Interparental conflict and mindful parenting practices: transactional effects between mothers and fathers

Objective: This study examined the transactional relations between mothers ’ and fathers ’ destructive interparental conflict tactics and mindful parenting practices. Background: According to family systems theory and spill-over hypothesis, interparental conflict sets the stage for parents ’ future levels of conflict as well as parenting difficulties. However, research on the link between conflict and mindful parenting practices remains scarce. The present study aims to investigate the longitudinal relations between mothers ’ and fathers ’ interparental conflict tactics and mindful parenting practices. Method: Participants were 386 families involving mothers and fathers of adolescent children at 12 – 17 years old. Mothers and fathers completed questionnaire reports of destructive interparental conflict and mindful parenting at two time points spanning 12 months apart. A structural equation model was conducted to examine the relations between conflict tactics and mindful parenting practices. Results: Findings indicated that mothers ’ and fathers ’ destructive conflict predicted their own and their spouses ’ subsequent destructive conflict, as well as mindful parenting practices. Mothers ’ and fathers ’ mindful parenting predicted their own mindful parenting longitudinally, but did not predict the other variables. Conclusion: Supporting the spillover hypothesis, this study revealed the longitudinal effect of interparental conflict on mindful parenting, regardless of parent


INTRODUCTION
According to the spillover hypothesis, the negativity associated with destructive and unresolved interparental conflict carries over to other family processes, such as childrearing consistency, parental responsiveness, and negativity and rejection toward children (Erel & Burman, 1995;Katz & Gottman, 1996;Krishnakumar & Buehler, 2000;Y. Li et al., 2011;Sturge-Apple et al., 2009;Warmuth et al., 2020).As indexed by behaviors including verbal and physical aggression, avoidance-capitulation, child involvement, and stonewalling, destructive conflict in the interparental relationship not only sets the stage for future levels of conflict (Boker & Laurenceau, 2006;Cheung et al., 2016), but may also result in parents' difficulties in parenting their children mindfully.
Mindful parenting refers to an integration of mindfulness into the practice of parenting by providing nonjudgmental, intentional, and present-centered attention and awareness to parentchild interactions (Bögels et al., 2010).Mindful parenting involves practices such as listening to the child and oneself with full attention, being emotionally aware of moment-to-moment parent-child interactions, accepting present moment experiences nonjudgmentally, having compassion toward the child and oneself, and being able to self-regulate during parent-child interactions (Duncan et al., 2009).In the face of interparental conflict, potential ruminative thoughts associated with conflict (Nolen-Hoeksema et al., 2008) may reduce parents' capacity to mindfully engage with the child.By contrast, parents who are mindful in parenting may be more capable of detaching themselves from autopilot, rumination, and potential biases, and being more present and responsive (vs.reactive) to their child's needs and their own needs (Duncan et al., 2009;Parent et al., 2014).Previous research indicated that mindful parenting fosters positive parenting practices during parent-child interactions, such as proactive parenting, positive reinforcement, warmth, and supportiveness, and weakens negative parenting practices, such as hostility, physical control, and lax control (Parent et al., 2021).By cultivating mindfulness, parents may also be more capable of pausing and sensing potential spillover from destructive interparental conflict to parenting behaviors and parent-child interactions (Parent et al., 2014).
Despite longstanding research showing that interparental conflict undermines parenting behaviors, parenting can also spill over to marital relationship outcomes, namely interparental conflict and marital satisfaction.The bidirectional relation between parenting and parents' relationship outcomes may be explained by family systems theory (Cox & Paley, 1997), which posits that the dynamics between the parent-child subsystem and the interparental subsystem of a family are interdependent.As one of the most common sources of interparental conflict (Papp et al., 2009), differences in childrearing decisions and parenting styles are likely to lead to a greater level of conflict.Previous research showed that parenting difficulty served an important factor for declines in marital intimacy over the first 3 years after childbirth (O'Brien & Peyton, 2002).In addition, child-related stress as indexed by stressful experiences in parenting, childrearing, and dealing with children negatively predicted parents' relationship satisfaction (Zemp et al., 2017).Although parenting attitudes and