Constructing Family from a Social Work Perspective in Child Welfare: A Juggling Act at Best

The transformative reality of diverse Canadian families is outpacing national and provincial statutes and policies. Social workers in child welfare agencies are faced with the complex task of making decisions about families while working within the confines of national/provincial statutes and social policies, as well as within agency structures. They attempt to balance the rights of diverse Canadian families and still protect children at risk of harm with the principle of the ‘best interest of the child’. The purpose of this qualitative case study was to explore the construction of ‘family’ and decisions about family life in protection services from the perspective of professional social workers in the prairie region of Canada. Social workers from several urban communities were invited to participate in focus groups. During the focus group discussions, themes of social worker’s nuanced and somewhat fluid understandings of family did not always converge with current legal and professional notions of families. Study findings suggest that social workers’ construction of family and the decisions they make about family life involve three primary themes: ‘acceptance of diverse understandings of family’; ‘safety and the best interest of the child’, and ‘professional discretionary decisions’ This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.


Introduction
Beginning in the early 1990s, the Global North (developed countries in the northern hemisphere) shifted its policies to reflect a neoliberal framework; ideologically, neoliberalism promotes individualism and lauds the value of choice for both worker and consumer (Dubrowolsky, 2008;Harvey, 2005).Working within a neoliberal framework, Canadian policy makers are faced with legal, social and political challenges in understanding new social and political realities within the 21 st century.Their work includes, but is not limited to, defining what are 'legally significant personal relationships' or stated another way, understanding 'family'.Within a neoliberal framework, social policy and understanding family structures is a complex and evolving endeavour (Brodie, 2010).

Canadian social policy reforms have not yet adequately responded to contemporary
Canadian families and their diverse challenges (Brodie, 2010).This paper describes a small qualitative case study comprised of four focus groups, with a total of 29 Saskatchewan social workers working in child welfare services.This case study is just one study in a collection of studies within an international research project on social work with families (for other examples, see Nygren & Oltedal, 2015;Tembo & Oltedal, 2015).The purpose of this qualitative case study was to explore the construction of 'family' and decision-making process about family life in the prairie region of Canada in professional social work practice.This paper will take the following format: a literature review that focuses specifically on contextual factors such as political, legal and policy realities, which impact social worker decisions about family life, research methods, findings and conclusions.For the purposes of this study, it was theorized that the 'best interest of the child' principle would override all other considerations in professional practice when working with families within the context of child protection.
When constructing knowledge and interpreting social situations with regard to family and family life, social workers must consider the social, cultural, political and linguistic contexts of a given situation.They must also consider the diverse understandings of individuals and societies as reflected in various methods, manuals (i.e.risk assessments, guidelines, legalities, etc.) and theories.They need to consider their own positional impact on a given situation, as well as the impact of using different methods and/or manuals (Nylund & Nylund, 2003).Thus, the decision-making process for social workers in child protection services is complex, difficult, contextually driven and inherently reflexive in nature.In addition, they must include factors such as trying to balance the safety and best interest of the child while respecting and upholding the rights of families from intrusion.They need to understand that child abuse is not a static concept, and is defined across both time and context.Lastly, social worker decisions about family life can result in extreme media scrutiny.Their decisions can be viewed as overzealous or neglectful (i.e.failure to keep children free from harm) (Parada, Barnoff, & Coleman, 2007), thereby adding to the complexity of the situation.However, Herz and Johansson (2011) suggest that the challenge of understanding the social worker's decision-making process about family life can be seen as a window of opportunity to explore diverse understandings of how social workers in child protection services make decisions about family life.
The answer to the question 'What is a family?' has a significant impact not only on various laws but also on social policies (Bala & Bromwich, 2002), and ultimately impacts social worker decisions connected to family life.This paper focuses specifically on a population of social workers who must understand 'family' and make decisions about family life in protection services, while navigating the complex contextual terrain that shapes their decisions for intervening in families' lives.Liberal welfare regimes (i.e. as in Canada) that reflect free market values, and individual liberty and responsibility for choices made, shape social work practice decisions (Rush & Keenan, 2013).Other contextual factors that also shape practice decisions include, but are not limited to, personal values, social work ideologies, agency mandates, national and provincial laws, social policies, service provision manuals and guidelines (Healy, 2005).
Many countries, including Canada, are attempting to move away from the historical ideal construct of family: what was described as 'permanent, monogamous, married, nuclear, heterosexual, Christian, and with defined gender roles' (Bala & Bromwich, 2002, p. 118).
Professionals, such as social workers, seek to understand significant personal relationships based on the current and changing realities of how Canadian families are formed in the 21 st century.Understanding diverse family forms and responding to multicomplex identities within the field of child welfare is vital for best practice approaches in engaging families (Morris, 2012).Although understandings of the family are diverse (Hakim, 2003), the 'idealized' nuclear form of understanding family is still recognized in the Global North, and continues to be used as a yardstick in evaluating relationships and child-rearing in current child welfare practice (Hargreaves & Page, 2013).
The Convention on the Rights of the Child, Article 3.1, states: 'In all actions concerning children, whether undertaken by public or private social welfare institutions, courts of law, administrative authorities or legislative bodies, the best interest of the child shall be a primary consideration.'Thus, the global concept, 'in the best interest of the child', is important not only from a legal perspective but also from a child welfare perspective, as it paves a path toward shaping a 'family form' that is ideally reflected in many countries (Bird, 2010).Gumuscu, Khoo and Nygren (2002) suggest that the construct 'best interest of the child' propels child welfare laws and policies, even though this construct can have varied understandings from location to location and from social worker to social worker.
There are a few international studies that link to this study's exploration of child protection workers' decision-making processes that illuminates their construction of family (Gavriel-Fried, Shilo, & Cohen, 2014;Khoo, Hyvonen, & Nygren, 2003;Khoo, Hyvonen, & Nygren, 2002;Nygren & Oltedal, 2015;Tembo & Oltedal, 2015).For example, Khoo et al.'s (2002) study examined social workers' intervention in decision-making processes in child protection services in both Canada and Sweden.They found that in Canada, unlike in Sweden, the more intrusive principles of protection and best interest dominated intervention decisions in child protection cases.The 'best interest of the child' principle was addressed through protection (supervising parent behaviour) and permanency planning (Khoo et al., 2002).Tembo and Oltedal (2015) indicated that factors such as patriarchal structures, culture and the economic status (i.e.lack of resources) impact Malawi social worker's decision-making processes when placing children out of home.
The 'best interest of the child' principle was premised within the belief that children's interests are best served by remaining in the home with the provision of needed financial resources.And lastly, Gavriel-Fried, Shilo and Cohen's (2014) study found that Israeli social workers defined family as simply 'any relationship or living arrangement involving children'.Factors such as their personal values, socio-demographic variables and personal acquaintance with non-traditional types of families influenced Israeli social workers' decision-making process in defining 'family'.

