Research Advisor & Senior Research Fellow
National Child Protection Clearinghouse
Abstract
In general, developments in child protection practice often result from the adaptation of overseas research for Australian conditions, with relatively few Australian studies being completed, or informing practice. Australian research into aspects of child protection practice has been hampered for a number of reasons, including: a lack of resources; agency culture; and the difficulties associated with investigating such complex, highly sensitive, real-world phenomena. Using his own and colleagues’ experiences of conducting research into child protection decision making and case management practices, the author first provides a brief overview of some of the major areas of investigation of child protection practice and the methods that have been used, to date. He contends that while evidence-based practice and experimental rigour are appropriate approaches for the investigation of some facets of child protection work, they are not (and cannot be) the only forms of investigation undertaken. The realities and complexities of child protection work require the adoption of methodological flexibility and the use of a variety of approaches in order to better understand the child protection system. These include: case tracking, multiple methods, case studies and action research and the incorporation of multiple perspectives (that is, child protection staff, other professionals and children and their families), where possible.
In the second half of the paper the author explores issues affecting the collaboration between researchers and child protection departments and ways to facilitate better partnerships and ensure policy and practice are informed by research. Issues identified include: access to information; agency defensiveness; the need for a truly collaborative partnership between child protection and research; intellectual property and tailoring research messages for child protection policy and practice; and the effective utilisation of research once it is completed. The paper concludes with a brief discussion of some methods that may facilitate child protection departments’ use of existing external research capacity and knowledge.
Introduction
In general, significant shifts in Australian child protection practice result from the utilisation of overseas research, such as the Messages from Research series developed by the UK Department of Health, with relatively few Australian studies being completed, or informing practice. It appears that there remains a ‘cultural cringe’ with respect to the valuing of Australian research by Australian governments, child protection departments and the wider professional sector. This may partially account for the relatively low investment in Australian research in this area.
More specifically, Australian research into aspects of child protection practice has been hampered for a number of reasons, including: a lack of resources; agency culture (defensiveness and a failure to value research); the difficulties associated with investigating such complex, highly sensitive, real-world phenomena, and the practice of researchers themselves.
What research has been done in Australia?
Despite this body of work and evidence of some degree of strategic research planning, there does not appear to be a clear, coordinated, dedicated research program operating across the nation, while most state/territory child protection departments have a clear research framework and have not developed strong collaborative relationships with external research institutions. (However Queensland for example, appears to be moving to change this situation, with close links developing between the Families, Youth and Community Care Queensland and the Department of Social Work, University of Queensland).
It is also apparent that a variety of methods and varying degrees of experimental rigour are evident in the body of research that has been completed. This is likely to result, at least in part, from the recognition that empiricism is not, and cannot be, effectively employed to investigate all aspects of complex, real world phenomena, such as child protection practice.
Clearly, an evidence-based approach is important, as is a degree of experimental rigour. However, despite its ability to demonstrate clear cause and effect relationships, a randomised control design (RCT) or quasi-experimental approach is often not possible in situ, or even desirable in every instance. Even in medicine, the ‘gold standard’ of RCT research is evident in only a small proportion of published research, it is often too difficult and too expensive (in terms of time and money) to be utilised. The bottom line is that there are a variety of research methods that can provide a degree of experimental control, reliability and validity. The trick is to tailor the methods to the research question being investigated and the situational constraints. For those reasons I would advocate the use of a multiple methods approach, combining quantitative and qualitative methods to provide the best understanding of applied social phenomena, such as child maltreatment and child protection work (see Tomison 2000).
Second, I believe that what is required is the development of a comprehensive picture of what works, not just a measure of outcome or ‘success’ as this does not tell us why a particular initiative is successful. What is required is the equivalent of a basic program evaluation model, consisting of a hierarchy of steps that build to a comprehensive evaluation.
Specifically, these include should include:
Baseline or Input stage: where the situation to be addressed is documented, aims, objectives etc. reported and the development of the program or initiative described.
