The shadow fathers: Barriers to whole family approach in social work?

Objective: The paper aims to explore the barriers to father involvement in family social work at the personal, cultural, and structural levels, and their interconnectedness to answer the main research question of how gendered constructions of mothers and fathers in social work affect the involvement of fathers in interventions. Background: This paper contributes to debates about fathers in social work by examining gendered constructions of fathers in social services for families with children in the Czech Republic. The problem is viewed from the perspective of social constructionism, anti-oppressive practice, and the whole family approach. Method: To meet the objective, we used a qualitative research strategy. 44 in-depth interviews with social workers, mothers, and fathers were carried out. After that, three focus groups with 21 social workers were formed. Results: The results show that mothers and fathers are treated differently in social work, and femininity and masculinity are perceived as dichotomous opposites. This hinders fathers’ involvement in family services and puts responsibility for solving family problems to mothers. Conclusion: Most of these barriers at the micro- and mezzo-levels are in some way related to system barriers at the macro-level, which are often shaped by gender discourses. case on in this way. “...the office for social and legal protection of children primarily works with the mother and refers some part of it t o us to deal with directly along with the mother’s contact. We all meet – an office for social and legal protection of children, mother, and us, so they hand over the model where they work with the mother” (Radmila, social worker).


Introduction
Fatherhood is socially and culturally constructed (Lupton & Barclay, 1997). Its definition changes over time and in different geographical contexts (Lucas et al., 2021). Today's conception of fatherhood in modern Europe is more fragmented than motherhood (Sheldon, 2005) and is becoming increasingly controversial (Ewart-Boyle et al., 2015). The "culture of fatherhood" is changing, and these changes have implications for social work practice (Shapiro & Krysik, 2010).
Fathers' perceptions and their experiences with social work are considered of a peripheral importance in general. Studies on fathers in social work tend to focus on issues such as child removal, domestic violence, divorce, and unemployment. The social work literature has only recently begun to focus more on the difficulties fathers encounter in family services and their involvement in interventions (Baum, 2016).
Father involvement in social work is a complex and multidimensional construct. There is a range of evidence about under-representation of fathers in childcare and protection services (e.g., Scourfield, 2003;Brown et al., 2009;Clapton, 2009;Coakley, 2013;Ewart-Boyle et al., 2015;Philip et al., 2018;Nygren, 2019;Critchley, 2021). Previous research shows that child welfare services focus primarily on mothers and not involving fathers. The focus on mothers can be seen in the content of interventions, the presentation of services, and the high level of presence of client mothers in the service (Sicouri et al., 2018). Mothers are perceived as the ones responsible for their child's well-being (Sinai-Glazer & Peled, 2017). Davies and Krane (2006) refer to the culture in child protection services as "mother-blaming". D' Cruz (2002) also confirms the gendered constructions of "responsibility" attributed to mothers and "invisibility" attributed to fathers in social work practice with families.
Despite the neglect of fathers by social work, they play an important role in the lives of their children (Flouri, 2005). Research suggests that fathers' involvement can be critical in enhancing the effectiveness of an intervention (Pleck, 2010;Lamb & Lewis, 2013;Wingrove et al., 2016). In relation to the lack of father involvement by social workers, Coakley (2008) found that where fathers were involved, more children achieved shorter stays in foster care. Marshall and English (2001) concluded that higher levels of father involvement were associated with lower maternal depression as well as better outcomes for children. Other research has suggested that greater father involvement is associated with children's overall well-being and reduced risk of neglect (English et al., 2009) as well as reduced risk of poverty (Blanden, 2006). Ferguson and Hogan (2004) found a connection between father involvement in interventions and lack of inclusive practice in family social work. Whilst the rhetoric is the 'whole' family, social workers prefer a conceptualisation of the family that assumes the primacy of mothers (Lee, 2014) and their role as gatekeepers (Schoppe-Sullivan et al., 2015). This leaves fathers as an untapped resource in strengthening families, minimizing risk factors, and increasing protective factors to improve family resilience.
Decision-making processes at the practice level in social work are highly dependent on the social context in which they operate (Hämäläinen et al., 2012). Legislative conceptualisations of "fathers" and national policies on paternity leave are key indicators of how the state perceives fathers and fatherhood. Although the paternal role is multi-faceted and includes important emotional and social resources, among others, current expectations of fathers are still primarily focused on their financial obligations as breadwinners (Sicouri et al., 2018). Public policy emphasis often concentrates on identifying fathers in order to enforce child support payments (Day & Lamb, 2004).
The barriers to fathers' involvement in social work exist at different levels (individual, organisational and societal) and are multiple and complex. The social worker approach and/or their lack of competence are among the most frequently cited barriers to fathers' involvement in social work (Coakley et al., 2018;Tully et al., 2018;Lechowitz, 2019).

Whole family approach
The importance of engaging with both parents is supported by the promotion of a "whole family" approach, but one obstacle to achieving this has been the difficulty in engaging fathers who continue to be underrepresented in family social work (Symonds, 2018).