parents' contributions to childcare predicted their relationship satisfaction (Ehrenberg et al., 2001), the prediction was dependent on parent gender and attachment avoidance (Fillo et al., 2015), thereby suggesting third variables in affecting the prediction.Turning to the role of mindful parenting, although qualitative data showed that mindfulness-based childbirth and parenting education had a positive effect on couple relationship quality (Duncan & Bardacke, 2010), quantitative data indicated that a mindful parenting intervention failed to directly improve marital functioning, as indexed by marital conflict, marital satisfaction, negative communication, and open communication (Bögels et al., 2014).In Bögels et al.'s study (2014), however, most parents took the mindful parenting program without their spouse.The investigators speculated that similar programs organized exclusively for couples might yield positive findings on marital functioning (Bögels et al., 2014).
Given the mixed findings in the literature, the present study gravitates toward understanding the link between interparental conflict and mindful parenting as a function of parent gender.Gender plays an important role in the relation between conflict and parenting.Of note, previous research conducted in the United States suggested longitudinal and bivariate effects of interparental conflict experienced by mothers and fathers (Cheung et al., 2016).That is, mothers' and fathers' conflict behaviors not only predicted their own future conflict behaviors, but also their spouses' conflict behaviors over time.However, another study conducted in Hong Kong suggested only mothers', but not fathers', emotion dysregulation predicted their spouses' emotion dysregulation longitudinally (Cheung et al., 2020), thereby implying the negative effect of mothers' negativity might be stronger.In relation to parenting, a meta-analytic study showed a stronger relation between interparental conflict and parenting behaviors for fathers than mothers (Krishnakumar & Buehler, 2000).However, recent studies revealed complex relations between mothers' versus fathers' level of conflict and their spouses' parenting, depending on the type of parenting behaviors (Gao et al., 2019;Pedro et al., 2012), child gender (Sturge-Apple et al., 2004), and third variables such as interparental relationship insecurity (Davies et al., 2004).Hence, the strength of association between interparental conflict and parenting may differ as a function of gender, type of parenting behaviors, and other factors such as the design of the study.
Besides gender, culture may also affect the relation between interparental conflict and mindful parenting.In a Chinese context where the present study took place, interpersonal harmony is highly valued in families (e.g., Bond, 2010), as reflected by Confucius philosophy and proverbs such as "和為貴" (i.e., harmony is precious) and "家和萬事興" (i.e., everything prospers in harmonious families).Previous research indicated that people from collectivistic cultures tend to avoid confrontations and behaviors that strain relationships (Leung et al., 2011).As such, interparental conflict may be particularly detrimental to parents' wellbeing and parent-child relationship in the Chinese context.Relatedly, conflict tactics such as avoidance-capitulation may be more pervasive in the Chinese culture (Cheung, 2021), even though other tactics including stonewalling, verbal aggression, physical aggression, and child involvement are common in both Western and Chinese contexts (Kerig, 1996;X. Li et al., 2019).The potential pervasiveness of conflict avoidance may affect not only the dynamics between couples, but also parenting and parent-child dynamics.Hence, the study of interparental conflict and its parenting correlates deserves unique research attention in the Chinese context.
Parents of adolescent children are also of particular importance to this study, as the transitional period of puberty can elicit parent-child conflict intensity, family stress, and mood disruptions (Arnett, 1999;Laursen et al., 1998), which may further exacerbate interparental conflict (Sherrill et al., 2017) and parenting behaviors (Arnett, 1999).In this study, it was hypothesized that interparental conflict and mindful parenting would be bidirectionally related over time.Given the complexity of the strength of association between mothers' versus fathers' conflict and parenting behaviors, we did not specify an a priori hypothesis as a function of parent gender.It was also hypothesized that mothers' and fathers' interparental conflict would be longitudinally and transactionally related.Although previous research suggested that mothers' and fathers' parenting practices were transactionally related over time (e.g., behavioral control; Van Lissa et al., 2019), little has been done to examine the transactional effects between mothers' and fathers' mindful parenting.Therefore, an a priori hypothesis was not made for the transactional relation between mothers' and fathers' mindful parenting in this study.