Process (also known as implementation or formative evaluation): The extent to which a program or initiative is operating as intended via the assessment of ongoing program elements and the extent to which the target population is being served. That is, how the program is achieved, any modifications undertaken, which program elements have led to a successful outcome.
Outcome (impact): The extent to which a program or intervention affects participants on a set of specified outcomes, variables or elements; the effect on clients, workers, wider society. Outcome studies are often the focus of research where a primary aim is to demonstrate success to funding bodies.
I would argue that it is a mistake to ignore input and process in order to focus on outcomes. The latter is only part of the picture and does not provide a determination of why a program works or doesn’t work.
Overall, while an evidence based approach is desirable, you can’t walk before you can crawl. That is, there needs to be an investment in a research base that supports comprehensive investigation - input and process analyses in combination with impact assessments. Determining why something works is just as important as determining if it works at all.
How can the use of research be facilitated?
Governments
Taking as ‘given’ the issues of inadequate funding of research in this field, other issues require addressing if research and the use of research is to be facilitated.
Substantial changes to policy and practice in child protection systems have often been implemented without careful, evidence-based consideration of the efficacy of existing systems, or evidence that the new initiative will have a significant, positive impact. This tendency has been exacerbated by the crisis-led approach to development in child protection. Child deaths and regular, adverse media coverage on a variety of aspects of child protection practice (leading many workers and services to feel that when making decisions or taking action they are ‘damned if they do, damned if they don’t’) help to create a climate where one has to be seen as making some form of response to alleviate concerns. These responses have often not resulted from the careful consideration of evidence-based practice. Exacerbating the trend towards action, the political imperative of Ministers needing to be seen to make changes and influence the direction of government portfolios has also hampered attempts to make strategic decisions that are informed by evidence-based practice.
Clearly, methods need to be found to combat (or at least minimise) the tendency of governments to adopt a ‘quick fix’ or ‘innovation-led’ approach to policy making and service development in the child protection field. Similarly, child protection departments must be able to take the time to comprehensively assess service limitations and the implications of advocated changes. Further, the tendency to restrict practice initiatives to pilot projects of limited duration that fit within the term of government office also requires action, given the negative impact such a policy has on the ability to adequately trial programs and to determine their efficacy.
Addressing such issues leads however, to a Catch 22 situation. Fighting the ‘quick fix’ tendency, or at least ensuring such decisions are formed from a research base, requires the development of a body of research knowledge (raising the question of adequate resourcing of research programs). Producing that research however, will require the funding of longer term research and practice initiatives with timelines that allow adequate assessment, an investment in research and a slower approach to the implementation of changes to practice. It would appear better to invest in research now in order to develop the knowledge to intervene more effectively over time.
Uniformity
Even allowing for some regional or statewide differences in populations, community needs and service infrastructures, the challenges and solutions facing the different child protection sectors are remarkably similar. Thus it should be that research findings are also, to a large extent, generalisable. Generalisability would be being further enhanced by the State/Territory child protection services moving to adopt uniform definitions of maltreatment, case outcomes and data collection processes (although this would require legislative changes).
In recent years, despite attempts by the Australian Institute of Health and Welfare and SCCSISA to develop uniform data definitions that would enable the collection of uniform community services data from across the country (SCCISA 1998), child protection systems appear to be becoming more divergent in their approaches, such as the introduction of case streaming models in Western Australia and Tasmania, and the differentiated response system developed in Victoria. Such trends will hamper the generalisability of research investigating child protection systems and reduce (to an extent) the value of a national research plan of action.
Cross-sectoral collaboration
Child maltreatment is a complex phenomenon that may reflect the degree of underlying social problems in a family, community or society (Melton & Flood 1994). The adequate prevention of child maltreatment requires that an holistic approach be adopted in order to address what are often multiproblem, disadvantaged, dysfunctional families. It has been demonstrated that attempts focusing primarily on remedying a single family problem are often not as effective as approaches that utilise a multivariate, holistic approach. Such programs target the influence of constellations of family factors and/or problems, often working in collaboration with other services (Tomison 1996; Durlak 1998).
