Social Work/Maatskaplike Werk A professional journal for the social worker

Decisions made by social workers play a crucial role in foster care. Despite the challenging conditions under which social workers work in South Africa, they continue to make decisions to place children in foster care. There is a lack of research on the decision-making processes of social workers rendering foster care services. A need to understand what decisions are made and how these decisions are made prompted this study. A qualitative approach was applied using the multiple case study design, with explorative, descriptive and contextual designs. Most social workers make decisions using a mix of intuition and empirical evidence, justifying the child's best interest as the basis for their decisions. This study contributes to the knowledge of the decision-making processes of social workers rendering foster care services. This knowledge can enable social workers to be consciously aware of their own decision-making processes during all the phases of rendering foster care services, facilitating transparent decision-making, with the best outcome for the foster child and his/her biological and foster family.


INTRODUCTION
Child protection work is highly emotive as social workers investigate and work with families who are possibly abusive in an atmosphere filled with strong feelings, anger, anxiety, concern and fear (Beckett, 2006). Even in these challenging circumstances, social workers routinely make highly complex and highly charged decisions (Kirkman & Melrose, 2014). In the quest to enhance social work practice, there is an increasing need to improve the nature and transparency of decision-making in social work. Social work is not only about making the right decision, but about making a justifiable one (Pollack, 2008).
Traditionally the social work profession was based on volunteerism and altruism, helping the poor with financial and emotional support. The decisions were mostly based on poverty relief founded on common sense, discretion, intuition and experiential knowledge. Without scientific tools to guide decisions, social workers had to rely on intuition (Taylor, 2007). In the late 19th century, the social work profession began to evolve in two ways. Firstly, social work cases were becoming more complex; secondly, there was increasing interest in making the practice more scientific. The move to more objective decision-making methods enabled social workers to introduce greater assurance in their decisions and thus minimise the errors made with intuitive decisions (Mattison, 2000;Wulczyn, Daro, Fluke, Feldman, Glodek & Lifanda, 2020). The significance of the move towards increased scientific rigour in social work processes is evident in the complex decisions that form part of the social work role. This move influenced all areas of social work practice, particularly foster care.
Foster care is a temporary placement for a child, while reconstruction and/or rehabilitation services are rendered to the natural family for reunification with the child or children, or there is planning for longer-term care such as adoption (Pardeck, 1982). The role of the social worker in foster care is to decide if the removal of the child or children from their biological home can be avoided by supporting and strengthening the family and child, or if a threat to a child or children is so serious that it is in the best interest of the child to be removed and placed in safe temporary care until the statutory intervention has been implemented (DSD, 2009). The social worker must then render reconstruction services while the child is in alternate care to treat and rehabilitate the child and biological family for the constructive reintegration of the child into the family (DSD, 2009). The construct of foster care services is described in the Guidelines for the effective management of foster care in South Africa (DSD, 2009). It guides the practice of foster care in South Africa by describing the processes and services the social worker must render in the management of foster care, namely assessment (intake, prevention, early intervention and investigations), placement or statutory services and supervision of the placement, which also includes the reconstruction and reintegration services (DSD, 2009). The literature confirms that social workers' decisions in rendering foster care services must be made explicitly based on professional knowledge and sound reasoning using robust assessment tools and empirical evidence (Taylor, 2012).
This view was confirmed when the Ministerial Committee on the Review of the White Paper for Social Welfare, 1997, conducted an extensive consultation from 2013 to 2016 with social service professionals in public and private practice, officials across all government sectors and with the public to determine the social issues and challenges confronting the social development sector (DSD, 2016). The Committee's March 2016 report raised concerns about the decisions social workers made in placing children in foster care. The report stated that most of the decisions social workers made were not in response to the need for care and protection services, but that social workers concentrated on using the foster care system to grant children living in poverty access to income support to alleviate poverty. It alluded to the fact that social workers rendering foster care services were making decisions based on intuition, namely their experiences and knowledge of the environment in which they functioned (Graham, Dettlaff, Baumann & Fluke, 2015;Munro, 2008).
A search was thus undertaken for available decision-making models developed for or used by social workers rendering foster care services in South Africa. The search rendered no results, but the related literature traced assumed that social workers working in the foster care field used the mixed decision-making method, where intuition and analytical decision-making are used in a complementary and interdependent manner (O'Sullivan, 2011). The researchers embarked on a research journey to determine the decision-making processes of social workers in rendering foster care services. The next sections present a literature review, the research methodology used in the study, the findings, the conclusion and recommendations.

