Highlights:

  • Most fathers report high ratings of parenting self-efficacy, use of positive parenting strategies and feel supported by a partner.

  • One in five fathers report mental health concerns since becoming a parent.

  • Fathers reported lower parenting self-efficacy and lower satisfaction with their parenting than mothers.

  • Parenting self-efficacy was related to positive parenting.

There are lasting positive impacts on children’s social, emotional and cognitive functioning when fathers are involved in parenting (Panter-Brick et al., 2014; Sanders et al., 2010). Yet, little research has explored fathering at the population-level beyond the impact of fathers’ involvement, and most often, early research has been focused on father involvement with young children (Lamb, 2000; Pleck, 2010). More recently, research has addressed fathering older children and adolescents, finding strong relationships between fathers’ mental health and the mental health and behaviour of their children (Flouri et al., 2019; Wickersham et al., 2020; Ayano et al., 2021). However, there is more to learn about other paternal characteristics and contextual factors to explain these associations.

Addressing research on a diverse range of parenting practices, parenting strengths and support needs of fathers of children of all ages, the current study examines the parenting experiences of a large and broadly representative sample of fathers of children from birth to 18 years, with a particular focus on how fathers’ mental health and the co-parenting relationship interact to shape fathers’ parenting and their sense of parenting efficacy. Knowledge about how fathers’ parenting experiences contrast with mothers’ experience is vital to understanding both the strengths of fathers as well as areas where they may be in need of further support in their fathering role.

Moreover, understanding the interrelationships between parenting and family demographic characteristics (including, but not limited to fathers’ educational background and employment arrangements, father age, child age, and child disability or complex needs), paternal mental health and the co-parenting relationship will help illuminate pathways via which the life circumstances of fathers may impact on their parenting, and ultimately on their children.

In line with other conceptualisations of the relationships between demographic factors, parent mental health, parenting self-efficacy, the co-parenting relationship and children’s wellbeing (e.g., Jones et al., 2003; Martin et al., 2017), Fig. 1 outlines the conceptual model that underpins the constructs examined in this paper. This model aligns with established frameworks on the role of context in understanding parenting. For example, Belsky’s process model of the determinants of parenting outlines how important parent, child, family environment and contextual characteristics such as social support and socio-economic disadvantage influence parenting and child wellbeing (Belsky, 1984; Belsky et al., 2006).

Fig. 1
figure 1

 A conceptual model of the influences on fathers’ parenting

Full size image

The model outlined in Fig. 1 also aligns with the “family stress” model (see Conger et al., 2002; Masarik & Conger, 2017), which asserts that contextual stressors such as socio-economic disadvantage may produce psychosocial effects in parents in terms of limited control, perceived inequality, increased stress, and exclusion experienced by families which may reduce parents’ capacity to provide enriched and stable environments for their children (Garbarino et al., 2005; Petrill et al., 2004). Research over the past 20 years supports the pathways outlined in the family stress model, including longitudinal research that supports the directional influence of economic hardship on parent psychological distress for families of different cultural backgrounds (e.g., Newland et al., 2013) and for parents of adolescents (e.g., Landers-Potts et al., 2015; Hardaway & Cornelius, 2014; Ponnet, 2014) as well as younger children (e.g., Iruka et al., 2012). Research also supports the pathway from economically-influenced parent stress and difficulties in the co-parenting relationship via the mediator of psychological distress (e.g., Landers-Potts et al., 2015; Lee et al., 2013), while other research has found a role for acculturative stress in addition to economic pressure for negatively impacting the psychological wellbeing of couples, with subsequent impacts on the co-parenting relationship (Helms et al., 2014). Despite some acknowledgment of pathway variability according to parent gender (for example, Ponnet (2014) found that in middle income families financial stress impacted fathers’ positive parenting but not mothers’ positive parenting, with no such gender distinctions in low- or high-income families), the extent to which the family stress model specifically applies to fathers remains unclear. Nonetheless, in combination, this research attests to the importance of considering the role of family culture, socio-economic disadvantage and other family demographics in examination of the influences on parent mental health, parenting and co-parenting. This is especially needed in research about fathers, given the gaps identified herein.

Fathers’ parenting experiences are likely to have changed dramatically over recent decades (see Burgess 2007). Widespread commentary in the academic and non-academic literature reflect that societal expectations of men’s roles in the family have increased (e.g., Cooklin et al., 2016; Genadek & Hill, 2017; Livingston & Parker, June 2019). Research has attempted to inform understandings about how increased expectations have impacted fathers, mothers, and children. Yet, historically much of this research has adopted a deficit-perspective to fathering, focusing on the negative impact of attempts to balance work and family life, or addressing correlates of fathers’ low rates of engagement with parenting programs (Panter-Brick et al., 2014). Nonetheless, the extant literature does offer a range of perspectives about fathering, and how fathers’ mental health and the co-parenting relationship influences their parenting.

Parenting Practices

What parents do during interactions with their children matter. Specifically, the approach parents take to interactions, plus the activities they engage in with children, help to shape children’s socioemotional and cognitive wellbeing. Evidence supports the view that parent-child interactions and the quality of the parent-child relationship have a profound effect upon children’s and adolescents’ social, behavioural and cognitive development (e.g., Morris et al., 2017; Sylva et al., 2004; Williams & Berthelsen, 2017). The specific approach parents use in interactions with their children that are thought to be important to children’s wellbeing can be broadly described as being sensitive and responsive. That is, children thrive when parents are sensitive to their needs (e.g., emotional and physical needs) and when parents provide an appropriate, consistent and timely response to address the child’s needs. For example, responsive parenting has been associated with increased growth of infant social, emotional, communication and cognitive competence, particularly so for very low birth weight babies (Landry et al., 2006). When parents provide environments for their children that include warmth and nurturance and that encourage linguistic and cognitive stimulation, children’s emotional and cognitive development benefits (see Guttentag et al., 2006).

