Systematic synthesis of qualitative research,Michael Saini and Aron Schlonsky

Reviewed by: Martin Webber, University of York, UK

When I first saw this book, I wondered whether the systematic synthesis of qualitative research was an oxymoron. By the time I had finished reading it, I had been convinced that it wasn’t.

The synthesis of quantitative findings in meta-analysis is commonplace in the canon of evidence-based practice. It is generally assumed that quantitative data are value-neutral and unproblematic in synthesis. Well-developed methods of meta-analysis provide an analytical framework which permit reliable conclusions to be drawn about specific types of studies. Typically used to evaluate the effectiveness of interventions, meta-analysis requires the same outcome measures to be applied to the same population in studies evaluating the same intervention, irrespective of its context.

In reality, though, systematic reviews of quantitative data are rarely capable of delivering methodologically pure meta-analysis. This is particularly the case in social work research where contextual variables disproportionately impact upon the homogeneity of interventions and their evaluation. If you were to undertake a systematic review of systematic reviews of quantitative data, it is likely that you will find many which did not employ meta-analytic techniques. The remainder are likely to have used a variety of synthesis techniques which collated findings and drew conclusions about the similarities or differences between studies. Arguably, this is not so different from the synthesis of qualitative data.

There are some fundamental objections to the synthesis of qualitative research findings. Data obtained via different epistemological and philosophical frameworks may be incompatible. For example, data produced by a researcher understanding the meaning which people ascribe to their experiences (‘interpretive’) can be fundamentally at odds with that which has been produced within a ‘post-positivism’ framework which seeks to make general laws about social behaviour.

Saini and Shlonksy take a much more pragmatic approach and argue that the problems of inequality and human suffering are too great to be stymied by the irresolvability of debates between interpretivism and post-positivism. The rich insights available within qualitative studies can, once synthesised, provide important insights to inform social work policy and practice, they argue.

In this contribution to the Oxford University Press Pocket Book series (which is more a reference to its diminutive size rather than an inferred dilution of its academic content), Saini and Shlonksy capably make the case for the systematic synthesis of qualitative data. They use its early chapters to present an overview of synthesis methods which is as comprehensive in its breadth as insightful in its depth. By providing examples and the rationale for the use of different methods, the authors introduce readers to the broader context of systematic reviews and the location of the synthesis of qualitative research in this panoply of research methods.

The second part of the book focuses on the steps required to undertake a systematic synthesis of qualitative data. Not just a ‘how-to’ guide, this section takes a critical approach to searching, screening and appraising studies relevant to the research question being investigated. Suitable for postgraduate social work students (both taught and PhD candidates), and social work researchers, this part of the book addresses common questions students and new researchers ask of their supervisors. Again, the authors not only provide questions to guide the critical appraisal process, they critically engage with their own 25-point quality appraisal form (the Qualitative Research Quality Checklist (Saini, 2011)). This is particularly useful as they illustrate why these questions are important in addition to providing examples of how they can be used in practice.

A limitation of this section of the book, though, is the insufficient detail on how to summarise a critical appraisal of a study. This is unfortunate as new and experienced researchers often struggle to sensibly collate their thoughts about critical appraisals of studies. It is often difficult to reach an overall conclusion about the quality of a study, particularly when a scoring system is not used. Unfortunately, the authors do not help much here, instead critiquing the conclusions of a paper (Bell and Wilson, 2006) to illustrate the importance of ensuring conclusions accurately portray the quality of a study. Arguably, it would be more useful to demonstrate how to move from answers to 25 critical appraisal questions to a brief summary of the study findings.

Running throughout the book is a worked example of the qualitative synthesis of family group decision making for children at risk of abuse and neglect (Shlonsky et al., 2008). This example comes into its own in chapter 8 where the data synthesis is presented. However, the process of synthesising data from multiple studies is largely absent. This is arguably a serious omission as it is often the most sought-after process which new researchers are seeking to learn. It is implied that this is covered elsewhere, but in the short paragraph on page 150 no references are provided. This pocket guide could be significantly enhanced with a fully worked example of the synthesis of qualitative data rather than the truncated example provided in table 8.2.

Qualitative synthesis can significantly enhance the evidence base for social work practice by providing a more comprehensive understanding of the range of client and professional experiences than individual studies permit. The synthesis of qualitative studies is yet to be fully embraced by the Cochrane or Campbell Collaborations, but this book helps to raise its status in the canon of research methods.

Social workers value insights from qualitative studies as they can broaden the field of vision beyond individual caseloads and efficiently bring multiple perspectives to bear on a single topic. The synthesis of qualitative data takes this a step further and broadens the scope to encompass multiple studies all focusing on a single issue or research question. This not only assists in the process of achieving theoretical saturation on the topic, it can provide practitioners with an enhanced understanding of the research topic taking them a step closer towards the final solution to their problem.