LITERATURE REVIEW
The following paragraphs will provide an overview of the various decision-making theories applicable to social work in general and foster care services in particular.

Decision-making theories
The literature indicates the most predominant decision-making theories in social work are the analytical theory, also referred to as the cognitive theory; the intuitive theory referred to as experiential, naturalistic, or discretionary theory; and the mixed theory, with fluidity between the analytical and the intuitive theories (Hackett & Taylor, 2012;O'Sullivan, 2011;Taylor, 2007).
The analytical theory of decision-making focuses on the mathematical modelling of understanding how a rational person ought to make clinical decisions (O'Sullivan, 2011;Taylor, 2012). It is defined "as a step-by-step, conscious, logically defensible process based on evidence" (Hammond, 1996: 60). According to Collins and Daly (2011), evidence-based social work practice is increasingly influenced by three elements, namely research, the views of clients and practice wisdom. Practice wisdom is information or knowledge gathered from multiple sources pertaining to a specific case. These include prior case histories, notes of social workers' own observations, reports of other professionals, the client's views, previous experience and knowledge of the social worker (Collins & Daly, 2011). Decision-making is highly influenced by legislation and procedures, which give social workers a backup for their decisions and a framework within which to work (Collins & Daly, 2011). These include legislation, procedures and guidance that indicate how you should perform, but there is still the individual judgment you must draw on to make decisions as a social worker (Collins & Daly, 2011). In social work many tools can make decisions explicit, informed and systematic, and these include assessment frameworks, decision trees, matrixes, checklists, evidence from research, and risk assessment and placement tools. Frameworks, decision trees and matrixes can help decide the best course of action to take, indicating the most desirable and the least undesirable outcome (Touhey, 2007).
Intuition has been variously defined as unconscious intelligence and the absence of analysis (Hammond, 1996), the non-conscious processing of data (Hamm, 1998), or making judgments and decisions without conscious deliberation (O'Sullivan, 2011). The intuitive school of decision-making studies shows how people reason and describe reasoning styles (Taylor, 2012). The very nature of intuition means that it is susceptible to mistakes, bias and distortion, and hence requires the safeguards of high levels of reflexivity (O'Sullivan, 2011). The lack of supervision gives social workers extraordinary discretion to make decisions and allocate their resources as they determine best (Dhludhlu & Lombard, 2017;Tupper, Broad, Emanuel, Hollingsworth, Hume, Larkin, Ter Meer & Sanders, 2017).
In the mixed theory of decision-making, intuition and analytical decision-making are complementary and interdependent (Hackett & Taylor, 2014;Helm, 2011;O'Sullivan, 2011;Taylor, 2007). Sheppard (2006) argues that social workers must apply practical reasoning in their work, which requires analytical ability. Intuition and analytical decision-making can also be perceived as a continuum with purely formal, analytical methods at the one end and blind intuition at the other (Hackett &Taylor, 2014;O'Sullivan, 2011).

Phases in rendering foster care services
The Guidelines for the Effective Management of Foster Care in South Africa (DSD, 2009) divides foster care into the assessment phase, placement phase, as well as supervision and support phase, each phase with its own decision-making demands.
The assessment process in foster care services involves determining whether the child needs care or whether the biological parents can perform their duties to meet the child or children's needs, and whether the foster parents are suitable to care for the child or children, as well as identify areas for further development (Beckett, 2007;Brown, 2014). The assessment phase is conducted in three stages: intake, prevention and early intervention, and investigation (DSD, 2009). Risk assessment and safety assessment tools are often used during this phase (Epstein, Schlueter, Gracey, Chandrasekhar & Cull, 2015).
During the placement phase, matching the foster child with the foster family is a significant factor. A suitable match between the foster family and the foster child is associated with better outcomes relevant to child development. Hence, choosing the best foster parent for the child is essential (Zeijlmans, Lopez, Grietens & Knorth, 2018). The participation of foster children in the decision-making processes on their placement is the cornerstone for successful placements (Schiller & de Wet 2018).
The role of the supervising social worker includes checking standards, responding to concerns and allegations, and ensuring compliance with policies and procedures (Brown, Sebba & Luke, 2014). They make decisions when managing risks, ensuring safety, reviewing parenting plans and monitoring the impact of the placement on the child and the household .
The research methodology adopted in the study is outlined in the next section.