Family Systems Theory frames all members within a family as interdependent (Broderick, 1993). Conceptualizing the family as a system, facilitates understanding of the wide-ranging impacts of unfavourable social circumstances on all members and the external systems with which families interact. This perspective reinforces the importance of the needs of the family as a whole to best support the functioning of the family and the well-being of its members.
Whole Family Approaches argue that interventions cannot be successfully implemented with any family member in isolation from others (Morris et al., 2008). These are interventions that involve familycentred, family-focused, family-based social work. Family-centred social work practice includes the following attributes (Allen & Peter, 1996): the family is a unit of attention; services are provided in accordance with the family's decision; it builds on the family's strengths (includes empowerment); the relationship between the family and professionals is based on partnership (equality, reciprocity, and teamwork); it involves a holistic view of the family in terms of "circumstances, barriers, and resources" and the availability of services for all family members; and it involves individualised services that meet the family needs.

Czech context
The Czech Republic has long been criticised for the high number of children in institutional care (UN, 2011). Due to transformation of institutional care in recent years we can observe a new emphasis on child removal from a family being the last resort to deal with the family situation. At a time when the removal of children from their families is perceived as extremely unsuitable, the best interests of the child are protected by the use of support services to work with the whole family through the activation of its members.
Czech social activation services for families with children are particularly focused on working with the whole family. This is a social prevention service that provides support to families with one or more children under the age of 18 whose development is at risk due to a long-term, unfavourable social situation (poor housing conditions, unemployment, debts, etc.). The parents are unable to overcome this situation on their own without assistance. As of December 2021, the Czech Republic registered a total of 250 social activation services for families with children. As a rule, social activation services are provided by non-governmental non-profit organisations. However, they cooperate closely with the departments of social and legal protection of children, which are state bodies (run by municipal and regional authorities) with powers under Act No. 359/1999 Coll., on Social and Legal Protection of Children.
Clients of social activation services for families with children are often people experiencing varying degrees of exclusion in society, which contributes to problems with raising children and functioning of families. In the Central and Eastern Europe, this exclusion concerns to a large extent Roma families and is based on a high level of racism toward Roma both at the societal and institutional level (Cashman, 2017;Prokop, 2019;Creţan et al., 2022). In the Czech post-socialist context, this form of discrimination has historically manifested itself primarily through the emergence of socio-spatially excluded neighbourhoods, inhabited mostly by Roma (Urban, 2015).
In the Czech environment, socio-spatially excluded neighbourhoods take the form of individual or residential houses, streets, or entire parts of cities (Office of the Government of the Czech Republic, 2016; Gojová et al., 2021). They can be defined as a concentration of more than 20 people living in unsatisfactory conditions and occupying a physically or symbolically enclosed space (GAC, 2015). In addition to spatial segregation related to discrimination on a housing market, Roma people are also discriminated on a labour market and by a school system (Cashman, 2017;Prokop, 2019), which is why they often find themselves in unfavourable social circumstances making up for most of the clientele of this type of service.
Many authors describe the phenomenon of racial discrimination against Roma in the context of Central and Eastern European countries as antigypsyism or anti-Roma sentiment. According to Nicolae (2007), antigypsyism is characterised by various forms of discrimination directed at Roma, such as hateful attitudes, exploitation, and discrimination, which are based not only on racial prejudice but also on specific stereotypes based on historical and cultural beliefs about Roma as an ethnic group. Some stereotypes concerning the functioning of Roma families, communities, and gender roles within them can also be adopted by social workers, including attributing the impacts of discrimination to an inherent nature of Roma people (e.g., unwillingness to cooperate with social services, little effort to provide adequate housing for the family, etc.).
Thus, the Roma origin of their clients is sometimes culturally-deterministically seen by social workers as the cause of perceived inability or undesirability of changing the level of father involvement. A similar ethnicizing approach is also manifested in the interpretation of social problems on a broader scale, which has already been critically reflected in the Czech context by some authors (Doubek et al., 2014;Frýbert & Pařízková, 2014;Kozubík et al., 2018). Valová & Janebová (2015) add that many social workers recognize structural oppression of Roma people, but Czech social work lacks the instruments to respond to it from the position of social services.
To outline the context of the construction of fathers in Czech social work, we need to make a few remarks about the client group profile. This group concerns men who are often unemployed or working illegally, living in insecure housing conditions, and lacking financial resources. These are men described by Connell (2005) as marginalised masculinity, a form of masculinity that is unable to reap the benefits of hegemonic masculinity. It includes, for example, belonging to a minority or having a disability.

Theoretical lens
To understand how barriers to father involvement are interrelated, we used Thompson's P-C-S analysis (2012) for inspiration and analysed the interactions between clients and social workers at three levels. These three levels (P, C and S) are closely linked and constantly interact with each other. Thompson (2012) assumes that our thoughts, emotions, and attitudes about particular groups in society are shaped to some extent by our experiences at a personal level. Personal level refers to the personal or psychological; it is the individual level of thoughts, emotions, attitudes, and actions. It also refers to the interaction of social workers with clients and to bias, inflexibility of one's mind that stands in the way of fair and non-judgmental practices.