There are clear associations between a variety of social ills and child abuse and neglect. There are similarities between the risk and protective factors underpinning the development of a range of social ills. Over time, a high degree of congruence has developed between the prevention of the various forms of violence and/or social ills, in terms of the priorities and strategies for action that have been proposed and undertaken. Clearly therefore, the prevention of child maltreatment and other social ills, and the enhancement of the professional systems would be facilitated by greater cross-sectoral collaboration and coordination from government, researchers and non-government agencies from policy-level linkages down to the enhancement of relationships between sectors and agencies at the service provision level. In research terms, this could be facilitated by the development of a cross-sectoral Research Council.
National Research Council
A National Research Council (intra- and cross- sectoral), perhaps linked in with the Australian Research Council, and consisting of government and non-government representatives and policy makers, but with a preponderance of researchers (with appropriate sector sub-committees), should be formed to regularly advise on new developments in research and practice. More importantly, the Council should be charged with the development of a National Research Strategy.
Such a framework or ‘master plan’ would work to ensure uniformity of purpose, clarity of mission, and enable large-scale cross-sectoral research initiatives (with pooled funding), while also forming a base for sector-specific interventions. The Council would also ensure communication was maintained between the various sectors and states, and enable the sharing of new research and lessons from practice. Significantly, if such a body was strongly supported by the States/Territory departments, it would provide a unique power base from which to lobby for increased spending on child protection and child abuse prevention research.
Agency issues
In order to make the most of research opportunities and to develop an evidence base, a number of issues should be addressed, some of which are discussed below.
To achieve these steps, a number of issues need to be addressed.
Agency defensiveness
Parton (1985) described defensive practice as professional and agency self-protection, where professionals are not prepared to take risks in case management or decision making for fear that they may make an error and subsequently be vilified or sued. This may lead to an inflexibility with respect to investigation, or the minimisation of abusive concerns; either practice may in turn create harm for children and families (Satyamurti, 1981; Jones, 1991; Hallett & Birchall, 1992).
Certainly, the regular criticism of child protection departments in the media (and to be fair, often in research publications) has led, at times, to a highly defensive stance by workers and managements. This has translated into a reluctance to expose the Department to further criticism via research unless it (and the dissemination of findings) are able to be entirely controlled by the Department. It is becoming common for research to be contracted on the basis that the Department has ownership of the work and the veto rights over publication and dissemination of the results.
This is becoming a major issue for research bodies, who if they cannot publish the results of contract studies, are unable to justify the time and academic resources expended. For those relatively few studies that are being undertaken, or that are proposed, the ‘ownership’ issue is beginning to seriously impact on what can be undertaken and achieved, deterring some of the best qualified researchers from applying to undertake the contract and resulting in a body of work that is not seeing the light of day, in some cases even where the results are predominantly positive. Thus, the ability to learn from these studies — in research and practice terms — is substantially reduced.
There is a clear need for a change in the ways in which research is undertaken — whether it be independent or contract research undertaken for a department. Despite the risks of, once again, having practice errors, or flaws in service delivery systems being highlighted, departments require research to inform practice. Admittedly, this needs to be undertaken in a manner that does not merely ‘catastrophise’1 the system, but also in such a way that researchers are able to publish and disseminate their work back to the department and the wider child protection field. The issue of partnerships in research is discussed further below (see Researchers below).
Developing a research culture
There is a need to develop a research-friendly culture within child protection departments, such that research is valued as a means of better understanding child maltreatment and of determining the most effective professional responses. Staff should be educated to perceive rigorous investigation of practice as a valuable component of service development.