RESEARCH METHODOLOGY
This study was undertaken to understand the decision-making processes of social workers rendering foster care services, a phenomenon that is not well understood and articulated. A multiple case study design was adopted, because it describes the relationships between issues, elements, personalities, characteristics, situations and processes (van Niekerk, 2009). The unit of analysis was the decision-making processes of social workers rendering foster care services in government and NGOs, respectively, within each of the five districts in Gauteng, namely Tshwane, West Rand, Johannesburg, Sedibeng and Ekurhuleni.
The researcher used purposive sampling, a method of non-probability sampling based on a set of criteria or characteristics that would provide the depth and optimal information required in the study (Mutinta, 2013;Yin, 2011). The sample comprised a social worker and his or her supervisor from both government and non-government organisations, respectively, rendering foster care services in each of the five districts in Gauteng. Each case was comprised of a case file, the social worker and his or her supervisor.
Ethical approval for the study was obtained from the Department of Social Work Research Ethics Committee of the at the University of South Africa (2019 -SWREC -05404371). The data-collection instruments were piloted. Permission was obtained from the client, foster parent and foster child or children to read and use the data in their respective files, assuring them of the protection of their confidentiality and anonymity by using codes as references for the information. All other ethical considerations were adhered to, including informed consent, debriefing of participants and the relevant protocols for permission from the Department, NGOs, participants and clients.
Triangulation of data-collection methods and sources, namely semi-structured face-to-face interviews with social workers and supervisors, and a document guide for analysing the case files, were used to enhance credibility (Yin, 2009). The researchers analysed each case using the document guide mapping out the nature of decisions and decision-making processes of the social worker during each stage and phase of foster care. During the interviews with social workers and supervisors, while following a semi-structured questionnaire in gathering data, references would be made to the nature of decisions and the decision-making processes recorded in the case file for clarity, confirmation and in-depth understanding. To ensure transferability, the researcher provided a thick description of the research methodology, datacollection and data-analysis processes.
The thematic analysis of data included a combination of the essentialist (top-down) and the constructionist (bottom-up) approaches to the analysis (Terre Blanche, Durrheim & Painter, 2014). In analysing the data, the researcher used the top-down approach drawing in the literature review, the research questions and the Guidelines for the Effective Management of Foster Care in South Africa (DSD, 2009) to develop the themes and sub-themes. The bottomup approach entailed reading and re-reading the transcripts several times for underlying messages, interpreting underlining messages and eliciting sub-themes and categories of data (Terre Blanche, Durrheim & Painter, 2014). The findings, with reference to the literature control, are discussed in the following paragraph.

FINDINGS
Seven social workers and five supervisors from the government participated, while eight social workers and six supervisors from NGOs formed part of the study. Pseudonyms were used to identify participants. The participants were asked the question indicated below during each stage of the assessment, placement and supervision phases of foster care services: • What were the decision-making processes and the nature of decisions made during each stage of the foster care services' assessment, placement and supervision phases?

Theme 1: Participants' descriptions of the decision-making processes and the nature of decisions made by social workers during the assessment phase of foster care services
The assessment phase is the most critical phase in foster care and informs the decisions that must be made in respect of the child's best interest, safety, care and protection (Brown, 2014;Ney, Stoltz & Maloney, 2013). As mentioned, the assessment phase is conducted in three stages: intake, prevention and early intervention, and investigation (DSD, 2009).