Cultural level includes shared ways of seeing, thinking, and doing. It is primarily about shared meanings. It refers to commonalities -values and patterns of thinking and behaviour, an assumed consensus about what is right and what is normal; it creates conformity to social norms. A cultural level is a complex network of taken-for-granted assumptions or "unwritten rules". Culture is very influential in determining what is perceived as "normal" in any given situation.
Structural level refers to the network of social divisions and the power relations that are so closely associated with them; it also refers to the ways in which oppression is "institutionalized" (fixed through patterns of thought, language, and behaviour). It refers to the broader level of social forces, the sociopolitical dimension of interconnected patterns of power and influence.
Our thoughts, actions, attitudes, and emotions are to some extent unique and individualised, but we must also recognise the powerful role of culture. The cultural level creates the context in which our personal experience is formed. Society not only controls our actions, but retroactively shapes our identity, our thoughts, and our emotions. Therefore, the paper is intended to explore the barriers to father involvement at each level and their interconnectedness as well as to learn how gendered constructions of mothers and fathers in social work affect the involvement of fathers in interventions.

Data and methods
The paper was written based on findings from the qualitative research study focused on understanding of the conditions and their interconnection that influence father involvement in social services in addressing the family's unfavourable social situation. The paper is based on partial data from the research. The research was subject to the University research ethics approval by the University of Ostrava. The research aimed to find out what conditions form the situation of (non)involvement of fathers in social activation services for families with children. The research study is part of a two-year applied research project aimed at developing methodology and training of social workers to support the involvement of fathers in family social work. We used a qualitative research strategy appropriate for use with topics that are complex and sensitive and requires information about personal and lived experiences of participants (Ritchie & Lewis, 2003).
From September 2020 to January 2021, the data were collected using face-to-face in-depth interviews. Interviews with elements of the ORID method (Stanfield, 2000) were used. ORID (Objective, Reflective, Interpretive, Decisional) allowed us to structure and naturally develop the conversation through a targeted discussion.
Data collection in this qualitative survey was carried out in two regions of the Czech Republic (Moravian-Silesian and South Moravian). The Moravian-Silesian Region counts amongst the regions with the highest rate of social exclusion and the highest number of socio-spatially excluded neighbourhoods inhabited mainly by Roma, where social services are conveniently concentrated (GAC, 2015). The Moravian-Silesian Region is also the region with the second largest number of social activation services for families with children in the Czech Republic. The South Moravian Region, in turn, is among the regions with an average number of socially excluded localities and concentration of services for families. We selected the two regions to include different levels of reported social exclusion.
We used purposive sampling when addressing/involving non-profit organisations providing social activation services to participate. These were mainly (but not exclusively) organisations that participated in the research project as application guarantors 1 . Some of the organisations work with socially excluded families (one is focused on supporting Roma families), whereas for others ethnicity is not a criterion of support. The fact that the majority of clients are Roma is influenced by the place of activities of the organisation (socially excluded neighbourhood).
The purposive sample included fathers, mothers, and social workers. 44 participants included 11 fathers, 14 mothers, and 19 social workers. Eligibility criteria included: social workers in the position of social workers or social managers in a registered social service that has a valid registration for social activation services for families with children according to the Registry of Social Service Providers. The selection criterion for mothers and fathers is that they are clients of the service, or their partner (spouse) is a client of the service.
Social workers were approached through collaborating organisations, while clients were approached by these social workers. The social workers handed them an information flyer about the project and the data collection. Social workers then either provided researchers with telephone contacts of clients who agreed to be interviewed to arrange interviews with them on their own, or the social workers accompanied the researchers to the families. In two cases, a social worker was present at the interview at the client's request; in other cases, the social workers departed to another room.
The research participants were over 18 years of age, and their native language was Czech or Slovak. Social workers were employed in a total of six non-profit organisations.
Interviews lasted between 13 and 81 minutes, averaging 34 minutes and were audio-recorded with participants' consent. Prior to each interview, researchers explained the study objectives and ethical issues. Pseudonyms were used to identify the participants.
In April 2021, we organized a total of three focus groups with social workers (n=21), each lasting two hours, to complement the picture obtained from the analysis of individual interviews. The focus groups were attended by social workers from collaborating organisations. Every focus group was moderated by a pair of moderators (male and female). The themes for focus groups emerged from individual interviews. We chose the focus group technique because, unlike individual interviews, it allows data to be generated through interaction between participants (Morgan, 2001). During focus groups, participants were able to verbalize their opinions and attitudes in response to the statements of others.
The data was elaborated using the grounded theory by Charmaz (2006), particularly initial, focused coding and memo-writing. We used the computer software program ATLAS.ti to support organisation, analysis, and storage of the data. Credibility of the data was ensured by audio-recording and verbatim transcript of the interviews and discussion of themes and interpretations among team members on multiple occasions (Guba & Lincoln, 1994;Thorne et al., 2004).

(Non)involvement of fathers at personal (micro) level
Social workers shared with us their personal life and professional experiences that have shaped the way they view fathers and fatherhood. Social service users shared their personal experiences of social services and perceptions of their involvement in interventions.
