coparenting	.04 (.04)	.08	.34
Covariances
Masculine Agency ↔ Hostile Sexism	.09 (.04)	.18	.02
Masculine Agency ↔ Gendered Provider Beliefs	−.01 (.02)	−.02	.76
Masculine Agency ↔ Father Nurturing Role Beliefs	.00 (.02)	.02	.84
Masculine Agency ↔ Maternal Gate Closing	.01 (.02)	.05	.67
Masculine Agency ↔ Observed Couple Behavior	−.07 (.06)	−.10	.20
Masculine Agency ↔ Father’s Education	−.03 (.02)	−.10	.19
Masculine Agency ↔ Father’s Religiosity	−.01 (.04)	−.02	.83
Hostile Sexism ↔ Attachment Avoidance	.18 (.05)	.28	.00
Hostile Sexism ↔ Gendered Provider Beliefs	.03 (.04)	.07	.40
Hostile Sexism ↔ Father Nurturing Role Beliefs	−.07 (.03)	−.21	.01
Hostile Sexism ↔ Maternal Gate Closing	.01 (.04)	.03	.76
Hostile Sexism ↔ Observed Couple Behavior	−.11 (.10)	−.09	.29
Hostile Sexism ↔ Father’s Education	−.04 (.04)	−.09	.24
Hostile Sexism ↔ Father’s Religiosity	−.02 (.08)	−.02	.83
Gendered Provider Beliefs ↔ Father Nurturing Role Beliefs	−.01 (.01)	−.07	.37
Gendered Provider Beliefs ↔ Maternal Gate Closing	−.01 (.02)	−.08	.49
Gendered Provider Beliefs ↔ Observed Couple Behavior	.09 (.06)	.14	.09
Gendered Provider Beliefs ↔ Father’s Education	.02 (.02)	.09	.24
Gendered Provider Beliefs ↔ Father’s Religiosity	.12 (.04)	.22	.01
Father Nurturing Role Beliefs ↔ Maternal Gate Closing	.01 (.01)	.05	.62
Father Nurturing Role Beliefs ↔ Observed Couple Behavior	−.05 (.04)	−.10	.22
Father Nurturing Role Beliefs ↔ Father’s Education	−.02 (.01)	−.09	.27
Father Nurturing Role Beliefs ↔ Father’s Religiosity	−.02 (.03)	−.06	.47
Maternal Gate Closing ↔ Observed Couple Behavior	−.03 (.05)	−.07	.55
Maternal Gate Closing ↔ Father’s Education	−.01 (.02)	−.08	.42
Maternal Gate Closing ↔ Father’s Religiosity	.02 (.04)	.04	.68
Observed Couple Behavior ↔ Father’s Education	.14 (.05)	.22	.01
Observed Couple Behavior ↔ Father’s Religiosity	.26 (.11)	.19	.02
Father’s Education ↔ Father’s Religiosity	.05 (.04)	.11	.16
Disturbance terms for Supportive and Undermining Coparenting	−.24 (.05)	−.79	.00

Note. Chi-square = 334.26, df = 219, p = .000; CFI = .940; RMSEA = .054. Bolded text indicates significant estimates.

In this initial model, fathers’ religiosity, education, hostile sexism, and gendered provider beliefs were not significantly associated with fathers’ observed parenting or coparenting behavior. Thus, we tested a trimmed model that eliminated these predictors. The trimmed model met two out of three of our fit criteria (chi-square = 250.08, df = 163, p = .000; CFI = .953; RMSEA = .054) and is depicted in Figure 1. In this model, fathers showed more positive parenting behavior at 9 months postpartum when the couple demonstrated more positive behavior during the third trimester of pregnancy, when fathers had greater masculine agency and held greater father nurturing role beliefs during the third trimester of pregnancy, and when mothers showed less gate closing behavior at 3 months postpartum. Fathers showed more supportive coparenting behavior at 9 months postpartum when the couple demonstrated more positive behavior during the third trimester of pregnancy and when mothers showed less gate closing behavior at 3 months postpartum. Finally, fathers were observed to undermine their partners more at 9 months postpartum when the couple demonstrated poorer behavior during the third trimester of pregnancy and when expectant fathers less strongly endorsed nurturing father role beliefs. Predictors accounted for 23% of the variance in fathers’ observed positive parenting, 22% of the variance in fathers’ observed supportive coparenting, and 21% of the variance in fathers’ observed undermining coparenting.

Figure 1.

Trimmed structural model depicting significant associations between predictors and fathers’ observed parenting and coparenting.

Chi-square = 250.08; df = 163; CFI = .953 RMSEA = .054. Path coefficients are standardized estimates. Covariances, errors, and disturbance terms not shown. F = Father; M = Mother; Nurt = Nurturing; N = Negative; Pos = Positive; 1 = Episode 1; 2 = Episode 2. *p < .05 **p < .01.