Sub-theme 1.1: Descriptions of the decision-making processes and the nature of decisions made by social workers during the intake stage of foster care
Intake is defined as the process of the initial assessment of the client and analysis of data that a client will provide (Alex, 2018; Glasson, 1965). Participants stated that in the case of straightforward applications for foster care by relatives and grandmothers, the main decisionmaking process at intake primarily entailed the social worker collecting the relevant information to decide whether a child needs care and protection, and whether the potential foster parent is suitable to foster the child.
They added that when a child is at risk of abuse, neglect and abandonment, their decisionmaking processes involve immediately removing the child to temporary safe care (TSC). They stated that the decision-making processes are often made procedurally and intuitively.
When it is a crisis, there is a strong possibility it is a removal. Then the intake worker talks to me and I talk to the crisis worker and she may have to go out and investigate the home circumstances and that may be the start of foster care when the crisis worker goes out and investigates. (Lerato) Kirkman and Melrose (2014) state that decisions made rapidly, generally unconsciously, on events based on prior experience and knowledge gained over time entail decision-making based on intuition. Without an analysis and complete information, quick decisions are based on the social worker's intuition (Collins & Daly, 2011;O'Sullivan, 2011).
All participants were asked what tools they used to assess the child's risk and safety, the child's need for care and protection, and the child's best interests. While some participants indicated the intake form as the primary tool for decision-making in this stage, others stipulated using intuition over and above standardised forms.

I don't use any tools. I call my clients in and listen to their problems and together
with them decide what would be the best solution to their problem. In straightforward foster care applications, I know what information is necessary for the court, I just get that information. (Gail) In cases of abuse, neglect and abandonment, there is a special pack of assessment forms developed by the government for all social workers who work with children in need of care and protection in both government and NGOs, but it seems that these are not always used.
We don't have a risk assessment form and don't do risk assessments. One of the intake social workers is quite experienced so she uses her own knowledge and experience to assess situations. (Rose) Decision-making processes in child protection should take place in an inter-professional context because of the nature of risks that threaten the wellbeing of children (Corby, Shemmings & Wilkins, 2012). Government norms and standards support the use of multidisciplinary decision-making (DSD, undated). However, the participants stated that decision-making processes did not involve the use of multi-professional, multidisciplinary decision-making, case conferencing or panel decision making in assessing the child or in case planning.
Intake workers work alone and make decisions alone. There is no panel or multidisciplinary decision making even if a child is placed in TSC in a child and youth care centre. If a social worker feels she requires the wisdom of the supervisor, she will do so. (Gail) Participants were asked to describe the nature of their decisions and provide some substantiation, and the responses from government and NGOs and between social workers and supervisors varied somewhat. Three participants in NGOs and six in government described the nature of their decision-making as intuitive; three participants in NGOs and none in government described the nature of their decision-making as analytical; four in NGOs and five in government described the nature of their decision-making as being mixed. Some participants described the nature of their decision-making as intuitive, because at intake they followed procedures, applied criteria, responded to threats and ensured the child's safety, which Kirkman & Melrose (2014) refer to as skilled intuition.
The decisions at intake are mostly intuitive. We just accept what the client tells us and then refer for investigation…We don't use research, previous case histories or assessment tools or even therapy during the intake stage. (Boitumelo) Some participants described the nature of their decision-making as analytical, while others saw the process as a combination of analytics and intuition.

Sub-theme 1.2: Descriptions of the decision-making processes and the nature of decisions made by social workers during the prevention and early intervention stage of foster care services
The Guide for the Effective Management of Foster Care in South Africa (DSD, 2009) describes preventative services as outreach and awareness campaigns to promote foster care programmes. Early intervention strengthens weak and dysfunctional families with resources and services without having to remove children and place them in alternative care (DSD, 2009).
Participants in both the government and the NGOs stated they did not have a tool to assess individual or community needs for preventative programmes and development. They used statistics on their caseloads to determine community needs regarding foster care.
I use my caseload to determine the needs for preventive work at the community at group work level… We mostly respond to the demand from the community on social issues and how to address them. (Gail) The participants stated that prevention programmes included providing the community with information on foster care, namely the requirements for the placement of children in foster care and information on access to the Foster Child Grant, group work and parenting skills (Hearle & Ruwanpura, 2009;Patel, 2015).
The question posed to participants was whether the decisions in preventive services are intuitive, analytical or mixed within the prevention and early intervention stage. Four participants from the NGOs and six from the government stated that the nature of their decisions was intuitive. Two participants from the NGOs stated the nature of their decisions was analytical, and none from the government made this point; four participants from the NGOs and four from the government stated that the nature of their decision-making was mixed.
Some participants stated that the nature of their decision-making was intuitive.

