Introduction

The co-occurrence of child abuse with domestic and family violence (DFV) is well recognised; however, until recently, research and service responses have mostly been siloed (Renner, 2021). Domestic and family violence creates pervasive harm and presents particularly complex issues for families where abusive fathers have ongoing contact with their children and children’s mothers through families remaining intact, or where child custodial arrangements require mothers to facilitate fathers’ access with their children (Noble-Carr et al., 2020; O’Connor et al., 2020; Scott & Crooks, 2007). In Western contexts such as Australia where this investigation is located, despite public and social will, challenges remain in developing effective practices to abate the unacceptable and high rates of violence against women and children and hold men accountable for their abuse.

A pertinent focus for responding to risk (Kaspiew et al., 2015; Katz, 2015) lies in holding fathers responsible for violence and supporting shifts in their patriarchally entitled attitudes, towards more child-centred parenting, and promoting respectful attitudes and safe behaviors towards women and children (Radatz & Wright, 2016; Safe & Together, 2016). In Australia and other Western nations, DFV is a punishable offence with coercive control increasingly legislated. A central mechanism for responding to perpetrators of DFV has been men’s behavior change (MBC) programs where family services, frontline DFV response services, statutory services and court systems standardly refer men and fathers who use violence. This often occurs while women’s and child protection services work parallel with them to support their safety and agency (Alderson et al., 2013; Humphreys & Bradbury-Jones, 2015). Although MBC programs are the most common non-custodial legal-system referral for men who use DFV in Western nation contexts, there is mixed evidence, mostly determined through partner reports, on their effectiveness for securing men’s change and increasing child and partner safety (Costello, 2006; Day et al., 2018; Featherstone & Fraser, 2012). The focus on fathers as a target group within MBC programs with children as significant beneficiaries of programs has only more recently emerged (Stanley et al., 2012). Given the relative powerlessness of children to prevent violence within families and its substantial impact on them across the lifespan, this development is a crucial, if late, emergence (Noble-Carr et al., 2020; O’Brien, 2016).

Despite consensus on the importance of intervening with fathers who are abusive, there is debate on what constitutes best practice in MBC programs and there is an increasing divergence in research on this topic (Costello, 2006; Day et al., 2018; Featherstone & Fraser, 2012). Velonis et al. (2016) suggest that much of the debate is “as political as it is scientific”, being largely pro-feminist or anti-feminist in rhetoric (p. 2). Beyond foundational theory for MBC programs, broad debates also exist in relation to many aspects of programs, including consideration of the different typologies of perpetrators who may be addressed through nuanced programs, which challenges the current “one size fits all” approach (Bowen, 2011; Holtzworth-Munroe et al., 2000; Wojnicka et al., 2016).

Available research on the effectiveness of MBC programs indicates that characteristics of perpetrators play a role. Better outcomes have been noted among men who are older, have reached higher levels of education, are employed and married, have committed only minor criminal offences, are less involved in alcohol and other drug (AOD) misuse, and experienced less violence in their family of origin (Burnette et al., 2015; Lila et al., 2014). Factors that influence rates of completion, adherence, and program engagement where poor outcomes are associated with domestic and family violence include:

  • timing of attendance, with the first 200 days between offence and intervention commencement identified as most significant for reductions in recidivism (Gondolf, 1999; Laing, 2016; Stansfield & Williams, 2014);

  • length of treatment, where longer program completion is associated with lower rates of recidivism (Arias Ramón and Arce Manuel, 2013; Babcock et al., 2005); and.

  • level of intervention compliance, where early drop-out or sporadic attendance are considered to impact negatively on behavior change outcomes (Diemer et al., 2015).

Another substantial domain of research in MBC is the spectrum of violence typologies for perpetrators of DFV, which may have significant implications for the targeting of interventions, given that high-risk perpetrators present the greatest risk to families for ongoing DFV. Bender and Roberts’ (2007) investigation aligns many of the foundational studies into typologies of violent men providing a summary of common groupings, including:

  • low risk perpetrators, whose behaviors are characterised by low levels of severity and frequency of violence, little or no psychopathology and almost no criminal history;

  • moderate risk perpetrators, whose behaviors are characterised by moderate levels of severity and frequency of violence and moderate to high psychopathology; and.

  • high-risk perpetrators, whose behaviors are characterised by high severity and frequency of violence, high levels of psychopathology, and often hold some criminal record (pp. 522–523).