Social workers reported that it was important to them that the whole family was involved in interventions, and at the same time, they considered the involvement of fathers to be desirable. "... we work with the family, and it is important for us to meet the whole family" (Radmila, manager). "We try to make it sort of comprehensive, because the father is part of it. The father is there and should be there" (Šárka, manager).
Although the importance of working with the whole family was declared, some social workers confirmed that they mostly paid attention to mothers and children in their practice. "...experience shows that women work with us in 90% (of cases)" (FG1).
Jana suggests that there is room for discretion on the part of social workers in this regard and the identification of family/parenting with the mother when she says: I think it's up to us to set how we do it, and to what extent it's going to be the whole family, or if it's just mother and kids… nobody differentiates between family and mother and kids…it's all the same…it means that we as social workers don't deal with it anymore, and we do what we can" (Jana, social worker).
Social workers' attitudes towards father involvement in interventions are to some extent shaped by their experiences at a personal level (see also Cryer-Coupet et al., 2021). "Certainly, when you work with these guys, you compare it to your own family, your own children, your own parents. It certainly has an effect on that" (Monika, social worker). According to the participants, personal experiences from their own family may reflect on the social workers' relationship building with fathers and how much attention they pay to them.
The relationship with social worker's own father can hinder ("I'm used to living without a father all my life … so it would be really weird if I didn't pass it on somehow, not adopting it to my practice… I don't think that's possible", FG1, social worker) or support the involvement of client fathers in the collaboration ("...my colleague who really works with all the grandmothers, grandfathers, fathers, mothers, she's from a family where her father was super active... and she finds it weird that father would be sitting unattended, somewhere on the side", Šárka, manager).
Similarly, Baum (2016) reports that daughters who have a positive relationship with their father are more likely to have positive relationships with other men. Social workers' performance in the profession is influenced by their personal experiences and it is therefore important that they are aware of the impact of these experiences on their professional behaviour. In doing so, non-reflected stereotypical thinking in gender relations can become a source of oppression (Hicks, 2000). For example, Monika talked about the risks of not reflecting on one's personal experiences and values: "I grew up in a family where I never even got slapped by my dad...and my husband never hits our kids, so I have kind male role models around...they're not aggressive, and yet I personally have a problem with that internally. When this woman sees me and she's sharing it (domestic violence) with me, it's internally irritating to me.... I never know what to do…I try to stay professional, so I never push them into anything… and I don't try to impose my opinion on them" (Monika, social worker).
Social workers' non-reflected projections of personal experiences, values and norms may be reflected in the evaluation and categorisation of clients. These abstract categories are created by classifying and grouping according to specific characteristics, which in this case were most often based on gender or ethnicity (Jacob, 2004). Categorisation provides workers with an orientation to social reality, saving time and energy in learning about it. The main binary categories into which social workers ranked their clients were the Roma-non-Roma, male-female and father-mother. In doing so, they often constructed them as asymmetrical and dichotomous categories.
Roma families were described as more traditional, with strong gender-differentiated roles. "There is a strong division of what a woman can do and what a man can do…they have it a bit different from the majority [of population]" (FG1). Most mothers were described as active, emotional, fragile, self-sacrificing, cooperative, etc. Fathers were constructed as aggressive, lazy, irresponsible, addicted (to substances), withdrawn, dangerous, etc. Mothers represent the main clients ("she is a key person"). The construction of the mother as the main client is linked to a perceived gender-traditional division of roles in the family ("it's usually the woman who sorts these things out"). The father is perceived as a secondary client. This is justified by his absence from a household. "…when we go to the household we most often talk to a woman, because either a man is not at home at all or if he happens to be there, he usually tends to hang out in another room, as he doesn't want to participate" (Kateřina, social worker). Fathers' absence was most often explained by their role as breadwinner. "…it's due to the fact that he is at work and can't deal with it…, because that they're the breadwinners" (Gabriela, social worker).
Other justifications for the father's absence were imprisonment ("...there are families where the father of children just isn't there, or else he's in jail", Gabriela, social worker), family breakdown, and male autonomy ("...they pretend to go out for a beer", Zora, social worker).
The data supports the thesis that the absence of men or the perceived specific role of women in Roma families and the related form of their cooperation with social and activation services are interpreted by social workers primarily through an ethnic differences of Roma people lens. This view sidelines interpretations based on institutionalized racism and inequalities in Czech society and especially on the intersectional burden of Roma families in an environment of antigypsyism. In other words, social workers tend to interpret specificities through ethnic differences rather than through a combination of different discriminatory conditions. An ethnicizing view of the social problems of families, and on the barriers to the father involvement or mother gatekeeping, may divert attention away from a contextually sensitive view of the situation of Roma families in Czech society thus preventing overcoming barriers to cooperation with the whole family.
Social workers described the construction of fathers as invisible (shadow fathers). They described encounters with such fathers as a physical act of separation. They cited a number of cases where shadow fathers, at the moment the social worker entered a flat, would go in another room or another part of the flat ("They'll just go somewhere else", "they'll make a cup of tea and then go sit and watch TV", "he hides in the bedroom, doesn't say hello, takes his mobile with him, lies there on the bed with the door closed", "he prefers locking himself in another room and actually doesn't come out at all", social workers), or would leave completely ("My experience is that when they occupied a 1+1 flat, her partner always left when I came in", Monika, social worker).