Sub-theme 1.3: Participants' descriptions of the decision-making processes and the nature of decisions made by social workers during the investigation stage of foster care services
Investigations are described as the process whereby the social worker examines the developmental and psycho-social circumstances of all parties concerned with the foster care application or placement (DSD, 2009). Saltiel (2014) states that home visits are one of the social worker's most important functions, yet there is a paucity of literature on this function. The participants also agreed on the need for home visits and what should be done during a home visit, namely gathering information to decide whether a child needs care and protection. Most social workers used Form 38 (DSD, 2009;DSD, 2012;RSA, 2006) to collect information on which they made written notes that also served as process notes on the file for the investigations. Form 38 is not an assessment sheet, but social workers in practice use it as an assessment tool. Some NGOs followed a report template similar to Form 38:

I don't use any tools but I follow the Form 38 template. I have done the same thing for 14 years and you end up implementing from your head based on your experience you have had. (Fikele)
Others referenced the Cane tool, the Circle of Courage, and the Heimler tool. The Circle of Courage is an assessment tool that moves away from pathology and focuses on development (Perumal & Kasiram, 2008).
Social workers stated that they rely extensively on school and teachers to assess, monitor and report on the child's academic progress and wellbeing over the period of supervision. The schools serve as a detection mechanism for problems in children's placements (Mampane & Ross, 2017). Social workers in government and NGOs conducted office interviews with the client and child or children concerned, when they could not obtain all the information they required during the home visit. They often used telephonic interviews as well.
The social worker is the only decision-maker during the investigations stage, and decisions are made by experiential cognition, following the procedures of the legislative frameworks, which Kirkman and Melrose (2014), O'Sullivan (2011) and Munro (2020) describe as skilled intuition.

If the need arises, we use the psychologist at the clinic or the hospitals, but the waiting period is long because the demand is high. We don't have the resources to use private services and therefore the use of other professionals in the multidisciplinary approach is rare. (Sizwe)
None of the participants stated that their decision-making processes were purely intuitive, based on experience and practice knowledge over time (Collins & Daly, 2011;Kirkman & Melrose, 2014;O'Sullivan, 2011). The responses of government and NGOs varied. Three participants in government and two in NGOs stated that the nature of their decision-making was analytical. Seven participants in government and eight in the NGOs claimed that the nature of their decision-making was mixed.
Some of the participants described the nature of their decision-making as analytical.

It is analytical because we weighed the pros and cons of the situation. Sometimes the social worker's experience and gut feelings also influence the decisions. (Rose)
Some of the participants described their decisions as mixed.

The decisions are mixed. I must gather evidence to support my decisions. I use many sources to gather the information and then I use my experience to make the decisions. I also use more than one method to gather the information, namely observations, home visits and interviews. (Anna)
Theme 2: Participants' descriptions of the decision-making processes and the nature of decisions made by social workers during the placement phase of foster care services.
The placement phase of foster care is also known as the statutory phase. It is the process of removing a child from his/her biological parents by bringing the child before the presiding officers of the Children's Court for a decision to be made regarding the child's need for care and protection, and possible placement in alternative care (DSD, 2009).
Foster care placements can broadly be divided into two types. The first is based on the applications where a child was urgently removed and placed in temporary safe care. Within 90 days the placement process must be finalised in terms of the Children's Act.38 of 2005. The second is the applications for foster care where the report and supporting documents are submitted to the court, a date is obtained for the court hearing, and the inquiry is initiated and finalised on the same day (DSD, 2012;RSA, 2006).
Participants stated that in the case of children in temporary safe care, the social worker must submit all the abovementioned information plus the medical report, forensic social workers' reports and other reports as requested by the court within 90 days from the initial court inquiry (DSD, 2009;DSD, 2012;RSA,2006). I spoke to the child. I spoke to the prospective foster mother and she agreed to care for the child and we removed the child to the home of the prospective foster mother. We allowed the child to stay with the foster mother. I called the child and found out how she liked the place and I also called the foster mother and asked her how the child was adjusting. The placement was an emergency one and we did not have the time to do an individual development plan with the child and foster parent.
The placement process is generally different for government social workers because the government authorises temporary placements and must issue a certificate which is more easily accessible to their social workers than the social workers in the NGOs. The social workers must ensure that the screening report and Form 30 for temporary safe care are completed, if the child is placed with an individual. This process is not required if the child is placed in TSC at a CYCC. In all court inquiries concerning placements in foster care, the social worker is required to submit her/his report and supporting documentation, which includes the identity documents of the child and the prospective foster parents, the screening report on the prospective parents, a certificate in terms of Form 30, police clearance certificate, copy of the advertisement for the missing parent, where applicable, and an affidavit by the prospective foster parent and child agreeing to the placement, three months or 90 days before a date for a court hearing is set down (RSA, 2006: section 155[2]).