Understanding family
Historically in Canada, the construct of family was understood to be what is described as the traditional 'nuclear' family that included a male adult (father), female adult (mother) and their genetic offspring (children) (Bala & Bronwich, 2002;Bird, 2010;Miall & March, 2003;Bures, 2009); the extended family included family members such as grandparents, aunts and uncles and cousins (Marcionis, 2010).Under Canadian law, the definition of the nuclear 'family' was formed via the Christian concept of marriage (Bala & Bronwich, 2002).Patriarchal, heterosexual and Judeo-Christian values, and in turn, customs, laws and social roles evolving from those values were intrinsic within this understanding of family (Bures, 2009;Bala & Bronwich, 2002;Wu, Hu, & Schimmele, 2008;Bird, 2010;Miall & March, 2003;Gazso, 2009).This 'nuclear' understanding of 'family' has paved the path in the development and implementation of Canadian social policies and laws that focus on the 'nuclear family' and its welfare, hence providing practical guidance for social workers in child protection services (Bala & Bronwich 2002;Bird, 2010).In addition to Canadian social policies and laws, social work associations (provincial, national and international) also provide guidance to Canadian social workers in their understanding of 'family'.For example, the International Federation of Social Workers (IFSW) website states that in the 'best interests of the child', biological parents are best if they can provide emotional and physical care (ifsw.org/statements/the-best-interest-of-the-child/),while the Canadian Association of Social Work (CASW) and the Saskatchewan Association of Social Work (SASW) suggest that a family consists of a parent-child relationship which may or may not be biological in form.

Family formation and composition
Most Canadian families are generally understood to be formed from two general perspectives: the biological perspective (blood ties, usually thought of as the nuclear family and an extended family) (Macionis, 2010) and the psychological perspective (the quality of individual relationships and support, and how these relationships are maintained) (Schoenhalls & Behar, 2000).Families are also understood from diverse cultural perspectives that can incorporate either perspectives or aspects from both perspectives.Cultural perspectives of family can include entities such as ethnicity, religion, sexual orientation or social experiences (He, 2005;McGoldrick, 1992;McGoldrick, Giordiano, & Garcia-Preto, 2005;Este, 2007).Some ethnic perspectives that include a collective understanding of family may also include multiple generations and extended family members (He, 2005;McGoldrick, 1992;McGoldrick et al., 2005) or their entire community (Simich, Beiser, Stewart, & Mwakarimba, 2005).
For example, Indigenous peoples in Canada, extended family members, clans and communities can be described as 'family' although there may be some differentiation in family roles between various Indigenous communities (Gupta, 2000;Hick, 2006 a).For many Indigenous peoples, a type of kinship model of family is still prevalent.For instance, Dragonfly (2012) states: Today, clans and societies still have a role in traditional governance activities.They also confer kinship.A member of your clan, even if not a biological relative, is considered a relation.The clan system ensures interconnectedness and balance among generations and even between distant nations, as people from other nations are also considered relatives if they are from the same clan (p.1).
However, in Canada, many First Nations communities are often divided and fragmented among themselves, in many cases often-referred to as 'colonial structures', as the 'nuclear family' has replaced traditional Indigenous structures such as extended family and clan systems (Dragonfly, 2012).
According to Bird (2010), the two-parent heterosexual biological model or the nuclear family unit is the current model of choice for provincial governments in Canada.There is limited provincial government acknowledgement via legislation and social policies that families can be formed from psychological relationships, which may support the formation of multi-parent families.A number of researchers advocate that the biological model of the family is the 'best base for a child's development' (Bala & Bronwich, 2002;Erfani & Beaujot, 2009;Eshleman & Wilson, 2001;He, 2005;Miall & March, 2003).Even so, Bird (2010) does question whether or not this advocacy for the 'nuclear' model of the family is exclusionary, and thus discriminatory to other models of family that can also be formed in 'the best interest of the child'.