METHOD Participants
The present study was part of a larger project aiming to examine family processes in adolescence (Cheung et al., 2020).Data were collected between 2015 and 2017 in Hong Kong, with a 12-month lag between Time 1 (T1) and Time 2 (T2).At T1, participants were 386 families involving 371 mothers and 305 fathers of adolescent boys (n = 185) and 201 girls (n = 201) from 12 to 17 years of age (M age = 13.64,SD = 1.15).As data were missing in some families, Little's missing completely at random test was conducted to test the null hypothesis of data missing completely at random.The finding was nonsignificant, χ 2 (1205) = 1282.99,p = .06,suggesting the data were missing completely at random.Parents were recruited through invitations via four secondary schools representative of all three areas of Hong Kong, including Hong Kong Island, Kowloon, and New Territories (Rating and Valuation Department, 2021).Parents were eligible to participate if they were (a) proficient in Chinese, (b) maritally intact, and (c) had an adolescent child ranging in age between 11 and 17 years.The parents were ethnically Chinese and were from Hong Kong.The monthly median household income ranged from HK$20,001 to $30,000 (approximately US$2564.23 to $3846.15), which was similar to that of the general population in Hong Kong (i.e., HK$26,000; approximately US$3333.33; Census and Statistics Department, 2021a, p. 103).In terms of education, 11.74% of mothers and 13.03% of fathers completed primary school, 30.20% of mothers and 32.04% of fathers completed lower secondary school, 42.61% of mothers and 40.14% of fathers completed upper secondary school, 13.09% of mothers and 11.97% of fathers had a bachelor's degree or above, and 2.36% of mothers and 2.82% of fathers were not otherwise specified.Compared to 30-to 59-year-old adults from the general population of Hong Kong, a greater percentage of the participants reported completing a lower level of education (e.g., 11.74% of mothers and 13.03% of fathers from this sample vs. 8.51% of women and 4.75% of men from the general population completed primary school education; 30.20% of mothers and 32.04% of fathers from this sample vs. 15.67% of women and 16.37% of men from the general population completed lower secondary education; see Census and Statistics Department, 2021b, table 3.3A).In addition, the average household size of 4.04 (SD = 1.02) in this sample was greater than the average household size of 2.70 in Hong Kong (Census and Statistics Department, 2021a, p. 97).
Mothers and fathers completed a set of paper-based questionnaires independently for two time points.Completed packets were returned in sealed envelopes through the secondary schools or by mail.The attrition rate was 14.51%, that is, 56 families dropped out at T2.The dropped-out and retained participants did not differ in all of the variables under study (ps > .05).

Measures
Given that Chinese is an everyday language among the participating families, the questionnaire was presented in Chinese.Following the back-translation procedures (Bartram et al., 2018), the measures were translated from English to Chinese by two trained independent research assistants proficient in Chinese and English.Discrepancies were resolved by the first author and the research assistants upon follow-up discussions.

Destructive interparental conflict tactics
Mothers and fathers completed five subscales of the Conflicts and Problem-Solving Scales (CPS; Kerig, 1996) regarding the conflict tactics employed by themselves and their partners.
The measure consisted of 32 items on Stonewalling, Verbal Aggression, Physical Aggression, Child Involvement, and Avoidance-Capitulation. Participants rated on a 4-point scale from 0 (never) to 3 (often).Higher averaged scores indicated a greater destructive interparental conflict tactics.To reduce social desirability and potential biases associated with self-report of interparental conflict (Sanford, 2010), mothers and fathers were independently invited to complete a self-report and a partner-report of interparental conflict tactics.To ensure confidentiality, they were asked to return the completed questionnaires to the research team in separate sealed envelopes.Sample items included, "sulk, refuse to talk, give the 'silent treatment'" (Stonewalling), "name-calling, cursing, insulting" (Verbal Aggression), "throw objects, slam doors, break things" (Physical Aggression), "become angry with child when really angry at partner" (Child Involvement), and "try to ignore problem, avoid talking about it" (Avoidance-Capitulation).The CPS has been validated in a sample of Chinese couples and yielded a reliable factor structure based on confirmatory factor analysis (X.Li et al., 2019).In this study, the CPS subscales had an adequate internal consistency over time, with Cronbach's alphas for mothers' selfreport = .80-.99, mothers' partner report = .80-.98, fathers' self-report = .77-.98, and fathers' partner report = .80-.98 at T1 and mothers' self-report = .74-.93, mothers' partner report = .78-.97, fathers' self-report = .77-.98, and fathers' partner report = .77-.98 at T2.