I submit the report and other relevant documentation to court and the court gives us a date for the hearing. Before the appearance in court, I discuss the report with the client. Then upon the court inquiry, if the presiding officer is happy with my recommendations, he issues an order for the placement… For us there is very little preparation because in most cases the child is already placed with the prospective foster parent. 95% of all the applications we receive are related to foster care placement. (Gail)
During the placement phase, the social worker makes decisions regarding her recommendation in her/his report to the court. The presiding officers must deliberate on the evidence and decide whether he/she accepts the recommendation or not (DSD, 2009;DSD, 2012).
Preparing children for foster care placement is one of the most important processes of this phase (Vanderfaeillie, Dament, Pijnenburg, Bergh & Holen, 2016). According to the participants, most placements for foster care do not require preparation by the social worker, because the children are familiar with the prospective foster parent and have been living with the family before the application for foster care (Drah, 2016;Hearle & Ruwanpura, 2009).
In our organisations between 70% and 75% of the children are already in placements when they approach us for the legal placements. So, if during the investigations there is nothing serious to threaten the best interest of the child, the child will be placed there. So, the decision is taken for us and we just follow the processes of placing the child in their care. The reason for the placement is the lack of visible means of support. (Unathi) Participants stated that they did not have or use a placement tool to match the child with the foster parent. There was no evidence of a placement matching tool being used to assess the stability of the placement. The presiding officers are the ultimate decision-makers in the Children's Court inquiry to determine whether a child requires care and protection. The social worker only makes a recommendation for the wellbeing of the child. The Children's Act 38 of 2005, Section 177, allows a child to appeal a decision (RSA, 2006).
As noted by one participant: The Commissioner of Child Welfare is the main decision-maker. (Ida) The nature of the decisions in this phase is legally based and must be evidence-based and in accordance with the provisions of the Children's Act 38 of 2005 (RSA, 2006). The social worker's role is to ensure that the evidence he/she presents supports his/her recommendation. Four participants in the NGOs and eight in government described the nature of their decisions made during the placement phase of foster care as being mixed, four in the NGOs and two participants in government described the nature of their decision-making as being analytical. One participant in an NGO and one in government described the nature of the decisions made as being intuitive.
Most of the participants described the nature of their decisions made as mixed.
The placement decision is mixed. It tends to be more analytical than intuitive because there is now a third party that validates the evidence the social worker presents against the law and the supporting evidence. (Ida) Other participants described the nature of their decisions made as being analytical.

I would say it is analytical because it is made by the presiding officers in terms of the legal prescripts and the social worker's report. (Mike)
Two participants described the nature of their decisions made as intuitive.
How would I describe the decisions? I think it is mostly based on experience, cultural practices, and one's values. In other cases, we use screening to fit the prospective foster parent and the child. We don't have too many choices but we try within the limitations. (Unathi)

Theme 3: Participants' descriptions of the decision-making processes and the nature of decisions made by social workers during the supervision phase of foster care services
The Guidelines for the Effective Management of Foster Care in South Africa (DSD, 2009) state that "supervision and aftercare services refer to the support and therapeutic services provided to the foster family, biological parents, family of origin and the child after statutory placement".
Such support should include a range of activities such as implementing a case management or intervention plan, monitoring the foster care placement, management of the extension orders, transfers and discharge orders, reunifying the child with a family of origin, and preparation for independent living . Most participants stated that during supervision they did not use any tools for assessments, and their social work services were not based on any plans developed from assessments.