Marriage
Statistics Canada (2012a) recognizes that even though married couples still comprise the majority of families (67%), the legal concept of marriage as a defining characteristic of the 'nuclear' family is no longer realistic in Canadian society due to an increase in commonlaw relationships, separation and divorce (Bala & Bronwich, 2002;Cameron, Coady, & Hoy, 2012), single parent families (Bird, 2010), cohabitation and lesbian and gay partnerships (up 42.4 % to 64, 575 since the 2006 census) (Goldberg, 2007;Rose, 2012).
Only 25% of Canadian families represent the 'traditional' biological nuclear family comprised of mother, father and children in the home (Statistics Canada, 2012b).
Common-law couples (up 13.9 % since the 2006 census) have now surpassed the number of single parents in Canada (up 8 % since the 2006 census).The 2011 census also found that one in 10 children live in a stepfamily.Of the approximately 3.7 million families with children, 87.4% of these families were comprised of two parents and their biological or adopted children, while 12.6% were stepfamilies.Other considerations in understanding the construct of 'family' include the caring of children that does not include blood relations (Miall & March, 2003), which includes sperm donors.For example, a family comprised of one father and two mothers challenges the biological model of the family that consists of only 'two' parents (Bala & Bronwich, 2002).
Statistics Canada (2006) introduced a change in the definition of the family to realistically reflect the changes seen in most families across Canada.A 'census family' acknowledges couples and/or lone parents with children, whereas an 'economic family' focuses on relations that go beyond parents and their children.Statistics Canada's (2006Canada's ( , 2012b) ) definition of the family appears to be inclusive for most mainstream Canadian families (Tillman & Nam, 2008).Nonetheless, the current census understanding of family does not adequately reflect a 'kinship' model of family that many Canadian Indigenous families support.Surprisingly, it does not appear that Statistics Canada's 'economic' and census definitions of 'family' are utilized in other Canadian institutions, specifically, the child welfare system.Important Canadian federal social policies such as social assistance, employment insurance and parental leave focus on the traditional heterosexual twoparent 'nuclear' biological model in understanding and defining the 'family' (Gazo, 2009;Service Canada, 2011).

Child welfare practice and family
The Constitution Act of Canada stipulates that all child welfare responsibilities are a provincial/territorial-, and not federal responsibility (Constitution Act, 1982); thus, each of the three territories and 10 provinces have jurisdiction over the delivery of child welfare services.This means that 36 million Canadians have 13 different Child Welfare Acts.
Canadian social workers who work in child protection services are legally sanctioned to intervene in families when children's safety and care are at stake (De Boer & Cody, 2007), as child safety is first and foremost in Canadian child welfare services (Black, Trocme, Fallon, & MacLaurin, 2008).Child welfare workers advocate for the rights of the child and have adopted their own understanding of what is and is not acceptable (i.e.professional discretion), based on the rights of the child and human rights in general (Durrant, Trocme, Fallon, Milne, & Black, 2009).
In its Guidelines for Ethical Practice (2005), The Canadian Social Work Association states that social workers exercise professional judgment (professional discretion) in their decision-making process, which is consistent with their provincial/territorial legislation if vulnerable members of society, such as children and their well-being, are at risk The term 'professional agency', rather than professional judgement or discretion, is also used in the literature.It refers to the capacity that social workers have to 'exercise their social work knowledge, skills and clinical judgement when making decisions in the context of their everyday child protection practices' (Parada et al., 2007, p.36).
Canadian legislation calls for a least intrusive model of social work practice.This model is based on the belief that it is generally in the child's best interest to be raised by its family of origin (Magnuson, Patton, & Looysen, 2011).Morris (2012) suggests that much of the relevant policy and practice literature directed at 'families' is in reality focused on children and parents, and/or with vulnerable adults, with minimal reference to extended family networks (p.908).
There is a legal expectation that parents do not harm children in any way so that their development and functioning is not negatively affected (Black et al., 2008;Chamberland, Fallon, Black, & Trocme, 2011), and that parents are responsible for the safe care of their child.There is also an understanding that parent-child relationships are a fundamental piece in the concept of the family with an element of responsible care (Cameron et al., 2012b;Durrant et al., 2009;Lavergne et al., 2010).Child welfare is aimed at the expectation that the mother needs to provide care and safety to her children.Fathers are often not included in child welfare, or excluded due to biases, policies or fear, as a biological relationship between family members is generally assumed (Brown, Callahan, Strega, Walmsley, & Dominelli, 2009;Cameron, Coady, & Hoy, 2012a).
Conversely, new relationships other than the biological parent-child relationship must be negotiated within families, and between families and social work professionals.These relationships create conflicting views about the worth and safety of these relationships for family members and professionals alike (Saltiel, 2013).In addition, according to Gladstone, Brown and Fitzgerald (2009), social work practice with kin does not appear to be collaborative, and thus contradicts the 'best interest of the family' approach which child protection policies assume.Lastly, policy or permanency planning in child welfare sometimes appears to override the best interest of the family (Gladstone et al., 2009).Bernard (2007) suggests that there is a lack of systemic research in best practice approaches that will inform social workers in child protection services, not only on how to reduce risk but also on how to balance the promotion of the 'best interest of the child' principle, while recognizing the rights and needs of the family as a whole.The decision to remove a child from a home in the 'best interest of the child' can be based on a perceived lack of parental competence, other than deficiencies in professional practice, services or support (Booth & Booth, 1994;Tarleton, Ward, & Howard, 2006).Not unique in Canadian child welfare practices, Saskatchewan child protection practice utilizes an interventionist approach to child welfare, rather than a 'preventative approach' (Faris-Manning & Zandstra, 2003).From an interventionist perspective, children's safety and nurturance needs ( i.e. rights) can be viewed as unique and separate from their parents and families, thereby potentially weakening the link between child and parent, and reinforcing a stronger link between child and state (McConnell, 2009).