An emerging area of research interest relates to children’s use of electronic devices and the behaviour of parents to monitor and limit children’s use of such devices and access to the internet. Parents may play a role in monitoring and limiting children’s access to technology in a variety of ways and for a range of reasons. It may be that time spent on devices is the primary concern to parents, or their restrictions may relate to the content accessed, including the people and online locations the child interacts with. Furthermore, parental monitoring and restriction of their children’s online and on-screen behaviour is likely to change as the child ages, thus, parents will typically need to adapt their strategies to keep pace with the ever-changing digital landscape, but also with their child’s developmental changes. Today’s children have been described as ‘digital natives’ and their parents as ‘digital immigrants’ (Nelissen & Van den Bulck, 2017), whereby children are the conduit via which parents understand the potential and the pitfalls of increased availability of technology in family life. How the expanding availability of web-based information and use of portable devices by children impacts on parenting requires further exploration (Oerther & Oerther, 2021). As such, parenting practices in monitoring and limiting children’s access to electronic devices is an important aspect of the contemporary parenting experience.

In sum, parenting interactions characterised by warmth, attentiveness and proactive monitoring, praise for positive behaviour, and firm yet calm and consistent discipline for negative behaviour, are widely endorsed by researchers as most likely to lead to good outcomes for children. Nonetheless, most of what is known about parenting from the extant literature is based on research involving mothers. While it is likely that effective parenting practices among men reflect similarly to effective parenting practices among women, less is known about how fathers interact with their children, and their preferences for use of particular parenting practices, compared to what is known about women’s parenting. Furthermore, improved understanding of the role of demographic factors such as fathers’ education, employment and age, as well as child age, gender and special needs, will increase knowledge about the anticipated complex interrelationships between child and family factors and fathers’ parenting.

Gaining greater understanding of fathers’ preferences and experiences of parenting is important because, according to past research, the amount of time fathers spend with their children has increased over time (Bornstein & Putnick, 2016). However, young children continue to spend more time with their mothers than their fathers (e.g., Baxter et al., 2007; Baxter & Smart, 2010), which may be explained by ongoing mother-father discrepancies in employment (Giallo et al., 2013). Nonetheless, the value of fathers’ time with children is rarely questioned.

Men’s parenting practices and style are as diverse as mothers’. While some past research has focused on the ‘rough and tumble play’ aspects of father-child interaction as a distinguishing feature of men’s parenting within the pre-school period (e.g., Paquette & Dumont 2013), other research reflects the diversity of father behaviour incorporating a broad range of potential interaction styles including warmth, responsivity, enriching engagement, and authoritative and permissive parenting styles (see Cabrera et al., 2018; Marsiglio et al., 2004). Alongside diverse ways of parenting comes a range of impacts on children that are both positive and negative (Panter-Brick et al., 2014; Amodia-Bidakowska et al., 2020). Further, research has addressed the role that fathers’ mental health plays in influencing their parenting practices.

Fathers’ mental health

Notwithstanding variability in reports of prevalence of mental illness among parents, researchers from developed countries generally acknowledge that up to a quarter of children live in a family where a parent has mental illness (e.g., Maybery et al., 2009). While much of this evidence is based on mothers’ prevalence (Bronte-Tinkew et al., 2007), some studies have specifically focused on fathers’ perinatal mental health (see reviews by Leach et al., 2016 and Paulson & Bazemore 2010), including factors that influence mental ill-health in perinatal fathers (see review by Baldwin et al., 2018).

Research indicates that in addition to the negative influence of maternal mental health, fathers’ mental health during the transition to parenthood and beyond can have an impact on children’s mental, academic and social functioning and on parenting (McLaughlin et al., 2012; Pape & Collins, 2011; Weissman et al., 2006). The pathway via which fathers’ mental health impacts on children is likely to be through the effect mental health challenges have on irritable or hostile parenting. The associations detected between fathers’ mental health challenges and the use of harsher discipline with preschool children (e.g., smacking and yelling; Davis et al., 2011; Giallo et al., 2014; Giallo et al., 2015) are likely to explain a significant proportion of the effects of fathers’ mental ill-health on young children’s emotional and behavioural functioning (Velders et al., 2011). Paternal mental health challenges have also been associated with reduced time spent with infants (Bronte-Tinkew et al., 2007) and young children (Giallo, Treyvaud, et al., 2013), which may be considered a risk or a protective factor (i.e., if fathers choose not to engage because they are unwell). Some research suggests fathers who engage in antisocial behaviour may place their preschool children at risk of conduct problems – an effect which was shown to increase with the amount of time a father spent living with the child (Jaffee, Moffit, Caspi & Taylor, 2003).

Fathers’ sense of parenting self-efficacy

Parenting self-efficacy (PSE) can also influence fathers’ parenting practices. PSE has been defined as the ‘… beliefs or judgements a parent holds of their capabilities to organise and execute a set of tasks related to parenting a child’ (De Montigny & Lacharite, 2005; p. 387). While PSE has been conceptualised in a range of different ways (e.g., see Jones & Prinz 2005), it is often described as a multi-component construct encompassing views about the self as effective at parenting, having personal agency in effecting good outcomes for children, being a good self-manager and being self-sufficient (see Hamilton et al., 2015). PSE influences parenting in a range of ways. Parents with higher PSE are more likely to employ positive parenting practices, including greater perseverance and involvement and more warm interactions, with these results evident across children’s developmental stages and a range of cultural groups (Ardelt & Eccles, 2001; Baxter & Smart, 2010; Jones & Prinz, 2005; Rominov et al., 2016; Wade et al., 2015).

Most PSE research has focused exclusively on mothers. In one notable exception, Junttila and colleagues (2015) examined PSE in mothers and fathers of preschool Finnish children. PSE in relation to presence, emotional support, routines and teaching was higher for mothers than fathers, although PSE related to play with the child was similar. Thus, fathers perceived themselves to be less competent than mothers in most domains of parenting. Conversely, Baxter and Smart (2010) reported on data collected about over 10,000 0- to 9-year-old children in the Longitudinal Study of Australian Children, finding few differences in ratings of PSE between mothers and fathers. Further exploration of the role of PSE in shaping fathers’ interactions with their children is warranted.