Some concerning research suggests that men most resistant to change make up the majority of MBC programs clients (Lilley-Walker et al., 2018), while a Spanish study finds that most dangerous participants typically have lowest rate of program attendance, and highest rates of drop-out and violence recidivism (Carbajosa et al., 2017; Ferraro, 2017). While these studies present diverse findings, it is clear that high-risk perpetrators are under-addressed by one-size-fits-all MBC programs. However, targeting programs to perpetrator typologies remains under dispute (Eckhardt et al., 2006; Featherstone & Fraser, 2012).

Many theorists and practitioners have called for programs that are more targeted to the needs of First Australian and ethnic minority group communities (Simon-Kumar et al., 2017; Stathopoulos, 2014; Zakheim, 2011); to respond to key comorbid concerns such as alcoholism, other drug addiction and mental health issues (Featherstone & Fraser, 2012; Ferraro, 2017); or that focus on violent or neglectful fathering and other expressions of DFV (Stanley et al., 2012; Stover, 2013). Given the high level of co-occurrence of child abuse and DFV, fatherhood status offers a crucially important focus for MBC programs that seek to promote greater safety for children who have experienced DFV. While most MBC programs include some sessions which address father responsibility and the impact of violence on children, this is usually limited (Humphreys & Campo, 2017). Yet it is increasingly acknowledged that fathers are often strongly motivated to be better parents, with evaluations indicating improved outcomes for children in father-focussed interventions (McConnell & Taylor, 2016; Stanley et al., 2012).

Developed in Canada for use with populations similar to those in Australia, Caring Dads (CD) is a MBC program that focuses on fathers (Scott et al., 2021). Recognizing that children are often co-abused alongside their mothers, it targets gender-based violence and control, and supports respectful, non-abusive, more nurturing, and child-centred parenting(Scott et al., 2018). Leveraging men’s motivation to be good fathers, the intervention is a 17-week empirically supported group program that works with fathers who have behaved abusively or neglectfully towards their children, and/or children’s mothers (Scott & Crooks, 2007; Scott & Lishak, 2012). The co-parenting relationship is distinct from the adult couple relationship (Teuber & Pinquart, 2010) and affects children’s psychological wellbeing directly and indirectly (Mohaupt, 2020).

The literature indicates that violent and neglectful fathers over-value their parenting (Humphreys & Campo, 2017; Meyer, 2017) and, where separated, perceive their co-parents to be over-responsible for co-parenting difficulties (Holt, 2013; 2018). The CD program aims to assist fathers in both acknowledging their abusive or neglectful behaviors and in making changes to these through group work that employs cognitive behavioral approaches (Scott et al., 2018). The program also aims to improve fathers’ quality of relationships with their children, and with their co-parents, as a fundamental aspect of child-centred parenting. Across international contexts, there is nascent evidence for the program’s effectiveness in reducing incidents of DFV and improving participating fathers’ behavior as key measures of program success, as well as increasing children’s and partners’ feelings of safety and wellbeing, as found in evaluations conducted by Diemer et al. (2020), McConnell and Taylor (2016), McCracken and Deave (2012) and Scott et al. (2021) for populations of fathers in Canada, Germany, and the United Kingdom. However, investigations of the CD program have not previously differentiated outcomes for separated and partnered fathers.

As a substantial response to the problem of DFV and the co-occurrence of child abuse, the authors seek to increase awareness about father-targeted MBC programs and the populations they provide efficacious outcomes for. The current study seeks to address a gap in knowledge about the types of fathers (partnered or separated) who may benefit from engaging in CD. Specifically, it investigates whether attending the CD program, affects improvements in fathers’ co-parenting relationships, reduces their abusive behavior towards mothers and/or children, and improves their families’ level of functioning. The findings of this study are intended to contribute to knowledge about groups of men who may benefit most from typology-targeted MBC programs in order to improve family wellbeing and safety in co-parenting behavior. The authors acknowledge major concerns present in ongoing contact between victim-survivors, their children, and perpetrator-fathers in the wake of DFV. However, they theorise that fathers’ attention to and investment in more positive parenting through father-oriented MBC programs has potential to create a spill-over into greater investment in, value for, and reduced violence and control towards their children’s mothers as well as their children.

Research Design and Method

The research design for the present investigation was mixed method, collecting and analysing both quantitative and qualitative data. The aim of the investigation was to support burgeoning understanding of change dynamics in co-parenting and family functioning for CD participants and their families. Specifically, this research sought to answer the question, ‘does a father’s relationship status, whether partnered or separated at the time of attending the CD program, affect outcomes in their co-parenting relationships and in their families’ level of functioning?’