Saskatchewan child welfare
In Saskatchewan, the Ministry of Social Services is responsible for providing protection services for children under the age of 16 (and in exceptional cases to youth under 18 years of age) under the mandate of the Child and Family Services Act (1989-90).
However, First Nations Child and Family Service (FNCFS) Agencies (currently 17 agencies in Saskatchewan) have delegated authority to provide child protection services to children and families on-reserve, including some areas off-reserve (MacLaurin et al., 2011).
The Child and Family Services Act (1889-90) forms the legal framework for the policy manual for Saskatchewan social workers, the Policy and Procedures Manual (2015).This manual emphasizes community-based prevention services (Final Report, 2002), and is framed within a number of theoretical approaches, such as a strengths perspective, and a focus on a client-centred practice intended to guide social worker decisions about family life (Kufeldt, Vachon, Simard, Baker, & Andrews, 2000).The structured decision-making (SDM) risk assessment tool is a key aspect of this manual intended to bring consistency and efficiency to the decision-making processes and practices of social workers (Parada et al., 2007).Among others, the objectives of the SDM risk assessment tool are to identify critical decision points and increase the reliability and validity of decisions.Social worker discretionary decisions are likely possible within the SDM tool, as workers can indicate whether or not they wish to increase or decrease 'response levels'.Nevertheless, Cradock (2004) suggests that risk assessment tools as a whole do not have the capacity to address the complexities of family lives because these tools appear to reduce families to mere scientific numbers.

Who defines family?
According to Barn (2007), it is important for social workers to understand how families understand the concept of 'family' before they intervene.Morgan (2011) suggests that rather than viewing family as a defined institution, social workers should ask instead, how 'do people do people?' or perhaps social workers should aim to understand families in terms of processes of social interaction that shape and re-shape family relationships on a daily basis (Saltiel, 2013).Social workers are aware that broad and diverse understandings of family often have little resemblance to the nuclear model of the family and/or to their understanding of family, including both their personal and professional understandings.However, the legislation and policies that guide their practice is reflected in the nuclear model of the family in its language and practice (Morris, 2012).Child welfare/child protection workers must continuously defend their decisions about family life before their clients, professional colleagues and the officials of family court (Magnuson et al., 2011).

Research question
The purpose of this qualitative case study was to explore the construction of 'family' and decisions about family life in protection services from the perspective of professional social workers in the prairie region of Canada.The primary research question was: How do social workers understand 'family'?The secondary question was: What factors influence social workers in the decision-making process in making decisions about family life?Research ethics approval for this study was received from the University of Regina, Research Ethics Board, as well as from the Research Ethics Committee, Government of Saskatchewan.