The co-parenting relationship

As distinct from the couple relationship, the co-parenting relationship has been linked to parental wellbeing, parenting practices, and to child outcomes (Feinberg, 2002; Parkes et al., 2019; Pilkington et al., 2019), even among separated parents (Whiteside & Becker, 2000). Defined as ‘the ways that parents work together in their roles as parents’ (Feinberg, 2002, p. 173), co-parenting clearly plays an important role in family life. For example, co-parenting relationships characterised as unsupportive and undermining when a child is 3 years old have been found to predict child externalizing behaviour at 4 years (Schoppe et al., 2001). In a separate study examining large datasets from two separate countries, Parkes and colleagues (2019) found that up to 60% of the effect of couple supportiveness on child behaviour problems when children were 8–10 years was attributable to co-parenting and parenting when children were 3–5 years. Furthermore, evidence from a study involving parents of preschoolers and pre-adolescents indicates co-parenting conflict (specifically) may mediate the relationship between partner conflict (broadly) and parenting (Margolin et al., 2001). Additionally, representative Australian data indicate fathers facing mental health challenges are likely to experience more issues with their co-parenting relationship, with the relationship likely to be one that is bidirectional (Price-Robertson, Baxter & Matthews, 2017). As such, the co-parenting relationship offers a modifiable target for intervention with potential for improvements in family functioning (Feinberg, 2002). The current study aims to add to the extant literature on the role of the co-parenting relationship by specifically examining how fathers perceive their relationship with their partner in parenting, and exploring correlates (i.e., potential antecedents and/or consequences) of poor co-parenting ratings by fathers.

Building on evidence from previous research, and in line with the conceptual model outlined in Fig. 1, the current study explores characteristics, experiences and preferences of fathers through examination of a rich and representative dataset. In addition to examining how parenting by fathers differs to parenting by mothers and how mothers and fathers compare in relation to their own mental health and views about co-parenting, this paper explores how particular characteristics of fathers – like mental health, PSE, and the co-parenting relationship – relate to how fathers parent. As such, the paper aims to provide improved understanding of the complex interrelationships between child and family factors with fathers’ mental health and their parenting in order to better target supports for parenting to fathers, including those with poor mental health and children with complex needs.

This study has four broad aims: (1) to explore mother/father differences in views about their own parenting, their mental health and the co-parenting relationship, in order to articulate differences in the parenting experience of mothers and fathers and to better understand the relative strengths and needs of fathers when compared to mothers; (2) to examine how fathers’ mental health is related to demographic characteristics such as employment, educational attainment, and child age as well as to family circumstances such as the child having additional needs, and the fathers’ contact with the child; (3) to examine how fathers’ views about the co-parenting relationship vary according to demographic and wellbeing factors, including fathers’ mental health and PSE ratings; and (4) to examine possible influences on parenting by fathers, including how parenting practices vary dependent on fathers’ PSE as well as how parenting practices and PSE among fathers vary according to key demographic characteristics, their mental health and their views about partner support.

In relation to the first aim, and building on past research, we hypothesise mother-father differences in parenting approach and activities, lower rates of mental health concerns among fathers compared to mothers, but similar perceptions about the co-parenting relationship. For aim 2, we hypothesise that fathers’ mental health will be associated with a range of family contextual factors, including that poorer paternal mental health will be associated with greater child needs, less paternal contact with the child, lower levels of fathers’ employment and lower child age. In relation to aim 3, we hypothesise that fathers’ views about the co-parenting relationship will vary in association with family contextual factors, such that lower levels of employment and educational attainment will be associated with poorer co-parenting relationships, and that having a non-English speaking cultural background will be associated with better co-parenting relationships. We also hypothesise that ratings about co-parenting will be positively associated with fathers’ mental health and sense of parenting efficacy. Regarding aim 4, in line with our conceptual model, we hypothesise that fathers’ parenting practices will be positively related to their PSE, and that fathers’ parenting practices will have a stronger positive association with their mental health, PSE and co-parenting than with parent and child socio-demographics such as father or child age, fathers’ educational attainment and employment.

Method

Sample

This study uses secondary analysis of cross-sectional data collected from the 2016 Parenting Today in Victoria computer-assisted telephone interview (CATI) survey (N = 2600 mothers and fathers) involving 1044 fathers (40.2% of the total sample) recruited through random dialling of landline and mobile phone numbers (see Wade et al., 2018). All respondents were from separate households, thus matching of mothers’ and fathers’ responses within the same family is not possible. The CATI was conducted by an independent polling company, Ipsos. While the demographic characteristics of the final sample matched the distribution of most demographics of Victorian parents and their partners in the 2011 Australian Census (Australian Bureau of Statistics, 2011), sample weighting was performed using three variables which showed > 5% discrepancy with population norms: parent age group, educational level and type of residential location (metropolitan or regional). Weighting resulted in 2535 usable cases (i.e., with no missing data for weighting variables), 1006 of which were fathers. Weighted data are used for all analyses in this paper.

Fathers were defined as male respondents who had a parenting role and most (95.3%) were the biological parent of the focus child (compared to 97.0% of female respondents). Parents were directed to answer some items about one of their children (if they had multiple children) – the child (0–18 years; 20.1% were 0–2 years of age, 17.7% were 3–5 years, 35.4% were 6–12 years, and 26.8% were 13–18 years) in their care who had their birthday most recently. Of these children, 24.6% were reported to have a medical condition or learning difficulty. Around 95% of fathers recruited spent most days in a month (> 21 days) residing with the focus child. Most fathers (86.8%) were living with a partner (not necessarily the child’s parent), and no information was available about the gender of their partner. Of those not living with a partner, 10% shared care of the child with their ex-partner. Most fathers (79.1%) and a fifth (20.0%) of mothers were in paid full-time employment. 9.6% of fathers and 41.0% of mothers were in part-time or casual work. Fathers listed ‘home duties’ as their main occupation in 8.2% of cases, compared to 39.0% for mothers. Most mothers (45.0%) and fathers (47.0%) had a diploma or degree (higher than the broader Victorian parenting population of 26%, Australian Bureau of Statistics 2011), 13.1% of fathers and 7.4% of mothers spoke a language other than English at home, and 1.0% of mothers and 1.1% of fathers identified as Aboriginal or Torres Strait Islander.

