Abstract
The previous editorial discussed two recent reports on foster care which offered resolutions to difficulties in recruitment and provision (Baginsky, Gorin and Sands, 2017; Narey and Owers, 2018). In June (2018) the Family Rights Group (FRG) and Nuffield Foundation published a third overview of the care system, producing in combination with the earlier reports a total of 603 pages of analysis and discussion and 56 recommendations (this should be more but the Baginsky report did not present its recommendations in summary form).
It is not possible in a short editorial to comment on the value of such overviews and the originality of their proposals, so I will concentrate on one aspect of the discussion: the role, use and relevance of research as this fits with one of this journal’s aims of ‘disseminating new research and practice developments’.
What is initially striking is the contrast in the perceptions of the role of research in each of the reports. Baginsky and colleagues are very research conscious, laying out the work that needs to be done if progress is to be made. The FRG also push this line: their 20 recommendations contain four topic proposals, four encouragements for agencies to do research-type monitoring and 11 exhortations for agencies to incorporate findings into policy and practice. Narey and Owers, in contrast, draw on evidence to inform their conclusions, but have only one specific research review section (p. 82 on Contact and Siblings) and do not mention the word (research) in their 36 recommendations, although three of them (8, 26 and 27) suggest the benefits of some basic monitoring. No proposals are made about what more needs to be known.
Social work and science have never been comfortable bedfellows. Some argue that social work is so complex that it can never be analysed by orthodox ‘scientific methods’ while others claim that without a strong evidence base, it is little more than witchcraft, and immoral too as it imposes untested interventions on vulnerable people and wastes public money. This tension is symbolised by the contrast between those who maintain that randomised controlled trials (RCTs) and high standards of evidence are the be-all and end-all and those who say that the world is far too messy and complex for such a reductionist approach – the different contexts in which practice takes place are all too unique for evidence to generalise.
This either/or stance confounds a lot of social work practice. Possibilities are put in opposition when they should be options to match the needs of different cases. Foster care is often seen as always better than residential care, adoption is deemed wholly preferable to long-term fostering and birth families are perceived as either bursting with blocked caring potential or wicked to the core. So what can be said about research?
When research is mentioned it is often viewed as a monolith that is always helpful. Unfortunately, this is not so. A first question, therefore, is whether the issue under scrutiny is actually researchable. We have to accept that some situations are influenced by such a plethora of factors and have such long-term implications that definitive conclusions are a chimera; all that can be hoped for are well-informed indications.
But if research is possible, a second question concerns the sort that is relevant to the particular enquiry. There are at least three types: meta, grand or programme theories; empirical surveys; and small-scale evaluations and explorations. The first, the meta theories, lay out general perspectives and can only be judged as to whether they are useful; it is the hypotheses derived from them that are testable. Many that influence child care are quite old and have taken time to seep into professional thinking: Freud first published in 1894, Piaget in 1926, Bowlby in 1944 and Bronfenbrenner in 1970, and these will probably be joined in years to come by present-day neuroscientists and epigeneticists. The second type involves gathering and analysing empirical data, usually using quantitative methods sometimes complemented by qualitative examples. This can have an immediate effect when it exposes something surprising but it also charts trends and establishes statistical associations that contribute to understanding causal processes, thus offering insights into the way systems work and how users experience them. The third type comprises small-scale evaluations or descriptions of specific situations and usually relies on observation, documentary material and personal information offered in interviews and focus groups. Unfortunately, it is usually too weak to provide answers to the big questions and tends to be dismissed by pure scientists; nevertheless, it can illuminate group dynamics, sub-cultures, views and emotions, thereby providing something interesting.
The important point is that no one of these methodologies is preferable per se to the others. All can be of good or bad quality and what is likely to prove fruitful depends on what is being studied as each has particular strengths and weaknesses. Meta theories attract academic respect but may exaggerate a particular perspective, quantitative studies rely heavily on measures and statistical analysis that can easily be questioned, and small-scale studies may be illuminating but are often perceived as unrepresentative and difficult to generalise. Thus, even the most sophisticated studies have limitations on what they can tell us: for example, those responsible for tackling the upsurge in youth knife crime in the UK have reams of quantitative data on incidents, locations, times, perpetrators and victims but struggle to understand why young people seem so hell bent on killing one another. This requires a sensitive qualitative approach.