Design
This study utilized a qualitative exploratory case study design; case study designs are philosophically based within a constructivist paradigm.The essence of a case study is that 'it tries to illuminate a decision or set of decisions: why they were taken, how they were implemented, and with what result' (Yin, 2009, p. 17).In addition, a case study design helps to provide insight into an issue or phenomenon; the case provides the mechanism into illuminating the understanding of the issue or phenomenon (Luck, Jackson, & Usher, 2007).Baxter and Jack (2008) suggest that utilizing a case study design ensures that the phenomenon is explored through diverse lenses, rather than one lens, which '…allows for multiple facets of the phenomenon to be revealed and understood' (p.544).One advantage of a case study design is the collaboration between the researcher and the participant; hence, participants are able to tell their stories (Crabtree & Miller, 1999).
Given that this study wished to explore, 'How do social workers construct family in professional practice in protection services?';an exploratory case study seems the optimal choice as a design for this research in order to explore those factors, in which the decision-making process about family life has no clear, single set of outcomes (Yin, 2003).Thus, the 'case' in this qualitative case study was the decision-making process in how the social workers constructed 'family' and made decisions about family life within the context of child welfare in Saskatchewan.

Constructivist paradigm
The philosophical foundation for a case study design lies with a constructivist paradigm (Yin, 2003;Stake, 1995).A survey of the literature indicates that 'social constructionism' has been used synonymously with 'social constructivism' (Franklin, 1995).Social constructivism is based on four main beliefs: realities are socially constructed; realities are constituted through language; knowledge is sustained by social processes and reflexivity in human beings is emphasized.In a social constructivist paradigm, an objective reality does not exist.Social constructivists believe that reality is produced via social negotiation and discourse; as separate entities, individuals, do not create reality (Gergen, 1994).It is through the process of social discourse that meanings are co-constructed (Gergen, 1994).Language is the means to construct worldviews and realities, as knowledge understood from a positivist perspective is no longer valid.For example, social workers adopting a social constructivist position maintain that their client understanding(s) of the world and self is an on-going process of communication in which clients and workers co-construct meanings.Lastly, social constructivists emphasize an ongoing and active reflection that questions different forms of knowledge and the inherent power of differentials with diverse forms of understanding (Nightingale & Cromby, 1999).
The meaning-making process is culturally and historically specific.For instance, social workers from a social constructivist position must constantly reflect on their personal and professional understandings, biases and assumptions about families and family life, as they co-construct meaning with clients.In summary, linear notions of reality do not exist, but rather reality and how we come to know reality is an act of co-construction, with an emphasis on the complexity and interrelatedness of the many aspects of individuals within their communities (Denizen & Lincoln, 2000).Therefore, through time and interaction with one another, social workers and clients create realities.

Sample
Social workers who were employed in either a mid-sized city (population 232,000) or a small city (population 32,200) were invited to participate in focus groups to explore their understandings of 'family' within the context of child welfare practice.There were four focus groups conducted between June and October, 2012.Participation was voluntary; the sampling procedure was purposeful and did not entail random selection.There were nine participants in the group for the not-for-profit agency and in the two government agencies (mid-size city); there were six and eight participants, respectively, while for the government agency in the small city there were six participants.
There were 29 participants in total: four male social workers and 25 female social workers.

Data collection
The decision-making process of understanding 'family' and making decisions about family life for social workers working in child protection was documented through qualitative research.Using focus groups as a data collection method was accompanied by a semistructured interview guide and three fictitious child protection case study vignettes.In the healthcare field in areas such as nursing (Web & Kevern, 2001), social work (Linhorst, 2002) and epidemiology (Dahlgren, Emmelin, & Winkvst, 2004), focus groups have proven to be an effective data collection method, particularly in exploratory studies.These focus groups were face-to-face, lasted 40 to 60 minutes, and were audio recorded and transcribed verbatim.Focus groups took place in the boardrooms located within each agency office: three government offices and one not-for-profit agency.In addition to the semi-structured interview guide, three fictitious family case study vignettes were shared (both written and orally) with the focus group participants.Participants were asked to respond to the three vignettes by stating what they would do if they were the child protection worker assigned to this case.
Vignettes are used to generate data that might not be discovered if only one data collection method such as interviews or observation is used (Renold, 2002).Vignettes are a useful way of making concrete the events and experience of practice, and facilitating the identification of individuals' situated understanding and practical theory (Reynold, 2002).They are equally valuable as a means of capturing assessment practice so that it can be reflected on for evidence of [the participants] enacted theories of assessment (Phillips, Schostak, & Tyler, 2000, p. 130).
These vignettes, which were prepared by an international research team in the social work with families' research project (Nygren, & Oltedal, 2014), were shared with participants as a method to stimulate and encourage a dialogue of their practice within protection services.Utilizing vignettes as a data collection method also allows the researcher the opportunity to compare different groups' (i.e.social workers employed in different locations or within different work units) responses (Renold, 2002).Criticism of the use of case study vignettes as a research method is concerned about the reality of 'what people believe'.What they ought to do with what they actually do may be quite different (Finch, 1987).However, the use of multi-methods, one of which is the case study vignette, can enhance our understandings of the relationships between beliefs and actions (Finch, 1987).