Mothers offered a similar description. "He hangs out with us for a while and then we [mother and social worker] go to the room" (Joanna, mother). Joanna claims that, even when the men separate, they monitor the situation. "He's in the bedroom, and sometimes can hear us." The polarisation between mother and fathers leads to the accentuation and deepening of differencesborderwork (Thorne, 1993), with the formation of a mother-worker coalition, facilitated by the knowledge of the family problem only from the mothers' perspective ("whether we want to or not we are on her side because we know her perspective", Radmila, manager). This results in blaming the father for the family's problems, the father (not involved) is perceived as both the cause of the family's problem and a barrier to its solution. One of the social workers, for example, said: "The father goes to a pub and leaves all the burden on the mother. It's actually a cycle, but if he cooperates more closely with the mother, it's not such a burden on her" (Alex, social worker).
Based on gender constructions of mothers and fathers, social workers applied different normative measures to mothers and fathers. The motivation for normalisation is a pursuit of clients' social functioning, and the goal is to achieve norm conformity and homogenisation through control of normality. At the micro level, normalisation manifests itself as a supervision of a norm compliance (e.g., the "good mother" norm) that is related to traditional conception of the family. Practices based on normative assumptions and behaviour ideals represent a means of reproducing gender differentiation and inequalities (Zábrodská, 2009).
The norm of being a "good mother" can influence mothers' perceptions of father involvement. Mothers often act as gatekeepers of worker access to a child's father (Maxwell et al., 2012). Traditional attitudes of mothers regarding parental roles predict a greater likelihood of maternal gate-closing behaviour (Schoppe-Sullivan et al., 2015). Maternal gatekeeping gives mothers the power to make decisions about father involvement in interventions (Malm et al., 2006). The role of the mother can act as an important source of strength and self-esteem for many women. Our participants reported that a mother, as a primary client, decides whether or not a father is included in the service contract ("Since it's usually the mother who decides on whether we include the father in the contract or not…it's primarily the mother in the contract.", Jana, social worker). The mother is the one who decides what a social worker learns about the family as well as what the father learns about working with a service. The mother can then support or hinder the social worker's contact with the father ("sometimes the guys are part of it, if the mother doesn't mind, and then I don't mind either", Egon, social worker). Most mothers we interviewed reported that their partners are interested in the collaboration, and they pass on information to them ("…yes, he asks about it. I say yes and that it's good", Beatriz, mother). Fathers also confirmed this to us ("…my partner always deals with it…when I come home, she tells me what happened and what's going on", Jáchym, father). At the same time, the women admitted that not all information reaches the fathers ("When I remember telling him, I tell him some stuff, but not everything", Sára, mother). One grandmother referred to traditional male conception of a female role in the family without identifying with it in any significant way ("...it's your job...that's his opinion", Svatava, grandmother).
Because of the gender-based construction of fathers as dangerous, threatening aggressors, social workers subjectively felt fear of involving fathers because they did not feel safe in their mutual interactions ("…as long as you don't turn your back at him"). Róza, on the other hand, reflected the fear that stemmed from the threat to the rest of the family in her absence, when she proclaimed: "I sometimes perceive the presence of her partner as dangerous and it's uncomfortable for me to leave. I actually mean -leaving the others there because I feel that they might be in some kind of danger" (Róza, social worker).
This subjectively felt fear, based on institutionalised constructions of fathers as dangerous, may prevent social workers from approaching them for collaboration. At the same time, social workers admitted that they had not yet experienced any dangerous situation. "We haven't been really assaulted" (Šárka, manager). Šárka points out that if men cooperate with the service, they are not dangerous to social workers: "…And once they cooperate with you, they won't hurt you. Like, I've been here for five years, nobody's ever assaulted anybody here." Despite this, the workers emphasise the need for an increased sense of security and the need for self-defence (self-defence courses, pepper spray). This also underpins the differentiated behaviour towards mothers and fathers within the organisations "…if the client is a man, the staff always go in pairs to see him…they tend to be afraid, because they don't know the person that much, so they are afraid of what reactions he might have or how he might act" (Jiřina, manager).

(Non)involvement of fathers at cultural (mezzo) level
Social workers' attitudes cannot be isolated from the broader cultural context (Ewart-Boyle et al., 2015). The mezzo level of father involvement in social service includes answering the question: How are things done in our organisation? It represents a set of values, norms, beliefs, attitudes, and assumptions that are not necessarily explicitly formulated, but which determine the way people behave and act and the ways they conduct their work. At the mezzo level, social workers in the interviews particularly indicated four areas that reflect an existing organisational culture of social services. These were namely service setup, i.e., codified and formalised regulations, as well as written and unwritten beliefs, i.e., what is traditional in the organisation, e.g., family concepts, client characteristics. Service setup and beliefs were then reflected in established work practices and communication forms (within the team, with each other, with management, with users, etc.). "I find it a very interesting thing because we actually create it internally within the team and then we disseminate it in some way, spread it to our colleagues and to our clients" (FG2).