Fathers were recruited from CD programs in Ipswich, Toowoomba, Warwick, Caboolture and Caloundra regions in Queensland, Australia, from May 2018 to May 2020. Father-participants’ children’s mothers were also invited to participate. Mothers’ reports were included as a more reliable source of data, as commonly supported in men’s behavior change evaluations (Kelly & Westmarland, 2015; Wojnicka et al., 2016). Research was approved by a Human Research Ethics Committee in 2018 (application number 2018/320) with method variations approved progressively.

The theoretical framework for this investigation was the Integrated Theoretical Framework (Gatfield et al., 2022). The Integrated Theoretical Framework is an intersectional, feminist-informed, and systems-oriented model which considers the impact of multiple systems on the emergence and continuation of family violence. In the context of the present investigation, it has provided an underpinning structure for the design, including the selection of quantitative measures that explore interactive family functioning concerns, co-parenting attitudes, interactions, and behaviors. The Integrated Theoretical Framework also guided the development of the semi-structured interview schedule, which explores mothers’ and fathers’ perceptions of coercive, controlling, abusive and violent behaviors in their family. as well as behaviors or experiences that strengthen family interactions and wellbeing. Of interests were families’ experiences of interactions with microsystems (immediate family and inner circle of close friends), mesosystems (i.e. interactions with extended family and friends, schoolteachers, social service professionals), exosystems (i.e. health and welfare services, schools, and criminal justice institutions) and macrosystems (i.e. policy legislation and governing systems).

Three waves of data (pre, post and 10 months post-intervention) were collected. Surveys and interviews were conducted in tandem and predominantly through face-to-face meetings across the five sites. At pre-intervention 33 sets of surveys and interviews were completed, at post-intervention, 22 sets were completed, and at 10 months post-interview, 10 sets were completed.

Qualitative interviews included 16 questions, some were open in style, such as, “Since you completed CD, what has been happening in your family?”; and others more directive, such as, “Since you completed CD, has there been any aggression, manipulation, coercion, control or violence between you and your ex-/partner?”

Quantitative data included four self-report measures to elicit broad understanding of co-parenting and family behavior. These included:

  • The empirically validated and reliable 50 item self-report, the Family Assessment Measure III General Scale (Pellerone et al., 2017; Skinner et al., 1995) used for assessing family system strengths and weaknesses in domains including task accomplishment, role performance, communication, affective expression, involvement, control, values and norms (Feindler et al., 2003);

  • The Co-parenting Relationship Scales, a 12-item short-form measure verified as having good reliability and strong stability with good psychometric and construct validity for assessing dimensions of co-parenting (Feinberg et al., 2012);

  • The Parent Problem Checklist, a 16 item self-report used to assess parents’ perceptions of experiences with their co-parents, which has been verified as valid and reliable for assessing inter-parental conflict about child behavior and parenting (Stallman et al., 2009);

  • The Parenting Scale (PS), a 10 item self-report scale modified from the original long form, to assesses parents’ dysfunctional disciplinary practices expressed through laxness, over-reactivity and verbosity with their children which has adequate reliability and internal consistency (Arnold et al., 1993); and.

  • The Parenting Perceptions Scale and the Parenting Relationship Perception Scale which were developed for use within this evaluation and based on the validated PICCOLO measure (Norman & Christiansen, 2013; Roggman et al., 2013), and from Gottman & Silver’s (2016) research-based relationships program, respectively.

Quantitative data was numerically coded and analysed using the Statistical Package for Social Sciences (SPSS) version 24 software and one-way ANOVA tests were conducted. Variations and trends were determined, with a focus on the variability between separated and partnered fathers and mothers. SPSS analysis assisted in drawing inferences from quantitative findings, with a detailed contextual picture provided when considered together with qualitative findings.

Qualitative interview data was transcribed and organised into question groupings with emerging themes for each question identified, coded, and organised using the software package, NVIVO 25 (QSR International, 2020). Data was analysed using a reflexive thematic analysis method (Braun & Clarke, 2017). To ensure inter-coder reliability, two coders independently coded 15% of questions, with strong agreement at an average of 88% across themes, with differences resolved through additional coding. A third coder successfully applied the codebook to 15% of interview questions, also reaching high levels of agreement (with the third coder finding 100% agreement with coding for 87% of questions and 69–70% agreement with coding for the remaining 13% of questions).

All men who attended the CD pilot program conducted in Western South-East Queensland from March 2018 to June 2019, and the mothers they co-parented with, were invited to participate in the investigation. The wave 1 sample was n = 26 fathers with n = 7 children’s mothers, with a small reduction to wave 2 to n = 17 fathers and n = 5 mothers, and a more dramatic reduction by wave 3 to n = 7 fathers and n = 3 children’s mothers.