Data analysis
For the purpose of this study, a thematic analysis (a process for identifying patterns or themes within qualitative data) was used to analyse the focus group data.Due to its flexibility, this method of analysing data is not tied to any particular epistemological or theoretical perspective (Smith, 2008).The focus group transcripts generated approximately 60 pages of text.Data was coded by following Braun and Clarke's (2006) framework for conducting thematic analysis, and was reduced into meaningful segments in order to identify themes and sub-themes.Because the research team was interested in addressing a specific research question: 'How do social workers construct family?' the initial coding frame was based on theoretical links and issues that were identified using a subset of the transcripts.The coding frame was then applied to the remainder of the transcripts and modified accordingly.Both a social work graduate student (who was not present during the focus groups), and a senior research team member separately coded the transcripts and identified themes/sub-themes that were then compared for consistency.The key themes were 'acceptance of diverse understandings of family', 'the safety and best interest of the child' and 'professional discretionary decisions'.A brief description of the case study vignettes and study findings/themes derived from the focus group data are presented next.

Three case study vignettes -brief descriptions
The first vignette case study focused on a 14--year-old girl, Maria who was six months pregnant and still in school, and who had not advised her family at that time that she was pregnant.There was some indication that she had been exposed to physical abuse from her father towards her mother, and was afraid to tell her parents about the pregnancy.
The father of her unborn child lived in her neighbourhood, but they were no longer in a relationship.The second vignette case study focussed on Maria, who was now 16 years old.Maria had dropped out of school and was having considerable difficulty in raising her daughter Penny, who was now two years old, and agreed that Penny be placed in foster care.Due to the increasing violence between her parents, Family Services felt that Maria should be moved to supportive housing in the community.The final vignette focussed on 18-year-old Maria who now had a stable job, and wanted Penny, who was now four years old and living in a foster home, to live with her.Penny had been moved to several foster homes during her care, but Maria had been visiting her every other weekend for the last two years.A psychologist has stated that Penny is hitting other children, and Maria felt that she could provide Penny with the stability and love that she needs.

Research Findings
The focus group participants were first asked to reflect on and share thoughts about their understanding of 'family' prior to reading each case study vignette.Participants shared their own 'personal' understandings of family and shared their professional understanding of family within the framework of their particular agency.If these understandings diverged in some way from one another, they elaborated their perspectives.After the discussion of 'family' concluded, the social workers were asked to read the first vignette, and reflect and discuss their reactions.They then described what course of action they might take.
After this discussion, the second and third vignettes were presented in the same manner.
Workers were asked to reflect and discuss the case study vignettes.After reading each vignette, participants were asked to discuss what advice they might give and what course of action they would likely pursue.
Interestingly, all of the social workers in the four focus groups stated that there was not enough information for them to be able to gain a full understanding of the situation.They stated that their comments were based only on what was presented, but which may have been altered if they had been given more information.For example, from the first vignette, social workers wanted to know if the father of the child had been informed of the pregnancy.They reflected on how Penny's father could have an important role in this case (i.e.potential support, an ensured legal right as the biological parent, the opportunity to have a voice; other potential support from his family, etc.).In addition, they wanted to know more about the school and other community support for expectant youth and/or young students who are mothers, and they wanted to know the current health status of Maria and the baby.

Acceptance of diverse understandings of family
The social welfare policy and practice literature suggests that the construct of 'family' is often used indiscriminately and with different assumptions about its meaning (Morris, 2012).In this study, social workers indicated that often their idea of 'family' did not coincide with their client's notion of family.As participant (2) stated, 'Things are constantly changing.It is really difficult to have one rigid definition of what a family is.' Families were often understood within the context of one's own social location, often referring to the heterosexual 'nuclear' family.Even though social workers realized that notions of family, a term that is dynamic, was not an absolute and was culturally diversified.Participant (2) continued, 'We are always learning more about families'.
The social workers stated that for some families, their 'family' included a child's family network (significant others, both related and unrelated), and was not limited to parents or other immediate caregivers (Euteneuer, & Uhlendorff, 2014;Gumuscu et al., 2014;Morris, 2012).Euteneuer and Uhlendorff (2014) suggest that in the 21 st century, issues such as increased migration, diversity in living arrangements as families, labour market uncertainty and changing relations within the family change our understanding of family.
Individuals and families must develop their own models of family in order to manage family life and negotiate, and adapt or alter these models.The idea of family is constantly In a study that examined how social workers understand the complexity in family's lives, Saltiel (2013) indicated that family networks and the various roles held by individuals in these family systems had to be negotiated within families and between families and professionals.Often, there were conflicting views about the value and safety of these relationships.As Williams (2004) suggests, the construct of family is not so much a focus on who constitutes the family, but rather an exploration of how individuals live as a family.physical needs, spiritual needs, you name it… when we are coming into court we are looking at what is going to be the effect of disrupting the continuity of care and also what the effect of a delay in making a decision will be.
Social workers were constantly renegotiating with families about safety and support.
Options were presented, discussed and then renegotiated, and new realities coconstructed; as a social worker (20) stated: ...at the same time, saying, we very much want to work with you and these are the supports that we can offer, is this something you would be interested in or is there some other support network that you would be interested in that you are aware of that I am not aware of that we can try to engage.