Compliance with Ethical Standards.

The study has approval from a Human Research Ethics Committee that is accredited by Australia’s National Health and Medical Research Council. Informed verbal consent was obtained by the CATI facilitator prior to data collection. There are no conflicts of interests for any of the authors in relation to this study. The first author takes responsibility for the integrity of the data and accuracy of analyses.

Measures

In addition to questions about demographic characteristics (e.g., child and parent age and sex, employment status, educational attainment, employment, household income, language spoken at home, child additional needs), the following text describes survey items analysed for the current paper, which form a subset of the total 109 items of the overall Parenting Today in Victoria 2016 survey. In general (and where relevant) parents were asked to answer questions in relation to one of their children. The survey included two established scales as well as individual items tapping into constructs of interest.

Parent mental health

The first established scale used was the Kessler-6 (K6; Kessler et al., 2002), which measures current psychological distress and was used as one indicator of mental health challenges. This psychometrically-sound version of the longer Kessler measure invited respondents to indicate how often during the past 30 days they had felt nervous, hopeless, restless or fidgety, extremely sad, worthless and that everything was an effort. Responses are given on a 5-point scale (1 = all of the time; 5 = none of the time), with an initial score range of 6–30 reversed and then converted to the more widely used convention (range 0–24) with higher scores representing greater distress. Respondents scoring above 12 were considered to be reporting serious psychological distress; those with a score of 3–12 were considered to have moderate levels of distress; and those scoring 0–2 were considered to be experiencing low distress. The K6 demonstrated good internal consistency in the Parenting Today in Victoria sample (α = 0.80).

A further mental health item was developed specifically for this survey. This item asked parents to respond to the question ‘Since becoming a parent, have you had symptoms of any of the following: (a) Depression; (b) Anxiety; (c) Substance addiction; (d) None of these’. Thus, respondents could respond ‘yes’ to one, two or three of the conditions, or reply ‘d’ to indicate zero of these. The question did not ask about professional diagnosis (etc.), therefore the item was responded to based on a respondent’s own interpretation of what ‘symptoms’ of each condition might be.

We derived an item from the K6 and the single mental health item regarding history of mental health problems. This derived variable reflected ‘better’ or ‘poorer’ mental health by classifying each participant (i.e., the total sample of mothers and fathers) according to their K6 scores (serious, moderate or low psychological distress) in combination with history of any past problem since becoming a parent (anxiety, depression or substance abuse). Those in the ‘poorer’ mental health group had serious or moderate K6 distress scores regardless of past problems. The ‘better’ mental health group had moderate K6 scores plus no past problems, or they had low K6 scores regardless of past problems. Internal consistency for this ‘Better or Poorer Mental Health Scale’ was moderate (α = 0.58).

Parenting self-efficacy (PSE)

The second established scale - the Me as a Parent Scale (MaaPs; Hamilton et al., 2015) - measures parents’ perceptions of their parenting or PSE. MaaPs items can be summed into four subscales and a Total score. Subscales incorporate four items each, and the Total score is the sum of all items. The subscales reflect the four constructs of Parenting Self-Efficacy (e.g., ‘My parenting skills are effective’), Personal Agency (e.g., ‘How my child turns out is mainly due to luck’), Self-Management (e.g., ‘When changes are needed in my family I am good at setting goals to achieve those changes’), and Self-Sufficiency (e.g., ‘I know how to solve most problems that arise with parenting’). Response options range on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). The Total score range is 16–80, where a high score means the parent feels more confident, effective and in control regarding parenting the focus child. The MaaPs demonstrated good internal consistency in the Parenting Today in Victoria sample (Total score, α = 0.87; subscales, α = 0.68 to 0.83).

Parenting Practices

Parenting practices were measured using a series of individual items either taken from existing measures or created by the survey developers. Some of these items were combined to form derived scales which have been used in this paper to facilitate multivariate analyses examining associations with PSE, demographic factors and fathers’ mental health, while some items are retained as individual items for all analyses.

For the first derived scale, ‘Parental Approach’, four items taken from the Parent Satisfaction Scale (Guidubaldi & Cleminshaw, 1985) were summed. Items asked parents to rate on a 5-point Likert scale (ranging from 1 = ‘strongly disagree’ to 5 = ‘strongly agree’) the extent to which they agreed they: (a) wished they did not become impatient so quickly, (b) wished they were more consistent in their parenting; (c) were sometimes too critical of their child; and (d) were satisfied with the amount of time they could give their child (this item was reverse coded). Internal consistency for this derived ‘Parental Approach Scale’ was moderate (α = 0.59). For the fathers in the sample a mean of 12.93 (sd = 2.86) was observed on the Parental Approach Scale (range = 4–20) with higher scores indicating more positive approach.

For the second derived parenting practices scale, ‘Parent Activity’, three items were used that were adapted from items used in the Longitudinal Study of Australian Children survey (http://www.growingupinaustralia.gov.au/studyqns/index.html). These asked parents how often they (a) play music, sing, dance or do other musical activities with their child, (b) play games indoors with their child, and (c) play a game outdoors or exercise with their child, with each item rated on a 4-point Likert scale (ranging from 1 = ‘often’ to 4 = ‘not at all’). Scores from each item were reverse coded and then summed to create the ‘Parent Activity Scale’, with higher scores indicating greater frequency of activity with children. Internal consistency for this ‘Parent Activity Scale’ was moderate (α = 0.60). For the fathers in the sample a mean of 9.93 (sd = 1.91) was observed on the Parent Activity Scale (range = 3–12).

Three individual items were taken from the Parenting and Family Adjustment Scale (Sanders et al., 2013) which asked parents to rate on a 4-point Likert scale (ranging from 1 = ‘not at all’ to 4 = ‘very much’) how often they: (a) rewarded good behaviour with praise/a treat/attention; (b) smack their child when they misbehave; and (c) argue with or yell at their child about their attitude or behaviour. One individual item was taken from the Parental Communication scale (Botvin, 2007), which asked parents to rate on a 5-point Likert scale (ranging from 1 = ‘never’ to 5 = ‘always’) how often they talk to their child about problems/issues that they might be dealing with.