In terms of a service setup, the involvement of fathers in organisations was mainly influenced by contract setup, location, and timing of the intervention. This is also related to the target group definition, the number of families (workload) per worker, the way the service is approached and the way the service is presented.
Even though social workers themselves identified the service definition as one that works with the whole family, they admitted that in reality the services are feminised, oriented towards mothers. Although feminisation of a profession is more of a macro-level issue, it also significantly affects the organisational culture. Feminised and maternally oriented services do not seem to meet the needs of fathers (see also Ghate et al., 2000). In this context, female social workers have thematized the need to employ male social workers. On the other hand, they expressed doubts about the ability of male social workers to be able to advise mothers on the care of especially young children ("Because I thought those guys couldn't handle it", Klára, social worker) and expressed concern about jealous reactions of male clients, especially among Roma families: But it's hard when you suddenly have to send a man to a Roma family where there is a mother and children. Her partner is at work earning money. And in his home, there's some strange guy visiting his wife giving her advice on how to bring up their kids... this situation... is not very good at all" (Šárka, manager).
On the management part, the neglect of male social workers can also be interpreted as a threat to their own role, with men typically being in leadership positions in social services (see also Christie, 2006). Female workers often spoke of male workers in hypothetical terms. Their accounts regarding men's lack of aptitude for this kind of work was not supported by personal experience. Their concerns about male client jealousy in the accounts of mothers were mostly not confirmed: "I don't think he was jealous. He's not like that with strangers, only when male friends come... friends and such, so then he's totally jealous... (sigh). Yeah, but not otherwise. It's what I think. I don't know, he won't tell me" (Amálie, mother).
The social workers based their stereotypes about male social workers on the essentialist statement that "a woman better understands another woman" or that "a mother better understands another mother". Róza, the social worker supported this by stating: "...actually building a woman to woman relationship... it seems to me to be easier in terms of certain attunement and understanding." and Kateřina (social worker) added: "I think one gets along better with the mum, maybe because we are women… it's just different...". Starting from such an assumption is risky, if only because of the diversity of womanhood and motherhood. Moreover, this attitude leads to the exclusion of a male subject from interventions and the construction of his inferiority (Zábrodská, 2009). It also complicates the involvement of male fathers in interventions.
The idea of father's no use during interventions can also be supported by mothers themselves, who refer to ethnicity: "With us it's mostly like ... not just with me, but with all of us, with the Roma… It's like, we have it the way that the guy takes care of ... let's just say the work, money…and the household, children and so on -that's our business. And of course, if I have something else to do, he plays with them, takes them out and helps me out…but not much. It would be nice if it was more" (Amálie, mother). Such explicit affirmation from female clients fits with the interpretation of cooperation as "the woman better understands another woman, and even more so, when she works with Roma" "... this strong division between what a woman can do and what a man can do ..." (FG1).
The feminisation and orientation towards mothers are reflected in a service setup that discriminates against fathers. An example of this is the contract setup. In order for workers to at minimum pass on information to the father, he must be listed in the service contract. The contract signing usually takes place in the first meeting. Given the perception of a mother as a primary client, the contract is usually only signed with her. Even in cases where men are present at the meeting, they are made invisible precisely because they are not listed as clients in the contracts. "So automatically for us it was always implied that we're not supposed to deal with the partner, that there would have to be an addendum attached to the contract. It also actually determines who we deal with in that family" (Kateřina, social worker).
One of the interviewed fathers emphasised the risks associated with tying contracts only to mothers, connected with mothers taking responsibility for the family's situation. "It should be valid for both parents…they should enter into a relationship where both of them are bound. It's usually only the mother who is bound because she receives all the welfare benefits ..." (Alex, father). Another father assumed that if they are married, the contract covers both parents. "If we're married, then one contract is enough, right? There's no need for two contracts, I assume. They gave us one contract and not two, right?" (Hugo, father). This shows an unclear father role in the cooperation with the service.
Non-involvement of fathers in the formal collaboration can be interpreted as a way of coping with professional demands. Social workers reported that they found working with the whole family very challenging and the situation where the whole family was brought together for an intervention was not common ("… but it doesn't feel like that to me…to really be able to work with the whole family. I can work with an individual, but I've almost never experienced working with a family, and actually with children either", Radmila, manager).
Radmila, the manager, sums up: "The current situation in social services, where we're forced to create waiting lists for individual families, the number of families per worker and interventions per worker are high, does not help to involve the whole family," and adds: "I'm thinking that's unrealistic. We would need a lot more time and a lot more capacity because it takes a lot of preparation for such a meeting…since we're all there, I believe we need to prepare for that." Also, the work demands of fathers from poor families and working hours of the service do not facilitate their involvement. This may be linked to a culture of the organisation, which does not sufficiently factor in different daily rhythms and thus a space for work with families (for example, men may do physically demanding work that leaves them tired; they are on "moonlight jobs" away from home -time for contact is limited).