Participant recruitment and retention were a challenge for the investigation, with only a small sample of men and their partners willing to participate and with significant attrition across waves of data collection, resulting in a small and diminishing sample group. Attrition of fathers from the program is considered to reflect general program attrition as a result of marginalisation (Gatfield et al., 2022; Hayward et al., 2018). While the sample size for this study was too small for findings of statistical significance to be made, the design had empirical rigor, established through the correlation of qualitative and quantitative data, through triangulation of findings with mothers’ accounts, and through use of a longitudinal design. As highlighted in the literature, women’s reports have longstanding acceptance as being more reliable than violent men’s (Kelly & Westmarland, 2015; McDermott & Meyer, 2019); thus, father findings were compared against mother reports for veracity, creating triangulation of the data. Finally, three waves of parallel-structured data were collected to establish a clear in-depth and long-term picture of change for fathers and their families over time.

Findings

Characteristics of Father Cohort

Fathers who attended the CD program and participated in this investigation (n = 26) were between 24 and 45 years of age (mean age = 34.2 years). 69% of participating fathers were partnered and 31% unpartnered at the beginning of the group. This changed to 57% partnered and 43% unpartnered by the time of the final interviews, representing an over 10% couple relationship breakdown across the evaluation interval. Regarding their employment status, the largest group of participating fathers were unemployed but looking for work (42%), with 42% of the cohort split evenly between fathers working full time and those working part time. In relation to education completion, less than a third of fathers (27%) had completed some high schooling with only 7% of the cohort having completed university study. 92% of participants in the cohort were born in Australia, with 12% identifying as being of First Australian descent. Overall, compared against general population statistics, this father cohort had a low level of school education completion, substantially high unemployment, a high representation of First Australians and a slightly higher than average percentage in couple relationships which dropped back to average by the end of the program (Australian Bureau of Statistics, 2020; Australian Bureau of Statistics, 2022).

Couple and Co-Parenting Relationships

Qualitative and quantitative findings in relation to the nature of couple and co-parenting relationships indicated that mothers and fathers perceived these somewhat differently. Overall, mothers perceived an improvement in their children’s father’s couple and co-parenting relationship behavior over the research period. As mothers’ reports are generally perceived within DFV research to be more accurate than fathers’ (Dobash et al., 1998; Kelly & Westmarland, 2015; Wojnicka et al., 2016), this understanding has been applied to the findings in this investigation, with father reports benchmarked against mothers’ reports. Thus, mothers’ identification of improvements in their ex-/partners’ co-parenting and couple relationship behavior is considered a more accurate and trustworthy finding.

When taken as a whole, mothers’ perceptions as expressed in the quantitative findings were that fathers’ co-parenting relationships improved when they attended the CD program. Themes identified from interviews with mothers also indicated a pattern of continuing improvements in co-parenting across the research intervals. However, when separated out, quantitative and qualitative findings show agreement between mothers’ and fathers’ reports indicating that only fathers who were partnered with their children’s mothers and remained so during their engagement in CD showed improvements in their co-parenting relationships. This was not reported to be true for separated fathers.

In terms of father’s self-perceptions, prior to attending the CD program, partnered fathers rated their co-parenting and couple relationships substantially more positively than mothers had. However, across the research interval, their perceptions changed, converging with those of mothers, such that by 10 months post-group completion, mothers’ and partnered fathers’ ratings closely resembled one another. This is illustrated in Fig. 1, below, which depicts co-parenting perceptions, where partnered fathers’ and mothers’ perspectives aligned with a 12% difference at wave 1, reducing to 9% difference by wave 3. By comparison, separated fathers perceived their pre-intervention co-parenting relationships as problematic, with a 25% difference between separated father and children’s mothers’ ratings. While these perception ratings dramatically improved by the end of the intervention, with only a 2% difference between separated fathers and their children’s mothers, their perception of the co-parenting relationship returned to being substantially negative by 10 months post-intervention, with a 45% difference between separated fathers and mothers.

Fig. 1
figure 1

Comparison between mean of parenting relationship perception scale reported by mothers and fathers who are partnered or are separated co-parents across waves

Full size image

The finding of a convergence between mothers and their partner fathers’ but a divergence between mothers and their separated fathers’ perceptions of co-parenting behavior are mirrored in the findings for the Coparenting Relationship Scale and the Parent Problem Checklist. These findings are also confirmed by qualitative results at a descriptive level. In interviews, partnered fathers described a continuing trajectory of improvements in their co-parenting by post-intervention and to 10 months post-intervention. By contrast, separated fathers described an initial improvement in co-parenting post-intervention but with a reversion to poorer co-parenting or no contact by 10 months post-intervention.