Mindful parenting practices
Mothers and fathers completed self-reports of the Interpersonal Mindfulness in Parenting Scale (IEM-P; Duncan, 2007) that assessed their mindful parenting practices.The 10-item scale consisted of three subscales including Awareness and Present-Centered Attention, Openness and Nonjudgmental Receptivity, and Non-reactivity.Participants rated on a 5-point Likert scale from 1 (never true) to 5 (always true).Higher averaged scores indicated greater mindful parenting practices.Sample items included, "I am aware of how my moods affect the way I treat my child" (Awareness and Present-Centered Attention), "I listen carefully to my child's ideas, even when I disagree with them" (Openness and Nonjudgmental Receptivity), and "When I'm upset with my child, I notice how I am feeling before I take action" (Non-reactivity).Though the 10-item IEM-P has not been validated in the Chinese context, its extended 23-item version has been validated and showed good validity and reliability (Lo et al., 2018).In this study, Cronbach's alphas were .74for mothers and .69for fathers at T1 and .69for mothers and .66 for fathers at T2.

Data analysis
MPLUS Version 8.3 (Muthén & Muthén, 2017) was used to conduct the analyses.First, the zero-order correlations, means, SDs, skewness, and kurtosis of the manifest variables were examined.Next, a structural equation model was conducted.To objectively capture interparental conflict, measurement models of conflict tactics involving latent and manifest variables were created at T1 and T2, with the five CPS subscale scores loaded on first-order latent variables of self-report and partner-report, respectively.The latent variables of self-report and partner-report were, then, loaded on second order latent variables of mothers' and fathers' conflict tactics, respectively.As for mindful parenting, the subscale scores generated from T1 and T2 self-reports were loaded on the latent variables of mothers' and fathers' mindful parenting practices, respectively.To account for shared reporter variance, residual variances were allowed to covary among manifest variables of the same reporter.To ensure weak measurement invariance of the longitudinal data, the factor loadings were constrained to be equal across time within each construct, although tests of measurement invariance showed that constraining several loadings did result in worse model fit (see Table 1 in the supplementary file).Rather than unconstraining the loadings, it was more crucial to draw meaningful conclusions by holding them to equality.Hence, the following factor loadings were constrained in the analyses: T1 and T2 mothers' self-report of interparental conflict, T1 and T2 fathers' self-report of interparental conflict, T1 and T2 mothers' partner-report of interparental conflict, T1 and T2 fathers' partner-report of interparental conflict, T1 and T2 mothers' self-report of mindful parenting, and T1 and T2 fathers' self-report of mindful parenting.Strong measurement invariance was not performed, given the invariance of mean structures of the latent variables was beyond the focus of this study.Maximum likelihood was adopted to evaluate the model fit to the observed matrices of variance and covariance.A covariance matrix and SEs for the covariance matrix of the latent variables were estimated.Although we did not conduct an a priori power analysis, post hoc power analysis using semPOWER (Moshagen & Erdfelder, 2016) indicated that the power was greater than 99.99% in rejecting the incorrect model when N = 386, df = 1166, RMSEA = .05,and alpha = .05.Full information maximum likelihood estimation was used to reduce bias from any missing data.

RESULTS
Table 1 in the main text shows the covariance matrix and SEs for the covariance matrix of the latent variables in the structural equation model.Table 2 in the supplementary file shows the zero-order correlations, means, SDs, skewness, and kurtosis of the variables.The structural equation model demonstrated adequate fit to the data, χ 2 (1166) = 2190.50,p > .001,χ 2 / df = 1.88, p < .001,CFI = .91,TLI = .90,RMSEA = 0.05, SRMR = .08(see Figure 1).Table 3 in the supplementary file shows the unstandardized estimates and SEs of the structural equation model.
Supplementary analysis was conducted to further examine whether patterns of findings statistically varied by gender.However, the model with measurement invariance across gender yielded poor fit (Bentler, 1990;Hu & Bentler, 1999), with TLI < .90 and SRMR > .08.Hence, we were unable to determine whether the pattern of findings varied between mothers and fathers.