No, we don't have any tools… We only do an individual development plan for the 159 extensions to meet the court requirements. (Elda)
I like using the Heimler tool and the skills wheel. The IDP is a plan that we manage, but does not measure the level of development of the child. We have to develop an indicator tool to measure the IDP.… (David) Tupper et al. (2017) found in their study that despite the presence of assessment tools, social workers preferred to rely on skilled intuition to guide them in decision-making, as in this narrative excerpt: Yes, there is a requirement for the IDP but I do not see it as a must. We tend to focus on the development of the child based on what we pick up and put it in our court report, but not a tool to follow up on. (Anna) The workload makes it impossible to implement any plans especially IDPs. (Lerato) Participants stated that they would only use multidisciplinary decision-making processes during the supervision of foster children, if there is an indication of abuse. Ross, Pretorius, Gerrand, Nathane-Taulela and Berhane (2008) also did not find evidence of multidisciplinary decision-making to assess the progress of children in foster care during this phase.
Several assessment, risk and safety tools are available for social workers to assist them in making more uniform, standardised and analytical decisions and minimise subjective and highly discretional decision-making (DSD, 2012;DSD & UNICEF, 2012). However, access to and training on the use of the tools and supportive systems to implement them is lacking. Hence social workers have come to rely on resources that are accessible and are largely outside themselves in terms of professional expertise, skills and experience to make decisions on foster care using skilled intuition (Collins & Daly, 2011;Kirkman & Melrose, 2014;O'Sullivan, 2011).
There was a mixed reaction from participants on the nature of their decisions made, with six participants in NGOs and five in government stating the nature of their decision-making was mixed, four participants in NGOs and five in government stating the nature of their decisionmaking was intuitive, one participant in an NGO and none in government stating the nature of their decision-making was analytical.
Some participants stated that the nature of their decisions was mixed in the supervision stage.
The decisions are mixed but more intuitive. The court function is less intense and more oversight and hence the social workers' decisions are less likely to be questioned with external evidence. (Ida) Other participants stated that the nature of their decisions made was intuitive in the supervision stage.

Most of the decisions are intuitive based on the cultural and personal values of the social worker. Most often social workers cannot explain the reasons for their decisions. (Unathi)
One participant indicated that the nature of her decisions was analytical in the supervision stage.