A noteworthy feature identifiable in the findings documented in Fig. 1 is that partnered fathers held an overly-positive, and separated fathers an overly-negative, perception of their co-parenting relationships when compared to mothers, though this phenomenon is less pronounced in fathers’ interviews. This quantitative finding discrepancy at wave 1 highlights potential skewed fathers’ perspectives which are explored further in the discussion in relation to the literature (Broady et al., 2015; Scott & Crooks, 2007; Trounson, 2017; Zalmanowitz et al., 2013).

Consideration should also be given to the possibility that findings may be influenced by men’s continuing participation in CD as a result of a self-selection bias that might favour families who are doing well. However, the exploration of factors that influenced men to drop out of the program requires its own distinct research and cannot be justified here.

Parenting and Family Functioning

Across all parenting and family functioning measures and interviews, mothers reported that, when fathers attended the CD program, fathers’ parenting, co-parenting, and their family’s functioning improved overall. Supporting this, mothers’ responses to interview questions regarding fathers’ parenting behavior indicated substantially poor relationships between fathers and their children at pre-intervention but improving over time. These were characterised initially by violent and aggressive behaviors directed by men towards their children and children’s mothers. However, relationships and interactions were identified as becoming increasingly positive by post-intervention, changing towards relationships that were free from violence and aggression, characterised by support and connection.

This sits in contrast to fathers’ perspectives, where both partnered and separated fathers represented themselves at pre-intervention as reasonably positive parents in their interactive behaviors with their children, despite mothers overwhelmingly indicating fathers as poor parents at pre-intervention. Among the two father subgroups, separated fathers perceived their relationships with their children most positively prior to intervention. However, their perceptions of highly positive parenting appeared to change little between pre- and post-intervention and then to improve dramatically by 10 months post-intervention. By comparison, partnered fathers viewed their parenting initially more positively (than mothers), but not as positively as separated fathers had, though they reported making substantial improvements in their parenting to post-intervention, with comments reflective of increased supportive behaviors, such as “…getting down to their (the children’s) level and talking to them, and explaining why I said no and why my wife said no” and, “If she’s done something silly, I sit down and ask her why she’s done something silly, you know? Rather than just telling her… Now I have more understanding and patience”.

In relation to families’ functioning, mothers and both father subgroups identified improving levels of family functioning across the research intervals with an overall 21% and 22% improvement for partnered fathers and mothers respectively, though only a 12% improvement for separated fathers between pre-intervention and 10 months post-intervention. While all couples in the cohort reported improvements in families’ levels of functioning, the increasing alignment between mothers’ and partnered fathers’ perspectives by post-intervention is highlighted in Fig. 2, below.

Interview findings support improvements in family functioning for all families by 10 months post-intervention with one partnered father commenting:

I have to pat us on the back… in terms of how my partner and I have dealt with things. She has seen the changes in me and how it’s working out for us and the kids … now we know things like, don’t argue in front of the kids. We don’t always sort things out straight away, but we give one another space that we need, a little bit of time apart to make sure that we do come back and fix it instead of just leaving it.

Supporting partnered fathers’ general positivity, mothers also communicated positive changes in their families’ functioning resulting from fathers’ engagement in the CD intervention, with mother statements such as:

Since the CD program … things have been going great. Things have been going really well … it’s been more of (him saying) “What do you think or how should we do this?” He’s been real [sic] open and it’s been good.… It’s been quite positive.… (Since the last time we spoke, post-intervention) it’s been more positive.

Fig. 2
figure 2

Comparison between mean of family assessment measure III general scale reported by mothers, separated fathers and partnered fathers across waves

Full size image

As identified in Fig. 2, improvements in separated fathers’ family functioning in comparison to partnered fathers’ were modest. Highlighting this in interviews, one separated father stated, “I’ve pretty much just been going forward in relation to everything … with the routine, with (children’s) behavioral issues. Everything has just been great …” However, in his relationship with the children’s mother he acknowledged ongoing issues.

For a while there it was going really good. Now that she’s got a partner I think that’s changed a lot of things… It is very rarely when we do talk, but because it is solely for the kids, we don’t have screaming matches or anything on the phone … She is short and abrupt in what she’s stating … And if I ask her how she’s been, she’ll avoid the question.

This illustrates how, while many separated fathers’ made improvements in aspects of their families’ functioning, especially in their parenting behaviors and interactions, not all fathers were able to improve their co-parenting relationship interactions, despite completion of the CD intervention.