According to the social workers, not only time but also the place of intervention play an important role. Some services only carry out field activities, others also run outpatient services. If the social workers were able to compare field and outpatient service provision, they reported that both forms had their disadvantages. According to the workers' experience, the outpatient form is not sought after by fathers. According to their statements, this is related to a formal and feminised environment of the service. In field work, although workers have a better chance of meeting fathers ("… the father gets to know the social worker in a different environment…that it's not just strictly here in the office, where I deal with him…and that he also knows that he can talk to the social worker..., and he somehow becomes more involved, so I believe that the cooperation works better there", Kateřina, social worker), fathers may perceive this as an invasion of their privacy.
Another way of legitimisation of father non-involvement at the mezzo level was to present the service as a "voluntary" service. The voluntary nature of the social activation service is usually defined in opposition to activities of an office for social and legal protection of children, which is seen as a controlling body. At the same time, the voluntariness of the clients' involvement in social services is considered questionable in view of the power pressure exerted by a child protection authority: "Well yeah, and that's exactly the family where cooperation is voluntary, but recommended by an office for social and legal protection of children…, so the voluntary nature is questionable (laughs)" (Zora, social worker).
Not only gender but also sexuality has a significant influence on a social work conduct. Sexuality is a regulatory mechanism that in some cases leads to avoidance of male clients. Both the desire to avoid insecurity on a personal and cultural level and the discourse of heteronormativity are involved. Professional boundaries are codified in organisations by the dress and behavioural requirements expected of female social workers ("They have to be more careful, of course, as to how they dress, how they smile at men"). Female social workers then lack knowledge of the male client's perspective and based on their own stereotypes of men they may avoid them because they feel they do not understand them, cannot communicate with them, and do not feel comfortable in their presence ("I think some topics are maybe more difficult to open up with a man", Sandra, social worker). Avoidance can also be blamed on situations where social workers do not know how to respond to intimacies and comments from fathers: "He sends me emails saying how he's been looking forward to meeting the blondie and I didn't quite know how to respond to that…so I requested individual supervision with a supervisor for this one" (Týna, social worker).

(Non)involvement of fathers at structural (macro) level
The macro level refers to the network of social and power relations, and to the ways in which oppression is "institutionalised" (fixed through thought, language, and action). The families we talked to often face economic and housing problems, unemployment, and institutionalised racism. These are families who live in insecure and substandard housing and/or are large families that cannot find a larger flat. However, the search for housing is not seen by social workers as a contract to be completed. "Quite often and most they deal with a housing issue… us as social activation services…we shouldn't deal with housing at all, but it's important for us that the family is able to function at all" (Radmila, manager). At the same time, social workers feel that it is not in their power (or the power of their clients) to address the housing issue. "But really to solve other matters if the family doesn't have a space to stay is terribly difficult. And of course, we can't arrange for that, can we? We can help them take small steps, but we can't secure flats for them" (Garbriela, social worker). Given the perception of housing in the Czech Republic as a commodity/investment, not a right, this is primarily a financial issue. The main barriers include discrimination against Roma in housing and labour markets. The inability to secure adequate housing or work can be perceived as a failure of fathers as breadwinners. Similarly, debt/foreclosures can be seen as a failure of the breadwinner-father. Here we can observe how a gendered aspect of shadowing fathers is interconnected with an element of (primarily racial) oppressionthis leads to circumstances that exclude men from contact, whether due to their unavailability (due to imprisonment, constant money-chasing, drug addiction) or due to their feeling of incompetence, which is a combination of expected attributes (providing for the family) and their disempowerment due to racial oppression. This puts social services in a tough spot of needing to respond to these contradictions, in order to do their work, but at the same time not having the mandate and instruments to do so (see also Valová & Janebová, 2015).
Parents' statements point to deepening poverty and segregation as well as the absence of systematic solutions to family situations: "It would be important to find a new flat, right? Because I'm tired of living there….my kids got infected there…it's a bad location" (Joanna, mother). "For example, welfare advice, welfare benefits, housing, or just accompaniment to places. Because on many occasions we go, but they don't tell us anything…to what we're entitled…nothing, they just don't tell us on their own" (Albert, father). "Me? What would I need? Another flat. To be able to leave from here" (Radovan, father).
Maternally oriented services do not contribute to unfavourable situations of families either. If, for example, a family is in a housing need and must solve its situation through emergency housing in a homeless shelter, the question often arises, what to do with the father? Social services in the Czech Republic are highly gendered. The father is forced to look for housing separate from the rest of the family. Shelters are usually divided into shelters for women and mothers with children and shelters for men. Families must thus be separated, and the children must stay with their mother. Shelters for fathers with children are not available. The social service system thus disqualifies fathers as parents. "... so quite often we end up in a situation where we tell these mothers that they have to decide to either opt for a partner or children, meaning that they'll go to a shelter for mothers with children. We're actually happy if she gives priority to her children, because it's the only chance for the family to live somewhere…but it's hard for them and often they don't want to leave their partner" (Radmila, manager). This complicates further work with the whole family. In contracts not to be separated, families often move to a hostel in such cases. A disadvantage for some families who would benefit from more intensive support, however, may be the absence of a social service that is offered almost 24/7 in shelters, whereas social workers can only come to visit such families at hostels.