DISCUSSION
Findings indicated that social workers were not always consciously aware of their own decision-making processes when rendering foster care services. This corresponds with the literature referring to data being non-consciously processed or decisions being made without conscious deliberation (Hamm, 1998;O'Sullivan, 2011). The predominant decision-making process of social workers was the mixed process, where the decisions had to be supported by evidence, especially during investigations, placement and supervision phases of foster care services. This finding is supported by Kirkman and Melrose (2014), Munro (2020) and O' Sullivan (2011), who stated that mixed decision-making processes are based on a combination of experiential cognition, procedures, legislative frameworks and professional knowledge. The intuitive process was the second most common decision-making process used during the intake, prevention and early intervention stages of the assessment phase. It is used by social workers when they must make decisions rapidly and in the absence of analysis or evidence (Collins & Daly, 2011;Kirkman & Melrose, 2014;O'Sullivan, 2011).
The decision-making processes of social workers during the different stages and phases of foster care services are elaborated on below. During intake, most participants stated that their decision-making processes were intuitive, followed by the mixed decision-making process. A small number of participants stated that their decision-making was analytical. The intake stage involves a quick data-collection process to determine if the case should be investigated. In the absence of analysis and complete information, quick decisions are based on intuition (Collins & Daly, 2011;Kirkman & Melrose, 2014;O'Sullivan, 2011). There was also an absence of conscious deliberation when the prospective foster family was related to the child. This finding is confirmed by O'Sullivan (2011). Some participants stated that their decision-making at intake is a mixed process when they receive reports of child abuse and neglect. Then decisions are based on analytical assessments of risk and safety, medical, psychological and forensic reports, professional knowledge, experiential cognition and discretion (Kirkman & Melrose, 2014;Munro, 2020;O'Sullivan, 2011).
The prevention and early intervention stages were also characterised by intuitive decisionmaking, where half the participants stated they only disseminated information to the community on access to foster care services. There was no conscious deliberation on decisions, but a routine process of creating awareness (Hearle & Ruwanpura, 2009;O'Sullivan, 2022;Patel, 2015). This was followed by a smaller number of participants who stated they used the mixed decision-making process by gathering and analysing data from their case files and compiling statistics on issues, doing research, obtaining information and observing (Hackett & Taylor, 2014). An even smaller number of participants stated they used purely analytical decision-making processes using the data from their case files and statistical compilation of these data to guide their community intervention. This finding corresponds with the work of Taylor (2012) and O'Sullivan (2011).
Most social workers used the mixed decision-making process during the investigation stage of foster care services, when they relied on multiple sources for information, including their professional knowledge and experiential cognition (compare Kirkman & Melrose, 2014;Munro, 2020;O' Sullivan, 2011;Saltiel, 2014). A minority of participants stated that their decision-making process was analytical, as they were using Form 38 as a guideline to gather the evidence. However, Form 38 is not an independent, defensible tool. It is a checklist of information required to compile the court report. Kirkman and Melrose (2014) state that applied criteria and procedures used in decision-making are skilled intuition.
In the placement phase of foster care services, the social worker's decisions are presented in recommendations in the court report, and social workers rely on the mixed decision-making process. In some cases, the court process involves several decision makers, namely the social worker, the child, the prospective foster parent, the Commissioner and the legal representative to interrogate the evidence before the court. At this phase the decisions made are based on the framework of the Children's Act No. 38 of 2005(RSA, 2006. Decisions are based on analytical assessments of risk and safety, medical, psychological and forensic reports, professional knowledge, experiential cognition, assessments on the suitability of foster parents and the match with the child, legislative procedures, preferences of the child, as well as discretion (Hackett & Taylor, 2014;Helm, 2011;Kirkman & Melrose, 2014;Munro, 2020;O' Sullivan, 2011). A minority of the participants stated that the decision-making process during this phase is analytical, because decisions are made on assessments of risk and safety, medical, psychological and forensic reports, and legislative procedures. The literature confirms that decisions must be made explicitly based on professional knowledge and sound reasoning using robust assessment tools and empirical evidence (Taylor, 2012).
Most participants stated that their decision-making processes during the supervision phase of foster care services were mixed. Some reported using the circle of courage to assess and plan services for the child, rehabilitation of the family, and the reintegration of the child with the biological family, together with their professional knowledge and skill, experiential cognition and discretion in determining the best interest of the child Perumal & Kasiram 2008). The supervision phase of foster care can involve investigations and the transfer of the child to another foster parent, or to a child or youth care facility, or to the biological parents. A minority of social workers used the intuitive process of decision-making. Social workers had the necessary assessment tools to assess risk and safety, but did not implement them because it is time-consuming and complex. Similar findings were reported by Tupper et al. (2017). There was also no evidence of multidisciplinary decision-making processes to assess the child's progress during the supervision phase of foster care services (Ross et al., 2008).
Social workers and their supervisors did not always agree on the type of decisions made, and their decision-making processes were not always transparently reflected in the case files.
Although analytic assessment tools are available, most participants did not use them because of a lack of time and high caseloads (Collins & Daly, 2011;O' Sulivan, 2011;Tupper et al., 2017). The lack of supervision gives social workers extraordinary discretion to make decisions and allocate their resources as they determine best (Dhludhlu & Lombard, 2017;Tupper et al., 2017).

RECOMMENDATIONS
A number of recommendations are made below.

Practice
• Social workers rendering foster care services should be made aware of their decisionmaking processes during the different phases of foster care.
• All decision-making processes should be transparently reflected in process reports with evidence, and all assessments should be filed.
• Quarterly supervisory audits can be conducted on all case files as per the supervision norm to ensure that recordings in the files are complete, accurate and current.
• Available assessment tools should be used in the relevant foster care phases.
• New assessment tools should be developed and used, e.g., a decision-making matrix on rendering foster care services, a matching tool for placing children in foster care, and a framework monitoring the development of children in foster care.
• Tertiary education should include the use of professional tools for the assessment of children and families and the design of intervention strategies for children in foster care, as well as for the development of social work skills to engage in multi-professional and multidisciplinary decision-making.

Further Research
• It is recommended that the involvement of children in social workers' decision-making processes in foster care services be researched, as children are the centre of the social workers' decision-making, for the sake of ensuring stability in their placement and their long-term developmental outcomes.
• The impact of new legislative reforms on the extended child support grant for orphans on social workers' decision-making processes in rendering foster care services should be further researched.

CONCLUSION
The decision to place a child in foster care or not ranges from being complex and lifethreatening to a routine decision or rubber-stamping of a prevailing situation. Social workers must be consciously aware of their own decision-making processes during all the phases of rendering foster care services. Transparent decision-making with supervisors and other stakeholders leads to accountable decisions. However, it is not about making the right or justifiable decision per se but making the decision with the best outcome for the foster child and his/her biological and foster family.