Discussion

This investigation examined the impact of the CD MBC program’s intervention on co-parenting relationships, family functioning and wellbeing, focussing on whether separated or partnered fathers experienced important improvements. Overall, findings from the investigation indicated that partnered men and their families benefitted most from the intervention. In the context of this investigation, co-parenting can be understood as the attitudes, values and behaviors exhibited by parents in relation to their children which are intended to support their nurturance, safety and wellbeing and includes the overlap between the function of these roles for a child’s parents, which require them to operate collaboratively with one another. This is distinct from family functioning which is defined as the attitudes, values and behaviors that contribute to various states of family wellbeing including, in this context, dimensions of interpersonal communication, development of family values and norms, internal family system operations and task accomplishment, and individual expression and role fulfilment. The investigation found that the CD program had some efficacy for promoting family safety, functioning and wellbeing for all fathers’ families. For men who were not partnered, the impact of the intervention was not as promising.

The State of Co-parenting Relationships and Family Functioning, Pre-Intervention

Findings from this investigation indicated that, prior to intervention, the state of couple, co-parenting, and family relationships where fathers had perpetrated domestic and family violence was highly vexed. Despite some minor positive, relationship-enhancing behaviors, family functioning and wellbeing were very negatively affected by a father’s use of violent and aggressive behaviors towards his ex-/partner and children. These findings align with and confirm the general longstanding picture emerging from the literature about the pervasively harmful effects that DFV has on children and their mothers and the role that fathers may play in ameliorating harm (Noble-Carr et al., 2020; O’Connor et al., 2020).

The investigation found that, prior to intervention, fathers over-valued their parenting and perceived themselves as generally positive parents, despite having perpetrated negative, abusive, and violent parenting on their children, and exposing them to violence towards their mothers. Mothers’ reports, which provide a more accurate reference point to compare father-reports against (Dobash et al., 1998; Kelly & Westmarland, 2015; Wojnicka et al., 2016), confirm fathers’ poor partnering and parenting. In addition to over-valuing their parenting, partnered fathers over-rated the positive functioning of their co-parenting relationships and their co-parents, portraying them as somewhat positive, while separated fathers rated their co-parenting relationships substantially negatively, describing disrupted, difficult, and conflictual relationships. These findings align with literature in the field, particularly relating to separated fathers’ negative perceptions of their co-parenting interactions, which can be emotionally damaging to children (Holt, 2013; Thompson-Walsh et al., 2018).

Partnered fathers may have been influenced in their representation of an overly positive relationship with their children and partners by the propensity of DFV perpetrators towards denial, minimisation of responsibility and displacement of blame for their violent behavior, which are clearly identified across the literature as a frequent defence among men who use violence and control (Douglas & Fell, 2020; O’Brien, 2016). They may also have held a positive perception bias (Broady et al., 2015; Scott & Crooks, 2007; Trounson, 2017; Zalmanowitz et al., 2013), which can be understood as operating when fathers who have acknowledged violence towards their children and/or ex-/partners still over-value their parenting (Humphreys & Campo, 2017; Meyer, 2017). This positive perception bias may extend or spill over into fathers’ positive perceptions of themselves as good partners and as engaging positively with their co-parent, despite their abusive behaviors (Broady et al., 2015; Scott & Crooks, 2007; Trounson, 2017; Zalmanowitz et al., 2013).

Some feminist understandings of this phenomenon construct a violent father’s positive perception of his couple or co-parenting relationships as possibly relating to his own positive experiences of exerting power and control over his ex-/partner and/or children (Kelly & Westmarland, 2015; Stark, 2009). This understanding may help to explain why separated fathers did not view co-parenting relationship as positively as partnered fathers did, since they would be limited in their capacity to exert power and control over their ex-partners due to separation, although some research suggests that in many cases power and control continues and even escalate post-separation (Stark, 2009). Qualitative findings from this study do not support a clear understanding about the internal or unconscious perceptions or motives of fathers in this regard. However, they do highlight that fathers perceived their own understanding about quality, nurturing parenting and co-parenting to increase across the intervention and reflect on their comparative lack of insight about this prior to the intervention.

There are important cautions implicit in the findings of this study. Fathers’ positive perceptions of their co-parenting relationship do not align with mothers’. As an implication for practice, this reinforces the importance of consulting and listening to mothers and children about their experience of fathers co-parenting, with special attention provided to their wishes regarding contact and concerns in relation to their safety (Alderson et al., 2013; Humphreys & Bradbury-Jones, 2015).