One item was created to measure parents’ approach to managing their children’s internet and electronic device use. Parents were asked to indicate whether (yes/no) they used one or more specific strategies, with responses to two options used in the current paper: (a) I talk about safe use of internet connected devices; and (b) I monitor online activity.

Two further created parenting practices items asked parents to reflect on (a) whether they know where their children are when not at school, and (b) if they had rules and set limits about where their children go in their free time, each rated on a 5-point Likert scale (ranging from 1 = ‘never’ to 5 = ‘always’).

Co-parenting

Three items developed specifically for this survey addressed the degree of parenting support received from the child’s other parent or their own partner (respondent could decide which person was the most significant other person in their child’s life – could be a non-biological ‘step-parent’ or non-resident parent, for example). These items were rated on a 5-point Likert scale (ranging from 1 = ‘all of the time’ to 5 = ‘never’) and asked parents how often they: (a) feel their child’s other parent understands and is supporting them as a parent; (b) are happy with the way parenting duties are shared; and (c) feel they and their partner agree on how to parent their child. These parent support items are consistent with Feinberg’s (2002) conceptualisation of co-parenting.

Items were asked of all parents, with the exception of the two items about children’s movements outside of school hours (asked only of parents of children from 6 to 12 years) and items about partner support – asked only of parents who indicated there was a partner or other parent helping them raise the focus child.

Data analysis

Analyses of the cross-sectional data were performed using SPSS, version 24. To address the first aim of the study, Pearson’s chi-square (χ2) was used to examine bivariate differences in the proportion of mothers and fathers reporting particular outcomes for ordinal or categorical data, while MANOVA was used to test multivariable differences for continuous variables (e.g., for MaaPs subscale and Total scores). MANOVA and χ2 were also used to address aspects of aims 2, 3 and 4, where differences between subgroups of fathers (e.g., those reporting poor, moderate or good PSE) on outcomes of interest (e.g., parenting practices) were examined. To address the fourth aim, we used ordinary least squares regression to examine factors predicting parenting practices. Multiple covariates were included in the regression including mental health, parent age, education, employment, household income, language spoken at home, child factors (age, gender, complex needs), and aspects of the co-parenting relationship. Non-parametric alternatives (e.g., Mann-Whitney U Test for independent samples in place of MANOVA) were also run where homogeneity of variance was violated.

We set statistical significance at alpha < 0.001 to reduce the likelihood of a Type I error as a result of the large number of analyses conducted for this study. However, where relevant (e.g., in tables) we have also reported p values of lower magnitude.

Results

Mother/father differences (Aim 1)

Parenting practices

Compared with mothers, fathers were significantly more likely to say they were too critical of their children and less likely to say they were satisfied with the time they could give their children, to talk to their child about problems, to set rules and limits about the child’s free time, to talk to their children about safe internet use, and to monitor online activity (all p < .001, see Table 1). Mother-father differences on other parenting items were not significant at p < .001.

Table 1 Fathers’ and mothers’ endorsement of views about their own parenting
Full size table

Parenting sense of efficacy

There was a statistically significant omnibus mother/father difference in MaaPs Total and subscale scores, F(4, 2596) = 12.933, p < .001, Wilk’s Λ = 0.980, partial η2 = 0.020. Follow-up univariate analyses indicated that there were statistically significant (p < .001) differences by parent sex on the MaaPs Total score and on two of the four subscales (see Table 2). In light of violation of the assumption of homogeneity of variance for three subscales, we used the non-parametric Mann-Whitney U Test, and found significant (p < .001) differences between fathers and mothers on the Self-Sufficiency and Self-Management subscales as well as the MaaPs Total score. Fathers’ mean PSE scores were consistently lower than mothers’ scores. Nonetheless, most fathers (73–96% across items) agreed or strongly agreed (or disagreed/strongly disagree for four items requiring reverse scoring) with each MaaPs item, indicating the majority of fathers reported high levels of PSE.

Table 2 Fathers’ and mothers’ ratings of their own parenting self-efficacy (PSE)
Full size table

Mental health

A concerning proportion of fathers reported they had experienced symptoms of depression (17.9%) or anxiety (19.0%) since becoming a parent. While this is lower than for mothers (depression = 34.0%; anxiety = 34.4%), of the fathers who reported depression since becoming a parent, 9.0% said this included post-natal depression. 3% of fathers reported serious levels of current psychological distress (5% in mothers).

Co-parenting

When asked about the parenting support they received from their child’s other parent or their own partner, the views of mothers and fathers differed in relation to views about shared parenting duties (see Table 3). Of the respondents who indicated there was a co-parent in their child’s life, the parent sex difference in ratings about the way parenting duties were shared was significant (p < .001) with 11.0% of mothers reporting occasional, rare or no satisfaction with the sharing of parenting duties, compared to 5.1% of fathers. Agreement about parenting and the extent to which parents felt supported by their co-parent did not show significant (p < .001) parent sex differences. Nonetheless, most fathers (93.6%) with a co-parent reported feeling supported and understood by the child’s other parent all or most of the time.

Table 3 Fathers’ and mothers’ endorsement of views about co-parenting support
Full size table

Correlates of fathers’ mental health (Aim 2)

Bivariate chi square analysis revealed a cluster of demographic variables that distinguished fathers with relatively better versus poorer mental health (see Table 4). Fathers classified as having ‘poorer’ mental health (20.1% of fathers in the sample) were significantly (p < .001) more likely to have a child with a medical condition or learning difficulty, and less likely to speak a language other than English at home (p < .001). Although just not significant, higher parent employment status (p = .001) was associated with better father mental health.