Then there is the other side of the equation and the question of what the research is intended to inform. Legislation is concerned with justice, rights, authority and process and is most likely to be influenced by representative surveys that reveal tensions in the system. Official guidance accompanying legislation tends to be general and specific to areas such as how to conduct a case review, so evidence from small-scale intensive studies might complement findings from larger surveys of how well things are working. Practice guidance, in contrast, needs to combine a strong theoretical base offered by meta theories with micro studies of family dynamics and professional–user relationships. Work with individuals has to be highly informed by meta theories and therapists look to the likes of Bowlby and Piaget for leads. This situation applies equally to professional self-reflection – the subject of two articles in this edition – where practitioners apply the theoretical perspectives they use to help clients to their own work. But it has to be said that there are circumstances in child care where research is of little immediate help; when challenged by an aggressive adolescent or an unco-operative family, practitioners have to rely on their wits and experience.
This discussion so far has fallen into a familiar trap. It has viewed research as if it comprises a collection of isolated projects. But objectivity and strength are enhanced by the institutional structure in which scientific enquiry takes place as well as by the quality of individual studies. Robert Merton argued nearly 70 years ago that objectivity does not rest in an individual project because although researchers strive to minimise distortion, there will always be bias in the methodology. Objectivity is more likely to be achieved by the institutional structure of science and the values that underpin it: communal ownership, organised scepticism, universal acceptance of the evidence and economic disinterest. Similarly, research has more impact when individual studies are part of a programme of interlinked projects, such as that commissioned by the Department of Health to provide background information for the Children Act 1989. Each project looked at a specific part of the care system but with some overlap across areas. It was so successful that the findings were widely disseminated in a series of conferences and publications that enabled the co-ordinator, Jane Rowe, to speak with more confidence than ever before about what constitutes good practice.
But there is another viper in the nest, namely the culture of social work. The profession has long been open to attack because of its eclectic rather than discrete knowledge base, its relatively modest entry qualifications and, in more gender-bigoted times, a perception that it was largely a woman’s occupation. Its status in universities and colleges was often low and no one was sure whether it was an art or a science. This issue is still salient, for example in the heated arguments about tick-box versus relational practice.
The fact is that despite research being seen as valuable in reports, it is still a minor issue in the big picture. Social workers do not get promoted because of their research knowledge and they are not held accountable for not having any; it does not matter if they have not read Bowlby. In 54 years of visiting social services offices, I have never seen a copy of an academic journal, including this one.
A good example of this lacuna occurred earlier this year in the evaluation of Frontline, an innovative way of training social workers (Dartington Social Research Unit, 2018). Here is an initiative designed to overcome an acknowledged problem by attracting to the profession clever people who would not normally have considered such a career. They are given intensive academic training, practical experience and supervision to achieve qualifications quickly. Again, this is not the place to discuss the merits of this approach but the interesting thing is that the methodology designed to evaluate the programme was highly sophisticated and would have shown fairly clearly whether or not the programme is effective. However, it failed because those asked to provide the information (mostly social workers) did not produce enough of it. Perhaps they were too busy, uninterested, anti-research or critical of Frontline or did not see evaluation as important, we cannot say; but their response was more cultural than an act of deliberate disobedience.
Naturally, social workers’ views are grounded in tough experience and must be respected, but if they as a profession choose to go down the non-scientific route, they must accept the consequences of their decision. Social work is littered with initiatives that have been enthusiastically implemented but have fizzled out. In recent years, Sure Start, Connexions, the College of Social Work, family centres and the Looking After Children materials were all hailed as promising projects but have collapsed or diminished, not just because they were overwhelmed by practical, financial and political difficulties but also because they could not produce robust evidence to justify their existence in a context of financial austerity and a chilly political climate. There are, of course, always new initiatives to improve the impact of research, a recent one being the establishment of What Works centres, and everyone should wish them well in strengthening the role of research in policy and practice, but the worry is will they last? 1
It is always easy to be gloomy but the lessons from history are more optimistic. As shown earlier, the effects of research are rarely immediate and the once castigated ideas of progressive thinkers like Jane Rowe, Nancy Hazel, Barbara Tizard, Richard Balbernie and Spencer Millham now constitute normal practice, despite the frustrating setbacks they encountered en route. As Michael Rutter once said, ‘Evidence wins in the end.’