Interview data from separated fathers and mothers identified that separated fathers viewed their co-parenting relationships through a generalised negative lens, referencing conflictual interactions with their ex-partners. This finding supports previous research that finds fathers who commit DFV often perceive their co-parents as bad mothers and blame them for co-parenting difficulties (Holt, 2013; Thompson-Walsh et al., 2018). The tendency for separated fathers to not only perceive their co-parenting relationship negatively, but to hold mothers responsible for problematic co-parenting, highlights ongoing challenges with holding separated fathers responsible for their violence when they are unable or unwilling to acknowledge it. Further, the findings of this investigation support that fathers’ perspectives of family wellbeing and relationships may be unreliable and possibly skewed, at least prior to intervention. This is evidenced in the divergence between fathers’ and mothers’ reports. It is also apparent in the incongruence between fathers’expressed positive perceptions and their concurrent descriptions of harmful and abusive behaviors they perpetrated on their family, which is also supported by findings from other related perpetrator research (Kelly & Westmarland, 2015; Wojnicka et al., 2016). As an implication for practitioners working in MBC, this finding underscores the importance of cultivating fathers’ awareness of their responsibility for violence, utilising tools such as motivational interviewing which have strong evidence for their efficacy and may amplify fathers’ motivation for change (Scott et al., 2018; Zalmanowitz et al., 2013). It also highlights the need for further research to be undertaken to develop more efficacious approaches to promote perpetrators taking responsibility for their harm as a critical goal of MBC prior to engaging in men’s behavior change programs to maximise opportunities for change.

Changes in Co-Parenting Relationships Through Fathers Engaging in CD

While both partnered and separated fathers over-valued their parenting and perceived themselves as generally positive parents, the current investigation found that partnered fathers improved their co-parenting relationships more substantially than separated fathers did, which has implications for how father-focused MBC programs should be targeted in general. This finding, verified by both fathers’ and women’s reports on co-parenting and safety, addresses a significant gap in the literature on the efficacy of father-oriented MBC programs, since the cessation of violence and abuse towards mothers represents a significant goal of the DFV agenda. The findings from the present study provide support for CD as an intervention that can facilitate fathers to improve their co-parenting behavior, reduce violence and increase family safety where couples are partnered. This is especially relevant given the high rates of fathers who continue to cohabit with families in the wake of DFV.

The findings from the current investigation indicate that research is needed to support further development of MBC programs to more effectively address the distinct needs of separated fathers who commit DFV. This requires a better understanding of how their motivations towards their co-parents are likely to be different to those of men in a committed relationship. Emerging research already points to the relevance of considering types of participants and shaping interventions to address their unique needs (Burge et al., 2016; Ferraro, 2017; Johnson, 2008). While the appeal to fathering which is embedded in the CD program may be somewhat effective for supporting improvements in co-parenting for partnered fathers, the findings also suggest that greater intrinsic motivation may need to be generated for separated fathers to experience equal benefit. It is also possible that early changes noted for separated fathers in co-parenting might be strengthened through longer engagement, rather than through different types of programs. While a small number of studies (Gondolf, 2011; Holtzworth-Munroe et al., 2000) have identified differential outcomes among separated fathers attending MBC programs, the evidence remains minimal. The current study strengthens the argument for increased attention to addressing violent fathers who are separated and often disconnected from any motivation of goodwill towards their ex-partners. This is an important consideration since separated fathers often continue contact with their children and, indirectly, their children’s mother through shared parenting arrangements. This is a site where ongoing violence and abuse towards mothers and children is commonly experienced, sometimes increasing, at times with lethal outcomes (Australian Bureau of Statistics, 2016; Kaspiew et al., 2015; Stark, 2009).

Changes in Parenting and Family Functioning Through Fathers Engaging in the CD Intervention

Where fathers engaged in the CD program, both parenting and family functioning improved for fathers, mothers, and children; and these changes were sustained to 10 months post-intervention. This finding supports existing evidence of improved safety for women and children where fathers attend CD (McCracken & Deave, 2012; McConnell & Taylor, 2016; Scott & Crooks, 2007) but also of improved and safer fathering and overall increased positive internal system functioning for families. As an implication for practice, the CD program is recommended as an intervention for addressing domains of parenting and family functioning where father-violence is a presenting issue. While, as identified earlier in this discussion, the program should not be recommended to improve co-parenting relationships and behaviors where fathers are separated from their co-parents, it does seem to impact positively on partnered fathers’ co-parenting relationships as well as improving their family functioning for all participants. At a policy and practice level, shifts should be made to adopt CD into the range of MBC programs offered to fathers in Australia in order to address whole-of-family wellbeing and safety. This is especially pertinent for families who remain together or in contact, as both present risk where fathers’ domestically violent behavior remains unaddressed (Kaspiew et al., 2015; Katz, 2015).