DISCUSSION
Supporting the spillover hypothesis (Erel & Burman, 1995;Katz & Gottman, 1996;Y. Li et al., 2011;Warmuth et al., 2020), the present study demonstrated longitudinal relations between mothers' and fathers' reports of interparental conflict and mindful parenting.Notably, mothers' and fathers' destructive conflict predicted their own and their spouses' subsequent destructive conflict, as well as their mindful parenting practices.In addition, mothers' and fathers' mindful parenting predicted their own mindful parenting over time, but not the other variables.These findings were consistent with the growing evidence attesting to the spillover effects of interparental conflict on parenting (Sturge-Apple et al., 2009;Warmuth et al., 2020).Supporting family systems theory (Cox & Paley, 1997) and studies indicating transactional effects between mothers and fathers (Cheung et al., 2016;Gao et al., 2019;Pedro et al., 2012), the present study also evidenced transactional effects between interparental conflict and mindful parenting over a 12-month span.
One of the key findings of this study was that interparental conflict predicted mindful parenting over and above previous levels of mindful parenting, regardless of parent gender.When parents exhibited a greater level of destructive conflict tactics, such as aggression, child involvement, and stonewalling, they were less likely to refocus on engaging with their child mindfully.Being provoked by conflict, for instance, parents might be preoccupied with negative interparental dynamics and were less likely to be attentive to moment-to-moment parent-child interactions.Similarly, destructive conflict might spill over to parents' lower self-regulation in parenting.Parents experiencing greater interparental conflict might also carry preexisting negativity to parent-child interactionsthey might be less likely to be nonjudgmental, to detach themselves from autopilot, and to offer kindness and compassion to themselves and their child especially in times of need.
Surprisingly, mindful parenting did not foster lower levels of interparental conflict, regardless of parent gender.Although the findings were consistent with a previous study indicating that a mindful parenting intervention failed to directly improve marital functioning (Bögels et al., 2014), the null findings also diverged from previous findings about general parenting, in that parenting difficulty predicted declines in marital intimacy (O'Brien & Peyton, 2002), and that parents' perceived child-related stress was linked to lower couple relationship satisfaction (Zemp et al., 2017).In speculating the discrepant findings, perhaps the negativity and potential threat involved in parenting were more crucial than mindful parenting in predicting interparental conflict.Hence, future studies should investigate mindful parenting and both positive and negative parenting practices in relation to conflict between parents.In addition, perhaps the parents in this study primarily experienced conflict on issues other than parenting and childrearing decisions (e.g., money, relatives, work; Papp et al., 2009).Therefore, mindful parenting might not have directly reduced interparental conflict, particularly on those issues.Future research should further examine the nature, frequency, and severity of conflict in relation to mindful parenting.
Another major finding was that mothers' and fathers' conflict tactics set the stage for their spouses' conflict 12 months later.This finding corroborated other studies conducted in both Eastern and Western contexts, in that husbands' and wives' destructive conflict behaviors and/or marital quality were reciprocal (e.g., Boker & Laurenceau, 2006;Cheung et al., 2016;  3 from the supplementary file for further details of the unstandardized estimates and SEs. X. Li et al., 2018).Supporting family systems theory (Cox & Paley, 1997), the conflict and hostility between parents compromised their own and their spouses' future levels of conflict, regardless of parent gender.Hence, the adverse effect of destructive interparental conflict applies to both mothers' and fathers' future conflict as well as mindful parenting.
Not surprisingly, the CPS subscale of Avoidance-Capitulation yielded low, albeit significant, factor loadings on self-and partner-reports of interparental conflict (see Figure 1).A closer examination of the zero-order correlations revealed that unlike the other CPS subscales, Avoidance-Capitulation was not consistently correlated with the rest of the CPS subscales for mothers and fathers.In addition, mothers' and fathers' average scores of Avoidance-Capitulation were persistently greater than the rest of the subscales (see Table 1 in the supplementary file).Consistent with previous research (e.g., Leung et al., 2011), the present study pointed to the pervasiveness of conflict avoidance in the Chinese context.Unlike the findings from another study (X.Li et al., 2019), however, our preliminary data suggested that due to the lower factor loadings, avoidance-capitulation might be different from the other tactics as indices of interparental conflict (e.g., verbal and physical aggression, child involvement, and stonewalling).Hence, future studies conducted in Chinese contexts or other collectivistic cultures may direct their attention to the role of conflict avoidance in relation to family dynamics.Finally, the average scores of mindful parenting were relatively high for parents, ranging from 3.04 to 3.58 on a 5-point scale, even though the parents did not reportedly participate in mindful parenting interventions.This finding corroborated other studies showing that parents in Hong Kong score relatively high on mindful parenting (e.g., Cheung et al., 2021;Lo et al., 2018).Perhaps the local parents are generally capable of being mindful in parenting, without the need to be taught or practiced through an intervention program.