Professional discretionary decisions
The notion that family is a support system was reflected in a number of the participant's statements.For example, one participant (18) stated that: Well I see family as a support system that goes well beyond the confines of family of origin.Her statement was echoed in a number of the participants, such as stated by this social worker ( 16): 'When we're looking for family and when we are involved in families, we look to who they feel is their family when we're looking for supports for them.' Cameron et al.'s (2012) research examined the impact on accessible service delivery sites in child welfare work.They found that although family is understood within the context of parent-child relationships, a community approach to practice is also important.
It was reflected in the sense that the primary caregivers were not responsible for a child's entire care and nurturance, but that the community was also engaged and involved in the care of children (Cameron et al., 2012).Creative and diverse approaches to helping and supporting families are significant in improving client engagement.Supporting families and locating positive support systems includes an element of responsible care that needs to be provided to children.Out of home care, such as kinship care, community care and foster care, may provide temporary care for children when the biological parents are not in a position to provide responsible care (Lavergne et al., 2010).We found that the social workers employed in the not-for-profit agency were not required to follow government policy manuals, nor adhere to court dates and limited time frames to complete assessments, and whose clients were voluntary as opposed to mandated clients.This group of social workers had much more freedom (i.e. less structural constraint) to exercise their professional discretion in making decisions about family life.
Decisions were holistically framed within the knowledge construction between social worker and family.However, all social workers were focussed on client-centred approaches with a strength-based perspective.Given that most, if not all, of these social workers received their university degree from the same university, their focus on clientcentred/strength-based support was not surprising.The strengths perspective is compatible with culturally responsive practice, and the client-worker relationship is reciprocal in this perspective.The principle, the best interest of the child, does restrict the degree of reciprocity within the client-worker relationship.Social workers in this study strove to exercise as much client-worker reciprocity in their relationships with families as was possible, as evidenced in their vocalized right to be self-determining.

Strengths and Limitations
Criticism of the use of case study vignettes as a research method is concerned about the reality of 'what people believe', and thus what they ought to do with what they actually do may be quite different (Finch, 1987).Given that this case study utilized vignettes as a tool to support data gathering in focus group interviews, the question of validity (i.e. would participant suggested actions encompass what actually would take place in real-life situations) could be raised (Nygren, & Oltedal, 2015).As Nygren and Oltedal (2015) indicate, the development of the vignettes used in this research was a collective endeavour from multiple social workers in diverse countries.It is assumed that they represent the context of child protection services around the globe (Hughes & Huby, 2004).The use of multi-methods, one of which is the case study vignette, can enhance our understandings of the relationships between beliefs and actions (Finch, 1987).Case study research has a poorly defined data analysis process (Yin, 2003), but on the other hand, it can follow different analysis methods (Merriam, 1998).
Lastly, a limitation of the research lies within our chosen sampling method.We utilized a non-probability sampling method: purposive sampling.Non-probability sampling methods will not yield a representative sample, nor can our study findings be generalized to all agencies that provide protection and related services (Engel & Schutt, 2017).Purposive sampling focuses on specific guidelines for participant recruitment such as the participant's knowledge of the research question, the participant's willingness to share their understandings and experience, and being representative of a range of views.As researchers, we attempted to focus on these guidelines to ensure that our sample represented the issue (construction of 'family') under examination (Rubin & Rubin, 1995).
Even though our focus group participants were purposively selected, focus group interviews do generate focussed discussions that 'mimic the natural process of forming and expressing opinions -and may give some sense of validity' (Engel & Schutt, 2017, p. 281).
There are a number of strengths to this study, which adds to the limited research in the area of understanding 'family' in the context of protection services worldwide.There are limited analyses that examine how evolving constructs of the 'family' have impacted child welfare workers' interventions with families (Holdsworth, 2004;see Khoo et al., 2003).
Utilizing qualitative methods allows the participants to respond to multifaceted issues and contextual issues (as presented within the vignettes), thus capturing nuances or the texture of responses (Nygren, & Oltedal, 2015;Schultz, & Avital, 2011).
In addition, the triangulation of data sources, data types and/or researchers supports the principle in case study research that the phenomena (decision-making process of constructing 'family') can be viewed and explored from multiple perspectives.An attribute of case study research is the use of multiple data sources such as interviews, participant observation, direct observation and documentation (Yin, 2003), which is a strategy that enhances data credibility.Triangulation can also include multiple theories or having more than one person interpret data and/or collect data.In this study, multiple methods such as semi-structured interviews, case study vignettes and a constructivist paradigm were utilized to enhance data quality through various sources.Triangulation speaks to how the researcher reduces bias and examines the integrity of participant response (Anney, 2014).The research team also analysed the data individually and as a group, utilizing another form of triangulation.
The researchers were also debriefed with one another with regard to study findings adding to the credibility of our study (Merriam, 2002).Lincoln and Guba (1986) suggest that the dependability and conformability of the research can be ascertained through audit trails.In this research, we utilized journal notes that can account for our reflective thoughts and questions, as well as research decisions and activities to show how the data was collected and analysed (Li, 2004).

