Decision-making in social work

One of the issues that chased my mind when I worked in a social work department many years ago was how social workers always needed to discuss what to do with colleagues. My assumption was that it would be easy to gather the relevant information, and to make a judgement followed by a decision. When asked how they made their decisions, they often referred to discussions in the group instead of giving me a straightforward answer. Nobody seemed to be able to explicate how decisions were made, even in cases where children were placed in care. Only in cases with the use of coercion, a decision was pinned out on a meeting in a statutory board. Sometimes social workers presented the case with convincing arguments documented in the casefile. Sometimes they needed help and involved colleagues and managers – the group.

These questions remained in my mind. What did social workers consider as important information, which data were relevant, and how did they systematise and analyse it? What happened when information was gathered and judged and in the end put into a decision of doing something for the benefit of the child? How did they communicate with the child and the parents after the decisions made? Did the case record reflect what happened e.g. did it comply with the law?

A few years ago these questions returned when we in two separate research projects investigated changes in how social work was organised according to the regime of New Public Management, and how the focus on costs influenced decisions in child protection cases (Nørrelykke, Zeeberg, and Ebsen 2011; Schrøder 2014). The projects showed how the social workers and their organisations described their work as characterised by linearity in decisions:

(1)	Referral of a child in possible need
(2)	Decision of further assessment or dismissal
(3)	A profound and thorough assessment
(4)	Writing down a child’s need
(5)	Decision of doing something or dismissal
(6)	Making suggestion to an action plan
(7)	Decision on action plan
(8)	Putting the action plan into practice – ordering to delivery agencies and making contracts
(9)	Sometime after make a follow up to check everything followed the action plan.

This order reflects Danish social and administrative law – often of statutory nature – supplemented by administrative guidelines. The order has developed over the last 3 decades with new demands and concepts: Assessment schemes, action plans, child involvement, child interviews, contracts with suppliers, negotiation of prices, deadlines, quality standards, and digitalized procedures. It appears as a line of order that social workers in their local settings easily can carry out and when doing so certainly will give the vulnerable child the best service. The same kind of thinking can be found in England and Sweden (Munro 2011; Ponnert and Rasmusson 2015).

The ideal line of order is also an order of who should make the decisions. The assessment of the referral and a following profound assessment is often the result of a caseworkers decisions – sometimes discussed with her colleagues (1–3). The decisions of judging and the provision of small-scale services are made in consultative groups of colleagues and low-level managers (4–5). The decisions concerning what to do in relation to a placement in a foster family or in a residential home often involve the frontline manager and sometimes external experts as consultants (6–7). This is a parallel to the costs involved. The more it costs, the more the managers are involved. The last two tasks (8–9) end up at the caseworker’s desk and can sometimes return to meetings with frontline manager if they for some reason have to be changed. This ideal order was reflected in our research (see also Hoybye-Mortensen 2015; Sørensen 2017).

Even though the order and the law seem to considerate many aspects of child protection, it is difficult for social workers to decide of what to do. The law claims what social workers have to do within certain time limits and rules of procedure. However, it does not specify the content in the dialogue with children and parents, the methods used for collecting information, how to make a judgement, etc. The criteria for decisions in each step are in the hands of the social workers and their managers. They have to fill in the content pertaining to each heading and to comply with reminders coming from a digitalized system requiring them to report on time.

In practice this ideal order is challenged. One challenge appears when the law is fragmented into different pieces which origin from different sets of rules. An example is when social workers simultaneously have to deal with differences in children’s legal rights in law of divorce and laws of child protection. In addition, differences between national law and rules regulated by international bodies, such as the European Court of Human Rights, occur. This lack of coherence in law is one reason for the breakdown of the linear order in a social work agency (Svendsen 2017a).

Another reason is that parents and children can act in unpredictable ways deviating from the ideal order of organising social work (Svendsen 2017b). This can happen when new information potentially requires a change of a former decision e.g. when a young person placed into care in a foster family gets into a fight with the foster mother (Bengtsson 2017). Children and parents who are subject to social services try to understand what is going on, try to change their mutual relations, may try to avoid interventions from the social system, etc. They can change position and attitudes for reasons that do not seem obvious, and differ from what they agreed to in conversations with the social worker. Also social work can be unpredictable. Social workers may have to violate the ideal order because of staff turnover, changes in services available, changes in law or local policy guidelines, cuts in budgets, etc. These potential changes and movements make social work difficult to predict and often challenge the ideal of linearity.

It is well known that decision-making in social work is a complex task (Munro 2008). Munro et al. mention two challenges from which complexity arises (Munro et al. 2016). One is the missing consensus about what is best to do. Another is how social workers have to compress information from children and parents to stories of what might happen and how to support the child. The complexity means that it is difficult to find one right way to deal with child protection. Social workers will have to use short cuts – heuristics – to reduce complexity and get along.

The literature on decisions has recently been defined from two positions. One is from the German Gigerenzer, who argues that humans act by reducing complex situations to certain signs guiding their actions e.g. by holding on to the first impression of a person (Gigerenzer 2014). This can be effective and is often build on experience and supported by discussions with others. But also the opposite can be the case. The other position originates from the American Kahneman, who perceives humans as Gigerenzer, but uses ‘bias’ to characterise the uncertainty that occurs when short cuts are used. He argues that all kinds of bias should be reduced (Kahneman 2011). In social work literature both positions are reflected and different kinds of heuristics/biases are described (Taylor 2010; O’Sullivan 2011; Munro et al. 2016). They all presume that social workers have to reduce complexity by using short cuts.