Table 4 Demographic characteristics associated with fathers’ mental health
Full size table

Correlates of co-parenting and partner support (Aim 3)

Bivariate chi square analyses identified a number of demographic and wellbeing factors were associated with fathers’ perceptions about co-parenting support. Across each of the three co-parenting items (listed in Table 3), a set of common factors were associated with fathers responding ‘occasionally’, ‘rarely’ or ‘never’ as opposed to ‘all of the time’ or ‘most of the time’. That is, fathers were significantly (p < .001) less likely to be feel supported and understood, less happy with how parenting duties were shared and less likely to say they agreed with their partner on how to parent the child if they were not living with a partner or spouse, spent fewer days per month with the child (10 days or less per month), and if they did not have shared care arrangements for the child with their ex-partner.

Fathers’ employment status was also significantly (p < .001) related to responses on one item - fathers who were employed full-time were more likely to say they were happy with the way parenting duties were shared. Other demographic factors (i.e., language spoken at home, parent age, child age and gender, the child having a medical condition or learning difficulty, father’s educational attainment, and family income) were not significantly related to father’s views about the co-parenting relationship, while PSE and mental health were not significantly associated with co-parenting support at the p < .001 level for any item.

Correlates of parenting practices and PSE (Aim 4)

MANOVA found differences in fathers’ parenting practices dependent on PSE. Using three groupings of MaaPs Total scores (poor = scores in the lowest quartile, range = 24–61; moderate = scores in the second lowest quartile, range = 62–65; and good = scores in the top two quartiles combined, range = 66–80) as the factor, a statistically significant omnibus difference in scores for a range of parenting practices (see Table 5) was found, F(12, 2092) = 23.206, p < .001, Wilk’s Λ = 0.779, partial η2 = 0.117. Follow-up univariate analyses indicated statistically significant (p < .001) differences for five of the six outcomes (all except smacking) by PSE. Thus, lower PSE was associated with less frequent activity between father and child, lower Parental Approach scores, less use of praise or rewards for good behaviour, less reliance on talking to children about problems, and more frequent arguing and yelling. Where it was possible to run non-parametric equivalent tests – if there was violation of the assumption of homogeneity of variance – significant between-groups differences were confirmed (p < .001).

Table 5 Associations between fathers’ parenting self-efficacy (PSE) and parenting practices
Full size table

Multiple regression was used to examine which demographic characteristics of the father or the child, as well as other factors (including fathers’ mental health and aspects of the co-parenting relationship) could significantly predict fathers’ scores on each of the following scales: the Parental Approach Scale, the Parent Activity Scale, and the MaaPs Total score (see Table 6).

Table 6 Results of the OLS regression to identify predictors of parental approach, parent activity and parenting self-efficacy (PSE) among fathers (n = 832)
Full size table

Results of the first regression indicated the model explained little of the variance (8%) in Parental Approach scores, yet the model was a significant predictor of parental approach, F(14,817) = 4.895, p < .001. Fathers’ mental health had the greatest and only significant (p < .001) contribution to the model (B = − 0.200, p < .001).

Variables in the second regression model explained 26% of the variance in Parent Activity scores, and the model was a significant predictor of parent activity, F(14,818) = 20.819, p < .001. Speaking a language other than English at home and child age had the greatest contribution to the model (B = − 0.121 and B = -412 respectively, both p < .001).

For the third regression, the model predicted 11% of the variance in MaaPS Total scores, nevertheless, the model was a significant predictor of PSE, F(14,818) = 7.013, p < .001. Speaking a language other than English at home and father’s mental health had the greatest contribution to the model (B = − 0.139 and B = − 0.172 respectively, both p < .001).

Discussion

Results from the 2016 Parenting Today in Victoria study – incorporating one of the largest surveys of fathers of its kind – demonstrate a generally positive picture of contemporary fathering. Findings from this analysis of fathers’ views about their own parenting provides evidence that most fathers are faring well. Most felt efficacious about their parenting and felt supported in their role as fathers. For example, eight in ten fathers reported frequent use of positive parenting strategies (e.g., rewards, praise or attention) to reinforce children’s good behaviour, two thirds said they often or always talked to their children about problems their children might be dealing with, and around nine in ten fathers felt supported by a parenting partner. These results go some way to reversing the deficit-perspective commonly seen in research about fathering, which has tended to focus on the negative impact of navigating a work-family balance or addressing fathers’ low rates of involvement with children and with parenting programs (Panter-Brick et al., 2014).

Findings about parenting by fathers in the current study are somewhat similar to findings from another randomly recruited Australian telephone survey of fathers of 0-12-year-old children (Sanders et al., 2010). For example, 65% of fathers in the 2010 study reported they were likely or very likely to shout at or become angry with their child when they misbehave, compared to 66% of fathers of 0-12-year-olds in the current survey who said they argue or yell at their child about their behaviour or attitude at least a little. Rates of smacking were slightly lower among fathers in the current survey (35% said they smack at least a little, compared to 42% in the 2010 survey who said they were likely or very likely to smack), which might be explained by community-level changes over time in the perceived acceptability of physical punishment.

Where fathers’ and mothers’ views differ (Aim 1).

Despite the generally positive picture of fathering, findings from this survey reveal some areas of concern and further attention in relation to fathers’ wellbeing and their parenting practices, with implications for the support we offer fathers. Importantly, many of the fathers who completed the Parenting Today in Victoria survey reported current or past mental health challenges. While in agreement with our hypothesis that fathers would report fewer mental health challenges than mothers, almost one in five fathers reported they had experienced symptoms of depression and/or anxiety since becoming a parent, and 3% were reporting serious levels of current psychological distress. Despite heterogeneity within the extant literature reporting the prevalence of paternal anxiety and depression in the community, and the limitation that most of this literature is about men during the perinatal period, the rate of mental health challenges among fathers detected in the current study does concord with previous research which has suggested up to 18% of perinatal fathers experience an anxiety disorder (Leach et al., 2016) and up to 26% of 3–6 month post-partum fathers experience depression (Paulson & Bazemore, 2010). Among Australian fathers, previous analyses (Giallo et al., 2012) of data from the representative Longitudinal Study of Australian Children, has found rates of serious psychological distress (using the K6 – the same measure used in the current study) among fathers of 4-5-year-old children (i.e., 2% of resident fathers and 3% for non-resident fathers) to be similar to that found for fathers of children 0–18 years in the current study (3%). These results add to the emerging evidence base about the prevalence of mental ill-health among fathers, and indicate men’s mental health is an important consideration in both research about fathering, and in the practice of supporting fathers and families.