Concluding Thoughts
The research suggests that effective child protection workers make use of collaborative decision-making processes, sometimes referred to as 'working in partnership' with clients (Spratt & Callen, 2004).They help clients to identify personal, social and environmental issues of concern to them, in addition to helping their clients develop goals and strategies to address these issues.In making their decisions, social workers attempt to work with the clients' understanding of family rather than their own as they balance numerous concerns regarding the safety and well-being of children.In general, social workers take a holistic and systematic approach in working with families and a focus on the issues that have led to the abuse or neglect, rather than the abuse itself (Trotter, 2002).Clearly, social workers in this qualitative case study indicated that they attempt to apply a clientcentred approach building on a strengths-based perspective (Brun & Rapp, 2001;Green, McAllister, & Tarte, 2004).It is an empowering strategy that is not easily applied within the state-sanctioned 'best interest of the child' principle (Burford & Hudson, 2009).
Balancing flexibility with a consistency of practice is a major challenge facing the child welfare field.Indeed, 'Much of good child welfare work with children in care is bridging the gap between the two perspectives, bending the rigidities of law and regulation to accommodate, even nurture and celebrate, the variability of human beings' (Martin, 2000, p.8).The conflicting tensions between parental needs, child protection concerns and the legislation create a 'juggling act'" for professional social workers that sometimes leans in favour of one side over another (Sawyer & Dalzell, 2011).Decision-making in child welfare services is difficult, challenging, and fraught with uncertainty, given that less than desirable solutions can be demoralizing for social workers (Sawyer & Dalzell, 2011).
There is much complexity inherent in sharing power with regard to the need to determine the 'weight' given to various possible decisions in child protection services.At the end of the day, social workers must be able to justify their decisions to families and courts (Sawyer & Dalzell, 2011, p.101).
How do social workers construct family in professional practice within child welfare services?The findings from this qualitative study suggest that three themes permeate social worker decisions about family life: acceptance of diverse understandings of 'family', the principle of the 'best interest of the child' and 'professional discretionary decisions', all of which must balance legislation and policy requirements to contribute to supporting social workers' construction of family and decisions about family life in protection services.
As Khoo et al. (2002) state, 'We are only beginning to grasp the significance of how different structural approaches (such as policy, judiciary and administration) shape the interventions of front-line social workers' (p.453).In the end, the social workers are still creating a juggling act.
Caucasian ethnicity predominated, with 25 of Caucasian ethnicity, two of Metis ethnicity, one of Indigenous ethnicity and one of African ethnicity.'"Métis" means a person who self-identifies as Métis, is distinct from other Aboriginal peoples, is of historic Métis Nation Ancestry and who is accepted by the Métis Nation' (Metis Nation, 2002).Participants' ages ranged from 21 years to 50 years plus (ages 21-30 (n=9) participants); (ages 31-40 (n=9) participants); (41-50 (n=7) participants) and ages (50+ (n=4) participants).Social work educational qualifications consisted of 19 participants with a Bachelor of Social Work degree, six participants with both a Bachelor of Arts and Bachelor of Social Work degree and three participants with other post-secondary degrees/certificates. Social workers were either employed with the provincial government (Family Services) or in a not-forprofit agency that supported women and children fleeing from domestic abuse in urban settings.
changing and is always in flux.Many of the social workers shared their personal understanding of what 'family' meant to them, and then they discussed how their understanding of family may or may not conflict with their client's understanding of family.Social worker constructions of family were often different from that of the families they support.The co-construction of understanding family was an ongoing communicative process that they engaged in with their client families.Participant (7) stated: I know I am just thinking back to when I first started with this work.It really challenged my idea of what family was and how you talk to family and how they define family.For me it was a little bit of a challenge to kind of get my head around what does a family look like and how do we work with family.When you first walk in here [agency], it is quite a shocker.When I think about work and most of our clients are aboriginal families, and they talk about family, they are talking about aunts and cousins, and for them the family seems much broader than when I think of family.
Social workers are required to use specific risk assessment tools such as the Structured Decision-Making Assessment (SDM) tool to support their discretionary decisions.As one participant (26) stated: We just brought in a bunch of new assessment tools… it is very client-focussed and client-centred.Working with the client and assessing what their needs are at this given point and going from this point, and constantly re-assessing where they're at and what their needs are and how they have changed.The SDM assessment tool uses indicators of risk such as alcohol abuse, social support, violence and drug abuse.Participant (26) continued by saying: We can look and see if there are no safety threats [using the tool].It is a tool to assess, but it is more of a measurable tool than what we had before.Social workers believed that the underlying theoretical orientation of the SDM tool is strength-based and solution-focussed.