Practice wisdom
There is currently a general reliance on ‘practice wisdom’ as the basis for enhancing child protection practice. While practice wisdom is a vital component of an agency’s knowledge, at times there appears to be a feeling among sections of the child protection departments that ‘we think this is what’s happening - therefore it is’. That is, workers operate under the assumption that their views and experience, in isolation, provide an accurate, comprehensive assessment of practice that can adequately inform changes to service delivery. Concurrently, research is perceived to be a waste of valuable resources that will, at best, merely confirm practice wisdom. Thus, the costs of participating or undertaking research not offset by a significant increase in knowledge.
Hence, overcoming agency defensiveness and developing an agency culture where research and evidenced-based practice is valued should be key features of corporate planning.
Research/Training unit
One key feature of the shift to a research culture should be the development and empowerment of internal staff in research roles. That is, utilising staff experience, practice wisdom, awareness of agency culture in combination with research expertise that can facilitate the conduct (internal and external) and uptake of research. [The latter incorporates the development of worker training programs.]
Appropriately trained staff provide an in-house expertise or capacity to conduct rigorous research, a capacity that may avoid issues of confidentiality and research ownership in circumstances where departments do not wish to publicise results. The development of specialist in-house expertise is also vital as a means of ensuring that what research is done, particularly research involving external researchers, is based on comprehensive information about the department, agency culture and practices and agency/staff perceptions and of aspects of child protection practice.
The internal researchers are best placed to facilitate the research process while ensuring the ecological validity of the interpretations that are reached are informed and/or conducted by researchers with the best knowledge of child protection practice under in situ (real world) conditions.
That is, such staff can help to bridge the research-practice gap, such that external experts’ research and the resultant findings benefit from the interpretation and assistance of those ‘at the coalface’.
Importantly it ensures that academic and practice concepts are more accurately translated, ensuring a better understanding of practice and ensuring that research findings are produced in a form that encourages adoption or use by practitioners. This in turn should minimise the publication of ill-considered conclusions that merely result in the scapegoating of practitioners. I believe that researchers would produce more balanced findings if they were able to obtain practitioners’ interpretations of the findings.
Utilisation and dissemination
Once research is completed it is vital that the results are used and disseminated widely so as to inform practice. This can be facilitated in a number of ways.
First, agencies should encourage researchers (internal and external) to produce academic publications. This provides status for the research and also contributes to the dissemination of knowledge to the field from a source that is considered reputable (that is, books, refereed journals).
Second, it is vital that the research is translated for practice. Researchers have an obligation to assist internal ‘experts’ to develop material or training programs and thus to disseminate the research findings effectively through the child protection system (that is, to contribute to practice wisdom). Most researchers are pleased to do this, but are often not taken up on the offer. Overall, as part of any research undertaking I would recommend the development of a publication/dissemination plan (allowing the agency a period of confidential review of the results prior to publication or use).
In summary, it is recommended that child protection agencies should:
Researchers
It is generally acknowledged that child protection work is exceedingly difficult and that no matter decision is made, a child protection worker is liable to have that decision criticised by other professionals, the media and wider community. It should also be acknowledged that child protection work is typically not done under ideal situations — workers must contend with involuntary, uncooperative and/or hostile clients, a lack of resources, high workloads etc.. Thus, when investigating aspects of child protection practice it is often the case that the focus is on understanding practice and identifying elements of the system that may benefit from alternative approaches. That is, the research is often focused on identifying less than optimal performance and offering some solutions.
While researchers should not attempt to move away from documenting the realities of practice — ‘warts and all’ — it is apparent that a proportion of the research that is produced is perceived by workers as scapegoating and a biased representation of their experience. This may be the case in some instances, however most researchers are merely attempting to develop as accurate a picture as possible of the realities of child protection practice. As I noted above, I would argue that some of the so-called ‘bias’ or perceptions that the research is not an accurate depiction of practice, is based in a failure (or a lack of opportunity) to obtain the workers’ or agency’s perspective on research. That is, a failure to take into account agency cultural issues and the language of practice. This emphasises yet again, the importance of an ‘action research’ approach, where researchers have access to practitioners perspectives on aspects of practice and/or their interpretation of the research findings.