While most fathers reported good partner support, this was sometimes in contrast to mothers’ views, thus refuting our hypothesis that mothers and fathers would have similar perceptions about the co-parenting relationship. Noting that our analyses do not facilitate comparison of responses between partners (the sample was made up of mothers and fathers who were not from the same families), of the parents in the study with a co-parent in their child’s life, fathers were more likely to be happy with the way parenting duties were shared between parents than were mothers (95% and 89% respectively). These views of fathers are consistent with findings from the survey of Australian fathers conducted by Sanders and colleagues (2010) whereby 94% of fathers indicated their partner was very supportive of their role as a parent. International research shows that the support parents provide each other through the sharing of everyday parenting tasks and responsibilities has an impact on child outcomes (Teubert & Pinquart, 2010). Thus, it is imperative that parents are ‘on the same page’ about their approach to parenting and these differences between mothers and fathers in perceptions about the division of parenting duties may indicate an area for intervention for some couples.

The study affords us valuable insights into how men can be supported in their parenting. For instance, examination of mother-father differences on individual survey items highlighted areas where fathers may be feeling less competent than mothers, and lends support to our hypothesis that there would be mother-father differences in parenting practices. Fathers were on average less satisfied with the time they could give their children, talked less to their children about problems and about safe internet use, set rules and limits about their children’s free time less, and monitored their children’s online activity less than mothers. Fathers were also more likely than mothers to say they were too critical of their children. Fathers were also less confident in their own parenting, as indicated by lower PSE scores compared to mothers.

The differences between fathers’ and mothers’ parenting suggest a different focus of support provision and parenting education may be needed for men compared to women. However, care should be taken to ensure any variation in the content of parenting information or education does not unintentionally reinforce any tensions between parenting partners. Our findings highlight the important role of co-parenting agreement in a range of parenting practices. Given this, support should consider how parents work in partnership in preference to distinguishing individual roles for mothers and fathers (Ramchandani & Iles, 2014).

Correlates of fathers’ mental health (Aim 2)

Findings supported the hypothesis that fathers’ mental health would be associated with family contextual factors, including child needs and fathers’ employment. Building on past research which has identified factors such as maternal mental health, PSE, and job and relationship quality as risk factors for poor paternal mental health (e.g., Baldwin et al., 2018; Giallo, D’Esposito, et al., 2013), findings from the current study indicate the need for consideration of the role of additional possible influences on fathers’ mental health – factors which may not be modifiable in themselves, but which do indicate intervention points with a view to potentially ameliorating the effect of poor paternal mental health on children. The sociodemographic factors of having a child with complex needs and being in part-time, casual or no paid work were generally associated with poorer paternal mental health. These findings indicate subgroups of fathers about whom support providers should be especially attuned to ‘check in’ on their mental wellbeing, and providing timely and appropriately delivered parenting supports when required.

Correlates of the co-parenting relationship (Aim 3)

We found support for our hypothesis that fathers’ views about the co-parenting relationship will vary in association with family contextual factors such as employment. Furthermore, the data provide novel understandings about influences on the co-parenting relationship, beyond the influence of employment circumstances and each parents’ mental health. Family living arrangements were associated with fathers’ views about the co-parenting relationship. Fathers rated their relationship with their partner or the child’s other parent more positively if they were living with a partner, if they spent more time living with their child or if they had a shared care arrangement for their child (if parents were separated). That is, these fathers felt greater understanding and support, felt the division of parenting labour was fairer, and said they agreed with their partner more than fathers who spent less time with their child, who were un-partnered or who did not have arrangements about sharing the care of their child. This is perhaps not surprising, as these aspects of the co-parenting relationship may be part of the reason why relationships do breakdown (that is, one or both partners not feeling supported, unfair division of household labour, parenting disagreements). Yet, it is of interest that these living arrangements and child contact factors were more strongly related to fathers’ views about the co-parenting relationship than other demographic factors and fathers’ PSE or mental health. These findings extend upon previous research which has linked partner support with fathers’ perceptions about parenting, paternal mental health and fathers’ attendance at parenting programs (Sanders et al., 2010). Extending on these previous findings, the current study suggests that a father’s contact with their child may be critical to having a positive relationship with the child’s mother.

Correlates of fathers’ parenting (Aim 4)

Collectively, findings from this survey help to fill gaps in knowledge about fathers, particularly in relation to the psychology of fathering. The study affords us valuable insights about correlates with fathers’ parenting practices.

Consistent with evidence from the extant PSE literature (Ardelt & Eccles, 2001; Baxter & Smart, 2010; Jones & Prinz, 2005; Rominov et al., 2016), and in line with our hypotheses, a fathers’ sense of efficacy in their parenting was related to key parenting practices. Fathers who felt less effective as parents were also more inclined to yell at and argue with their children, were less likely to use positive behaviour support strategies such as giving rewards, treats or attention for good child behaviour, and were less likely to talk with their child about problems or issues. Furthermore, fathers with higher PSE reported playing more often with their children and reported a more positive parenting approach (i.e., less impatient, more consistent, less critical of child and greater satisfaction with time given to child). Given our inability to draw causal inferences about the direction of the relationship between PSE and parenting practices using this cross-sectional data, we draw on evidence that parenting programs are known to improve PSE while increasing positive parenting (Barlow, Smailagic, Huband Roloff & Bennett, 2014; Sanders et al., 2019) to infer from our own findings that fathers’ sense of efficacy in their parenting may be enhanced through the provision of parenting information and supports that are known to increase positive parenting practices.