Limitations and future directions
Several limitations of the present study merit consideration.First of all, the present study utilized only questionnaire reports.To increase objectivity of the data, future studies should incorporate other assessments such as observational measures of family interactions and physiological data.Second, the sample size was small.Although post hoc power analysis indicated that sufficient power to reject the incorrect model, the findings should still be interpreted with caution.Future research should increase the sample size and the number of time points to estimate the trajectories of interparental conflict and mindful parenting.Likewise, the mediating and moderating mechanisms between interparental conflict, mindfulness, and parenting practices should be further investigated across family subsystems.For instance, as a trait, parents' mindfulness may drive conflict and communication tactics, which may be linked to mindful coparenting behaviors within the interparental subsystem (e.g., Parent et al., 2016).The relation between parents' trait mindfulness and conflict may further affect parent-child dynamics (i.e., beyond the interparental subsystem).Future studies should examine the role of trait mindfulness and utilize a process-oriented approach to investigate mediators and moderators within and between family subsystems.In addition, the present findings reflected betweenperson associations and the findings might differ at the within-person level.Hence, future studies should also examine both within-and between-person effects associated with interparental conflict, mindfulness, and parenting behaviors.Next, although gender was one of the core interests in this study, we were unable to determine whether the pattern of findings varied between mothers and fathers, due to the poor model fit upon constraints.Future research should be conducted to investigate this research question.As another limitation, in this study we only assessed parents' destructive conflict tactics, namely verbal and physical aggression, avoidancecapitulation, stonewalling, and child involvement.Though significant at ps < .001, the factor loadings of avoidance-capitulation were consistently low in the structural equation model.
Future studies should examine the specific role of avoidance-capitulation as a conflict tactic, particularly in collectivistic cultures.Other dimensions including frequency, duration, and level of resolution should also be examined, as conflict is a multidimensional phenomenon.Relatedly, some of the variables, such as physical aggression, were skewed with a floor effect (see Table 1 from the supplementary file).Although previous research recommended that skewness between À3 and +3 and kurtosis between À10 and +10 were acceptable values in structural equation modeling (Griffin & Steinbrecher, 2013), future studies may consider transforming the variables and minimize the skewness.Finally, the present findings might only be generalizable to parents from Hong Kong.To increase generalizability, future studies should be conducted with parents from diverse ethnic groups in other Eastern and Western societies.Researchers may also consider involving parents with children at different developmental periods or follow the families for a longer period (e.g., 10 years) to capture developmental trajectories and effects of age on interparental and parent-child dynamics.

CONCLUSION
The present study adds to the growing literature attesting to the spillover effects of interparental conflict on parenting behaviors (Erel & Burman, 1995;Katz & Gottman, 1996;Y. Li et al., 2011;Warmuth et al., 2020).Findings revealed the significance of interparental conflict on mindful parenting, regardless of parent gender.Supporting family systems theory (Cox & Paley, 1997), the study underscores transactional relations between mothers' and fathers' destructive conflict tactics, as well as their negative effects on mindful parenting behaviors.These findings inform practitioners and researchers of the importance in reducing both mothers' and fathers' destructive conflict tactics, such that their effects on subsequent levels of conflict and mindful parenting could be alleviated.