Discussion

Contextualizing the unique nature of fathering is critical in efforts to foster thriving father-child, and coparenting, relationships. We used longitudinal data and observational methods with a sample of dual-earner families to conduct a multidimensional test of resource theory (Palkovitz & Hull, 2018), which posits that resources including personal characteristics, interpersonal relationships, and contextual characteristics shape fathering. Beginning with personal characteristics, our results suggest that men who reported greater nurturing father role beliefs prenatally showed higher-quality interactions with their 9-month-old children. Moreover, when parenting together with their child’s mother, men who had more nurturing fathering role beliefs were less likely to undermine when coparenting. Overall, these findings support prior work demonstrating the predictive power of role beliefs on fathers’ behavior (Beitel & Parke, 1998; NICHD, 2000).

Although masculine norms continue to be strongly associated with fathering (Shafer et al., 2020), recent evidence suggests that fathers in the United States are shifting away from a traditional masculine identity that prioritizes work; in a 2011 survey, only one in six fathers reported being interested primarily in work, while four in six reported that work “is only a small part” of their identity (Harrington et al., 2011). This reflects a marked shift away from the father-as-breadwinner norms that restricted father involvement in childrearing through the 19^th and early 20^th centuries (Pleck & Pleck, 1997). Those traditional masculine norms, instilled from very early ages, taught men to be less emotional, more individualistic, and to avoid typically “feminine” behaviors (Connell & Messerschmidt, 2005; Mahalik et al., 2003). The dual-earner fathers in our study who had more progressive beliefs about fathers’ roles were indeed more sensitive and positive and less detached with their infants. The stoic male norm does not lend itself to parental sensitivity, and these fathers’ rejection of that norm in favor of more nurturing attitudes will benefit their children’s socioemotional development (e.g., Malmberg et al., 2016; Mills-Koonce et al., 2015). Further, expectant fathers who reported more nurturing father role beliefs were also less likely to undermine their child’s mother postpartum. Kuo et al. (2017) similarly reported that fathers who endorsed less traditional gender role attitudes were members of less competitive coparenting relationships. As ideals and expectations of fathering continue to evolve and men see a full range of sensitive, emotional, and positive parenting behavior as increasingly acceptable and comfortable, children and coparents will benefit. One potential avenue for future research is examining whether interventions targeting positive paternal nurturing attitudes will improve the quality of father-child and coparenting relationships.

In contrast to our hypotheses, we found that dual-earner fathers that identified more closely with stereotypically masculine agentic characteristics such as being daring, dominant, and aggressive showed more positive parenting behaviors. This finding stands in contrast to prior research and theory, based primarily on survey data, indicating that highly masculine orientations are associated with less engaged fathering (Petts et al., 2018; Pleck, 2010). It is worth noting that the agentic characteristics we examined, although stereotypically masculine, are perceived as highly positive (Diekman & Eagly, 2000). Future work should further explore whether contemporary fathers, particularly those in dual-earner families, combine traditional aspects of masculinity with new nurturing ideals of fathering to create new fathering identities. Perhaps hostile sexism and gendered provider beliefs were not associated with fathering and coparenting behavior in our model because American fathers, at least highly educated fathers in dual-earner couples, are in the midst of transforming fatherhood, and our conceptualization and measurement of key components of masculinity may need to be reexamined. This transformation of fatherhood needs additional qualitative, quantitative, and mixed-method research. Because most studies use unidimensional measures of masculinity, a qualitative investigation of contemporary masculinity and its dimensions is warranted and a revision or brand-new conceptualization or theory of contemporary masculinity may be necessary. Quantitative and mixed methods scholars could build on these insights to create new measures of contemporary masculinity and its dimensions. Further, a novel application of resource theory to masculinity could identify the contexts and resources that foster, or subvert, contemporary fatherhood.

Turning to resources in fathers’ interpersonal relationships, the couples’ preexisting relationship was also a key predictor of these fathers’ observed parenting and coparenting behavior in dual-earner families. Our findings support resource theory (Palkovitz & Hull, 2018) in that the parents’ observed positive couple relationship behaviors were a key resource supporting positive fathering and coparenting. Couples who exhibited more positive relationship behaviors before their children were born included fathers who showed more positive parenting behavior, more supportive coparenting behavior, and less undermining coparenting nine-months later. The quality of the coparents’ relationship, as the executive subsystem of the family (Van Egeren & Hawkins, 2004), is a key aspect of the family system and is associated with both child and parent wellbeing (Lau & Power, 2019; Mallette et al., 2019; Teubert & Pinquart, 2010). Several transition to parenthood programs focus on supporting and improving the couples’ interpersonal relationship, and our findings support the critical importance of these efforts for the development of subsequent father-child and coparenting relationships (Cowan et al., 2006).