While the identification of areas requiring improvement is vital for researchers to ensure their findings are utilised and translated into practice, it is important that a problem focus or ‘catastrophising’ does not become the primary response. The way results are presented — balancing negative and positive aspects of practice, ensuring that the language used, and the understanding of what has occurred, is congruent with workers’ perspectives will improve the utility of the research. Although a researcher may maintain a different interpretation of the findings, giving workers a voice is important for ensuring that the results are perceived as balanced/ and objective. It facilitates their acceptance and use in child protection systems and ensures that the researcher is able to make the most of the collected data. The tension inherent in such a process however, is avoiding being too influenced by practitioner views such that important findings are minimised or inappropriately modified merely to avoid conflict.
Overall, to conduct research with the child protection departments requires entering into an agreement based on mutual trust. From the researcher’s perspective, the benefits of engaging in a collaboration relate to: enhanced access to the field; ensuring the most accurate depiction of child protection practice; and thus ensuring the work is a valuable addition to the research and practice fields. The only possible downside is that rather than being more balanced, the research process and interpretation of the results may become too influenced by agencies or practitioners such that important findings are minimised. I would argue that with care, the potential benefits of getting the support/advice and wisdom of the system’s ‘insiders’ far outweigh the risks.
In conclusion, it appears that there is currently a shift to make greater use of Australian child protection research. In creating further opportunities I would contend that three principles for conducting research - from the perspectives of both researcher and practitioner - should be:
References
Durlak, J.A. (1998), ‘Common risk and protective factors in successful prevention programs’, American Journal of Orthopsychiatry, vol.68, no.4, pp.512-520.
Hallett, C. & Birchall, E. (1992), Coordination and child protection: A review of the literature. HMSO, Edinburgh.
Heatherington, T. (1998), ‘Child protection reform in South Australia: initial evaluation’, Child Abuse Prevention, vol.6 no.2 Spring, pp.7-10.
Jones, D.P.H. (1991), ‘Professional and clinical challenges to protection of children’, Child Abuse & Neglect, vol. 15, supp.1, pp.57-66.
Melton, G. B. & Flood, M. F. (1994), ‘Research policy and child maltreatment: developing the scientific foundation for effective protection of children’, Child Abuse and Neglect, Vol.18, suppl. 1, pp.1—28.
Parton, N (1985), The politics of child abuse, Macmillan, London.
Parton, N. (1999), ‘New Directions in Western Australia: a policy initiative to re-focus child welfare practice, in 7th Australasian Conference on Child Abuse and Neglect, Perth, WA, October 1999: Conference proceedings, vol. 2, p811-816, Promaco Conventions, Canning Bridge, WA.
Satyamurti, C. (1981), Occupational survival: The case of the local authority social worker, Basil Blackford, Oxford.
Tomison, A.M. (1996), Intergenerational Transmission of Maltreatment, National Child Protection Clearinghouse Issues Paper no. 6, Australian Institute of Family Studies, Melbourne.
Tomison, A.M. (2000), Evaluating Child Abuse Prevention Programs, NCPCH Issues Paper no.12., Australian Institute of Family Studies, Melbourne.
Notes
1. A term I picked up while working with a child protection departmental manager. She defined the term, in child protection vernacular, as presenting the most negative or pessimistic perspective when describing the child protection system.
Return to Papers and presentations
AIFS Home |
Email
queries to
webmaster@aifs.gov.au
|
Copyright, privacy,
disclaimer
|
Site map Australian Institute of Family Studies, Level 20, 485 La Trobe Street, Melbourne Vic 3000, Australia. Tel: (03) 9214 7888. Fax: (03) 9214 7839. URL: http://www.aifs.gov.au/ |