Although some social workers reported that they did not see any differences between contracts from mothers and fathers, "...I guess, I don't see any difference at all…women don't have any different goals from men…" (Sandra, social worker), father involvement appeared to be dependent on their contracts: "It depends on…if we really only deal with some of the care stuff, most of the time I really just end up working with the mother" (Laura, social worker).
However, the contracts do not come exclusively from families, but often from outside: child protection authorities, schools, courts, etc. Family social service workers also have their own ideas about how families should function. "And I just try to incorporate something into that for them, ʻYou should do this and that. You should go here and there, it's just needed'" (Klára, social worker). Some social workers reported that the construction of a mother as a primary client was related to the model passed on from an office for social and legal protection of children that also works primarily with the mother and passes the case on in this way. "...the office for social and legal protection of children primarily works with the mother and refers some part of it to us to deal with directly along with the mother's contact. We all meet -an office for social and legal protection of children, mother, and us, so they hand over the model where they work with the mother" (Radmila, social worker).

Discussion and conclusion
The results suggest that not only personal attributes, but also the design and form of social service provision may lead to fathers losing a contact with social and activation services for families with children. This happens on the basis of reinforcing traditional gender roles (also McKeown, 2001), but also on the basis of different forms of oppression combined with an ethnicizing approach to Roma in Czech society.
Regarding the first level of traditional gender roles reinforcement, the feminisation of the profession and gendering of services may have a deterrent effect on fathers' willingness to become involved. Given that social service contracts are predominantly signed with mothers, administrative barriers to father involvement are also linked to the problem of gendered services. Involving fathers and working with the whole family is therefore complicated at the service system setup level, at the intra-organisational processes level, and at the personal level of individual workers or clients. Setup of an entire service system that does not include fathers is supported by a "father deficit model" (Lechowitz et al., 2019). Panter-Brick et al. (2014) found that the policy frameworks under which interventions are implemented often assume this deficit, where fathers are considered incompetent in terms of their skills and knowledge related to children's health and development. Worker attitudes and behaviours towards fathers may perpetuate the deficit model and affect the level of father involvement (Pfitzner et al., 2015).
An example of the interconnectedness of individual barriers is, for example, the subjectively perceived fear of workers associated with the participation of fathers, which has its foundations at the macro and mezzo level. At the macro level, it is related to social construction of men in social work as aggressors; at the mezzo level, this construct is supported by organizing self-defence courses for female workers and the supply of protective equipment (e.g., pepper spray). Also, Maxwell et al. (2012) note that social workers sometimes label fathers as dangerous without having met them. Brown et al. (2009) suggest that social workers' fear of men is a fear of the unknown. According to Baum (2016), a feeling of threat has both external and internal causes. Some fears may stem from the higher proportion of men in families with children who have substance abuse problems, and/or a history of criminal activity related to violence. (Zanoni et al., 2013) Female workers may then generalize their experiences or unwritten organisational beliefs and become hostile and distrustful of all male clients (Baum, 2016). The need to explicitly formulate the gender constructions of male and female clients and understand their implications in family social work has been emphasised by Scourfield (2001). He argues that social workers should reflect on their own practice and the gender discourses in the culture of their own organisation. It is social work educators who should support this process of reflection.
As for the second level, it is related to structural and racial oppression, but also to an ethnicizing approach of social workers and the self-concept of their clients. The results show that there is an intersection of gender and race, where social workers are able to distinguish between majority and minority (Roma) masculinity within gender categorisation, with the Roma masculinity being at a disadvantage here (see Connell, 2005). This can lead to 'distancing' among Roma (Doubek et al., 2014) defined as an attempt to shift social downgrading of the "Roma" category to the "other Roma" category, in this case "Roma men". This ethnicizing profiling can also be accepted by female clients -see the statement of one of them, who applies gendered differentiation to all Roma people in line with the opinion of female social workers. It can be concluded that phenomena associated with oppression or disadvantaged socio-economic position can be misinterpreted in terms of an ethnicizing view of gender roles in families.
For educators, there is another impulse aptly expressed by Kozubík et al. (2018, p. 81) as "deconstruction of a stereotypical ethnicizing optics of marginalized Roma communities." They base their conclusions on an ethnographic study of life in segregated rural settlements in Slovakia, which showed only a negligible difference between the local culture and the traditional rural culture of Slovaks. Given the interconnectedness of gendered differentiation and ethnicizing approaches that we observed in our research, we believe that promoting gender-sensitive education can contribute to the deconstruction suggested by Kozubík et al. (2018).
Unless the social, political, and economic context in which families live, is explicitly acknowledged and addressed, these services will continue to focus on individual (personal) rather than structural deficiencies (Philip et al., 2018). Poverty, ethnicity, and culture are just some of the factors that Coakley et al. (2014) suggest may prevent fathers from fully participating in addressing the family situation. Gordon et al. (2012) add that racism and discrimination underlie many fathers' emotional and psychological problems, which may adversely affect their ability to function competently as parents.

Data availability statement
The data that support the findings of this study are available on request from the corresponding author (in Czech and Slovak language). The data are not publicly available due to their containing information that could compromise the privacy of research participants.