Maternal behavior, another interpersonal resource, also mattered for fathering and the quality of the coparenting relationship. Maternal gate closing behavior, including criticizing the fathers’ parenting, restricting his interactions with the infant, and showing displeasure and coldness, is rooted in women’s and men’s gendered parental identities (Kulik & Tsoref, 2010; Schoppe-Sullivan & Altenburger, 2019). When mothers closed the gate to fathers in the current study, fathers showed less positive parenting behavior toward their infants, and, importantly, reciprocated with less supportive coparenting behavior towards mothers at 9 months postpartum. Although prior research demonstrated that fathers’ perceptions of maternal gate closing were linked to poorer-quality fathering behavior (Altenburger et al., 2018), this study is the first to link observed maternal gate closing behavior to less positive observed parenting behavior by fathers. This suggests that women’s own adherence to mothering norms may have broader detrimental effects on the family system that are not limited to fathers’ lower involvement in parenting. Even as fathering norms have changed, expectations for mothers have never been higher, more demanding, or more time intensive (Bianchi et al., 2006). Although on the one hand dual-earner mothers may want and need greater involvement in parenting from fathers, they may also have trouble letting go of the ideal mother role and may engage in gatekeeping in order to protect their identities as mothers. Evidence for this was provided by Pedersen and Kilzer (2014), who found that among dual-earner mothers greater work-to-family conflict was associated with higher levels of maternal gate closing. Thus, greater attention to maternal gatekeeping, its antecedents, and its consequences, is needed. Our previous research suggests that fathers’ as well as mothers’ characteristics are associated with new parents’ perceptions of maternal gatekeeping (Schoppe-Sullivan et al., 2015), hence future research should examine both fathers’ and mothers’ characteristics to better understand the links between maternal gate closing and family processes.

This study was not without limitations. Results based on this primarily highly educated, non-Hispanic white, different-sex, Midwestern U.S. sample of dual-earner families may not generalize to other populations or contexts, especially to fathers in single-earner families, who may have greater adherence to traditional masculine norms, such as gendered provider beliefs. Some of the measures we used had lower reliability than ideal; the alpha for our measure of fathers’ nurturing role beliefs was relatively low and some of the gammas for fathers’ coparenting behaviors were in the lower range of acceptability. In addition, we only examined fathers’ adherence to traditional masculinity and father nurturing role beliefs in the third trimester of pregnancy. Thus, we were unable to examine shifts in adherence to masculine norms and fathering ideals that may occur when men find out they are going to become fathers, or across the transition to parenthood. Future work examining these constructs across the transition to pregnancy, across the transition to parenthood, and in more diverse populations of parents, is warranted. Furthermore, mothers’ attitudes about gender roles may also influence maternal gatekeeping, father-child, and coparenting relationships (Beitel & Parke, 1998; Kulik & Tsoref, 2010; Schoppe-Sullivan & Mangelsdorf, 2013), but we were unable to include those additional predictors due to limits on model complexity. Future research should consider the roles of maternal as well as paternal attitudes about gender roles in the quality of family relationships. Finally, our focus on fathers’ sensitive, engaged, and warm parenting behavior in the current study may have left out other important father behaviors. Theory and research suggest that there are other behaviors, like activative fathering or challenging parenting behavior (Majdandžić et al., 2018; Paquette, 2004; Volling et al., 2019), that fathers may specialize in and may promote children’s development. Thus, future research should consider how aspects of traditional masculinity, father role beliefs, and maternal gatekeeping are associated with a broader range of fathers’ parenting behaviors.

In conclusion, consistent with resource theory (Palkovitz & Hull, 2018), these results suggest that expectant fathers in dual-earner families with nurturing father role beliefs showed better parenting and coparenting behavior after their child’s birth. Mothers also played a role; when mothers showed less gate closing behavior toward fathers and had better interpersonal relationships with fathers, fathers subsequently showed more positive parenting and coparenting behavior. Given the importance of high-quality fathering and coparenting for child and family development (Malmberg et al., 2016; Schoppe-Sullivan et al., 2004; Teubert & Pinquart, 2010), understanding how to help fathers develop nurturing father role beliefs so that they can create a supportive relationship with their child and coparent is critical. Moreover, working with mothers to support the development of positive father-child relationships is also imperative. Yet, it is also important to recognize that fathers in our sample who held more agentic identities also had more positive fathering behavior, highlighting that contemporary masculinity is multi-faceted and deserves further research to more fully understand its role in fathers’ relationships with children and partners.

Public significance statement:

Traditional masculinity, father role beliefs, and maternal behavior are all important to new fathers’ relationships with children and partners in dual-earner families. Given the importance of high-quality fathering and coparenting for child and family functioning, supporting fathers’ nurturing role beliefs and reducing maternal gate closing are key goals.