Research indicates fathers’ mental health is an important influence on parenting, with associations identified between poor paternal mental health with harsher discipline and less frequent father-child interactions (Bronte-Tinkew et al., 2007; Davis et al., 2011; Giallo, Treyvaud, et al., 2013; Giallo et al., 2014; Giallo et al., 2015). While causality cannot be implied from the current analyses of cross-sectional data, given evidence from longitudinal studies supports a causal link between poor parent mental health with negative parenting practices, and with subsequent impacts on children’s wellbeing (Giallo et al., 2015; Weissman et al., 2006), the rates of mental ill-health among the fathers in the Parenting Today in Victoria 2016 sample (one in five had experienced depression and/or anxiety since having children) are concerning.

Our findings offer novel insights into the specific nature of the role of fathers’ mental health in parenting and reveal some associated factors that may, with further investigation, help illuminate the specific pathways via which paternal mental health impacts on children. When considered in combination with other factors, the role of fathers’ mental health in Parental Approach scores dominated other variables (e.g., child age, co-parenting and speaking a language other than English). However, the role of mental health in the relationship with fathers’ time spent engaged in activities with the child (e.g., playing games or music together) was less important (p = .012) than speaking a language other than English at home and the child being younger (fathers who spoke only English at home and those with younger children spent more time in activities with children). An interpretation of these findings might be that paternal mental health challenges can be viewed as a factor impacting on the style and content of interactions between fathers and their children, but not necessarily as a factor associated with the quantity of interactions. Fathers who experience mental health challenges still manage to spend time with their child, but the consistency of positive interactions may be affected by their experience of mental health challenges. This is important in challenging some of the widely held deficit-focused perceptions about disruptions to fathers’ involvement with their children. While research has found that fathers tend to spend less time than mothers engaged with their children in play and caregiving (e.g., Baxter et al., 2007), these differences are ameliorated once parental employment is accounted for (Giallo, D’Esposito, et al., 2013). Furthermore, reducing the impacts of mental health challenges on paternal caregiving may have greater benefits for children and fathers than emphasising the need to increase fathers’ time spent with the child. In this way, this study challenges the deficit perspective often assumed in discussions about fathers, which tends to focus on the impacts of absent or un-engaged fathers. With future research, including research about how the relationships observed herein vary with different child ages and developmental stages, greater knowledge about the influence of paternal mental health on fathers’ parenting approach will permit improved understanding of mechanisms for improving father-child relationships.

Implications

Findings from this survey illuminate areas for practice reform and further research. There are implications for how we think about and interact with fathers who may be struggling with their own mental health. For the one in five fathers in the sample who were experiencing poor mental health, parenting self-efficacy was significantly lower, and parental approach was more negative. Until recently paternal mental health was relatively under-researched, and routine assessment of fathers’ mental health (e.g., during newborn health visits) is not common and is hampered by the unavailability of valid and reliable screening tools (Fletcher et al., 2015). Given emerging evidence of the positive effects of supporting fathers’ mental health early in the parenting journey (Rominov et al., 2016), early screening and intervention for fathers’ mental health issues is likely to yield benefits for parents and children. Support providers should be particularly attuned to the mental health of fathers of children with complex medical needs or disabilities, single or unemployed fathers, and fathers of adolescents.

While most fathers reported positive parenting practices, compared to mothers they were less confident that they were doing a good job in the role of parent, and as a group fathers were more likely than mothers to describe their parenting negatively (e.g., in relation to being critical, time spent with the child, and talking with the child about problems). These differences between mothers and fathers highlight areas where parenting education and skills development programs can assist fathers, and areas where further interrogation of fathers’ parenting support preferences may be required.

Given the joint responsibility for raising a child (in most cases), attention to how well mothers and fathers support each other is crucial. While causality of the relationships between family living arrangements, the co-parenting relationship and parenting practices or sense of efficacy cannot be determined using the current cross-sectional dataset, our findings have important implications for how we support the co-parenting relationship, especially in situations where parental separation means men have reduced contact with their child, or where the parental relationship is negative (e.g., where family violence or conflict is present). Further inquiry into the co-parenting relationship is warranted, with a view to achieving family arrangements that optimise the division of child caregiving and that minimise conflict between parents. The impact for children will be maximising the opportunity for them to benefit from interactions with fathers as well as mothers.

Limitations

Despite the randomised approach to sample recruitment, a high response rate and data weighting to enhance representativeness, a limitation of this survey was that parents who understood or spoke limited English may not have been recruited at representative levels as the survey was not administered in languages other than English. Also, parents who did not have a landline or mobile number were not sampled. An additional limitation relates to the cross-sectional nature of the data, which limits our capacity to draw causal inferences. Our conclusions are somewhat limited by the sampling of only one parent from each family. Having data from both partners in a family would contribute additional knowledge about mother-father variability, which would be of particular interest for understanding the co-parenting relationship. Furthermore, surveys of this nature are subject to specific threats to validity and common reporter bias as a result of use of a single informant and social desirability bias. The use of individual items rather than scales for some analyses (e.g., the regressions that included examination of the individual co-parenting items) is another drawback, as it increases the risk of Type I error. Finally, factors that were not measured in the survey may contribute unexplained variability to some of the models constructed for the current analyses.

Conclusion

Evidence about the fathering experience afforded by the large, representative sample made available through the Parenting Today in Victoria study extends contemporary understandings about the role that men play in family life. A deficit model often dominates the discourse about fathering within research and the media (Panter-Brick et al., 2014). This deficit frame can portray fathers as ineffective or neglectful in the area of child health and development, and as difficult to engage in parenting support. Results from the 2016 Parenting Today in Victoria survey presents a more positive picture of fathering, illustrating that most Victorian fathers are faring well, using positive parenting strategies, and feel supported in their parenting role. This bodes well for the healthy development of their children, and attests to the accessibility and usefulness to fathers of supports for parents. Nevertheless, the study provided insights into areas where fathers are struggling to meet widely held expectations about good parenting. For instance, the role of mental health in parenting cannot be understated, and factors influencing the apparent mother-father differences in parenting and in views about the co-parenting relationship require further consideration in practice and in research. Future waves of the Parenting Today in Victoria study will present further opportunities to learn more about the psychology of fathering.