The disparity between partnered fathers’ and mothers’ reports, which almost disappeared by post-intervention in relation to most aspects of parenting and co-parenting, adds nuanced understanding to the well documented tendency for perpetrator fathers, pre-intervention, to minimise or divert responsibility to mothers and children for violence and parenting issues (Broady et al., 2015; Humphreys & Campo, 2017; Meyer, 2017; Scott & Crooks, 2007; Trounson, 2017; Zalmanowitz et al., 2013). Further, the findings of increasing alignment in partnered parent perspectives, supported by increased positive communication and shared language around co-parenting, provides substantial information about the ways in which co-parenting may have improved interactively for partnered couples. The implications for research and practice with violent men remain, that women’s reports should continue to be used as a benchmark for veracity (Kelly & Westmarland, 2015; McDermott & Meyer, 2019). Second, the convergence of father and mother perspectives indicates an important and positive shift for partnered couples, where there is increased mutuality in co-parenting which contributes to conferred positive outcomes for children in experiences of being parented. This is highlighted as an important focus for continuing research which may be capitalised on to improve the value and efficacy of MBC programs for partnered men who have used DFV. However, it also highlights the limitations of such programs for separated fathers and of the importance of further research and men’s behavior change programs development to address this typology (or subgroup) of men.

Limitations

This study had a small sample size and high attrition, limiting its generalisability. Attrition of fathers reflects general MBC program attrition as a result of marginalisation, including concerns such as transport and financial issues, insecure housing and alcohol and other drug misuse (Gatfield et al., 2022; Hayward et al., 2018). Further, it is acknowledged in other investigations that attrition of mothers from studies is a predictable phenomenon given that they commonly disengage from related services following a partner’s program completion (Diemer et al., 2015). Nevertheless, the study uses rich qualitative data over three data collection phases to strengthen the method. The study does not draw on formal records from police or court systems for recidivism data, which are commonly sourced in men’s behavior change programs evaluations. However, using mothers’ self-report data to cross reference fathers’ reports has strengthened the method. This may have been more effective than incorporating recidivism data since mothers’ self-reports are often understood to be more comprehensive and accurate than official court and police data (Cheng et al., 2019; Lila et al., 2019). Finally, the current investigation does not include direct children’s perspectives, which limits its representation of whole family perspectives. The current investigation provides a preliminary understanding of potential positive outcomes from the CD program for partnered fathers that should be viewed as indicative rather than definitive. Significantly, it builds the case for further research to extend the findings utilising larger cohorts, greater stakeholder input and inclusion of the voices of children.

Conclusion

This investigation has focused attention on an efficacious MBC intervention, Caring Dads, which has potential to support improvements in co-parenting and family functioning where fathers who use violence and remain in, or maintain contact with, their ex-/partners and children (McConnell & Taylor, 2016; McCracken & Deave, 2012). This is an important area for intervention as families that remain in contact with perpetrators post DFV are at high risk for further abuse. Further, coercive control and systemic disengagement can contribute to escalating violence in families, while targeted intervention for fathers can help contribute to the de-escalation of violence and healing for harmed family members, particularly children. The current investigation has provided support for existing findings of the significant impact of father violence on families where there is no intervention (Noble-Carr et al., 2020; O’Brien, 2016). It has also explored some of the potential psychological obstacles to fathers’ taking responsibility for their violence prior to intervention. These include biases in their beliefs that support a positive perspective of themselves; poor recognition of the harm they have caused; displacement of responsibility for their harm towards ex-partners and children; and minimisation of their personal responsibility. The study also provides strong support for the importance of listening to women’s and children’s voices to understand their experience and verify men’s reported change. It also reaffirms the significance of supporting perpetrators to take responsibility for their violence as a precursor to change-work.

Due to the limited sample size and pre-post, self-report nature of the design, findings must be viewed cautiously, and as early investigative evidence. The study found differential results for the effectiveness of CD for separated compared with partnered fathers. It indicates that partnered fathers are more likely to make sustainable positive change in their parental alliances and family’s functioning. This investigation draws attention to the significant need for targeted referrals to MBC programs; that is, programs which specifically address men’s typologies, and for future research to support program development for perpetrator-specific typologies, highlighting the subgroup of separated fathers which, in the current investigation appear intransigent in attitudes and behaviors towards their coparents, possibly implicating features of program length and modality. This study emphasises the potential positive impact on children, their mothers and whole family functioning when fathers engage in CD and supports existing calls for shifts in practice to support fathers to safely parent where they engage with their family in the wake of perpetration of DFV.