Not only on an individual but also on an organisational level bias/heuristics will occur. Routines, meetings, and hierarchies in decision-making develop and is maintained in everyday practices. Institutions forms how things gets done (Berger and Luckmann 1972). Simon suggested how routines encouraged satificing to make workers do what was accepted as sufficient instead of optimal solutions (Simon 1997). Especially this was the case if the task was difficult or resources were limited.

A third level to be taken into account with regard to what influences decision-making is the institutional environment of a child protection department. Neo-institutional theory argues how this happens through different kinds of imitation (mimetics) (Dimaggio and Powell 1991; Johansson 2002; Linde and Svensson 2013). Not only has the parliament’s change of laws power to impose uniform methods. Also ideas from other organisations or professionals can change social work through imitation. Managers and social workers are in touch with people outside their organisation, which might influence their decisions. In Denmark, the introduction and use of assessment tools like ‘Integrated Childrens’ System’ and ‘Signs of safety’ can be explained by the use of such channels of distribution.

I began this Editorial with introducing a linear approach to decision-making in child protection. To summarise: This is challenged from a lack of consistence in law; from difficulties in knowing what to do and from reducing lots of information to a solid base for decisions, often followed by a need to make changes as time passes by. This complexity is reduced by using short cuts (heuristics/biases) which are influenced at an individual level by the social worker, at an organisational level and by different channels outside the organisation of child protection. The linear approach makes it plausible to criticise what went wrong or right in hindsight, but overlooks and simplifies what actual happened, and the reasons for actions at the time being. This is an easy and traditional way of putting forward a certain ideal. However, such a reduction does not develop our understanding of the complexity of decision-making. Therefore, we might be better off by accepting and seeking to understand how decisions are made in practice, and by discussing how they can be made in a way that in the end benefits the children. By taking this approach we might find ways of improving processes of decision-making that actually reflects the purpose and practices of social work.

The aim of this issue is to expand our understanding of the complexity of social work. It presents research that from various points of view shows how difficult it is to understand social work as a linear chain of events. Moreover, it illustrates how often such a way of thinking dissociates from practice.

The first article by Magnussen and Svendsen is about short cuts in decisions in child protection (Getting there: Heuristics and biases as rationing shortcuts in professional childcare judgments and decision-making – an integrative understanding). The article is based on selected examples of research on decision-making. The purpose is to develop an integrated understanding of heuristic and biases. The selected examples present rationing mechanisms and heuristics that are meaningful at the time when action takes place, but can be interpreted differently in hindsight. In the second article, we move outside the child protection department. Markström and Münger’s study shows that teachers are insecure and emotionally influenced when they decide whether to report or not about domestic violence (The decision whether to report on children exposed to domestic violence: Perceptions and experiences of teachers and school health staff). The article presents teachers’ dilemma of acting as an important social person for everybody in a class and/or deal with a specific child who report about violence. In the third article Barfoed looks into the use of a specific tool – Addiction Severity Index (ASI) – aimed at making decisions concerning adults’ use of drugs or alcohol (From stories to standardised interaction: Changing conversational formats in social work). Barfoed’s study shows how an elaborated and tested assessment instrument needs ‘off-track interactions’ to make progress in the conversation with the client. Such off-track interactions depend on whether the social worker and the client know each other. She discusses how such instruments reduce complexity in social work but also exclude important aspects of a client’s story. In the fourth article, Liljegren, Höjer and Forkby explore laymen’s influence in child protection services’ decisions on coercive actions (’I don’t want to tell you how to do your job, but …’ – Laypersons challenging the authority of professionals in Swedish child protection). Their study shows that laymen depend on social workers’ judgement and collection of information. Laymen influence the decisions by pointing out inconsistencies, cost calculation, arguments such as ‘others have rights too’, and by using their local knowledge to correct judgements. By doing this, they are able to keep a boundary to the professionals’ facilitating the decision process. The fifth article by Mackrill, Ebsen, Antczak and Svendsen discusses statutory guidelines of care plans referred to as SMART plans (Care planning using SMART criteria in statutory youth social work in Denmark: reflections, challenges and solutions). The article argues that unpredictability, power issues and discretion challenge the embedded linearity. A three-stranded model is presented as an alternative. According to this model, ideal planning should build on the young person’s/child’s/parent’s view on possibilities and resources available. In the sixth article, we move from social workers’ decisions to management. Rønningstad presents a review of empirical research on leadership in social work (Leading for better outcomes: Social work as knowledge work). Complexity leadership is divided into functions of adaption, enablement and administration. His conclusion points to the importance of the function of enablement which seems to be the best aide to increase social workers performance. Often this is disturbed by an increase in leaders’ administrative tasks. The last article by Skillmark and Denvall studies social workers’ role with regard to implementation of an assessment tool – Children’s need in focus – in Swedish social services (The standardizers: social workers’ role when implementing assessment tools in the Swedish social services). Well-educated and experienced social workers were chosen to facilitate the use of the tool. The study shows that the social workers played different roles as facilitators identified as instrumental, adaptive and transformative roles. In particular the transformative role makes it clear that assessment tools should support reflective skills in local social work.

The initiative to this themed issue of Nordic Social Work Research was taken at the Nordic FORSA and NOUSA Conference in 2016 in Copenhagen. The intention with this themed issue is to contribute to an understanding of the complexity of social work decisions in practice. I would like to thank Martin Börjesson for special support in the process of editing as well as the Editors and the Editorial Board of Nordic Social Work Research for making this issue possible. I hope it will contribute to improving our understanding and development of decision-making for the benefit of people in vulnerable positions.

Frank Ebsen
Guest editor
Institute of Social Work, University College Metropol, Copenhagen, Denmark
[email protected]