This article first appeared in Funding Insight on 19 May 2016 and is reproduced with kind permission of Research Professional. For more articles like this, visit www.researchprofessional.com.
Researchers, research commissioners, and research funders all struggle with identifying good quality arts-based research. ‘I know it when I see it’ just doesn’t pass muster. Fortunately, Sarah J Tracy of Arizona State University has developed a helpful set of criteria that are now being used extensively to assess the quality of qualitative research, including arts-based and qualitative mixed-methods research.
Tracy’s conceptualisation includes eight criteria: worthy topic, rich rigour, sincerity, credibility, resonance, significant contribution, ethics, and meaningful coherence. Let’s look at each of those in a bit more detail.
A worthy topic is likely to be significant, meaningful, interesting, revealing, relevant, and timely. Such a topic may arise from contemporary social or personal phenomena, or from disciplinary priorities.
Rich rigour involves care and attention, particularly to sampling, data collection, and data analysis. It is the antithesis of the ‘quick and dirty’ research project, requiring diligence on the part of the researcher and leaving no room for short-cuts.
Sincerity involves honesty and transparency. Reflexivity is the key route to honesty, requiring researchers to interrogate and display their own impact on the research they conduct. Transparency focuses on the research process, and entails researchers disclosing their methods and decisions, the challenges they faced, any unexpected events that affected the research, and so on. It also involves crediting all those who have helped the researcher, such as funders, participants, or colleagues.
Credibility is a more complex criterion which, when achieved, produces research that can be perceived as trustworthy and on which people are willing to base decisions. Tracy suggests that there are four dimensions to achieving credibility: thick description, triangulation/crystallization, multiple voices, and participant input beyond data provision. Thick description means lots of detail and illustration to elucidate meanings which are clearly located in terms of theoretical, cultural, geographic, temporal, and other such location markers. Triangulation and crystallisation are both terms that refer to the use of multiplicity within research, such as through using multiple researchers, theories, methods, and/or data sources. The point of multiplicity is to consider the research question in a variety of ways, to enable the exploration of different facets of that question and thereby create deeper understanding. The use of multiple voices, particularly in research reporting, enables researchers more accurately to reflect the complexity of the research situation. Participant input beyond data provision provides opportunities for verification and elaboration of findings, and helps to ensure that research outputs are understandable and implementable.
Although all eight criteria are potentially relevant to arts-based research, resonance is perhaps the most directly relevant. It refers to the ability of research to have an emotional impact on its audiences or readers. Resonance has three aspects: aesthetic merit, generalisability, and transferability. Aesthetic merit means that style counts alongside, and works with, content, such that research is presented in a beautiful, evocative, artistic and accessible way. Generalisability refers to the potential for research to be valuable in a range of contexts, settings, or circumstances. Transferability is when an individual reader or audience member can take ideas from the research and apply them to their own situation.
Research can contribute to knowledge, policy, and/or practice, and will make a significant contribution if it extends knowledge or improves policy or practice. Research may also make a significant contribution to the development of methodology; there is a lot of scope for this with arts-based methods.
Several of the other criteria touch on ethical aspects of research. For example, many researchers would argue that reflexivity is an ethical necessity. However, ethics in research is so important that it also requires a criterion of its own. Tracy’s conceptualisation of ethics for research evaluation involves procedural, situational, relational, and exiting ethics. Procedural ethics refers to the system of research governance – or, for those whose research is not subject to formal ethical approval, the considerations therein such as participant welfare and data storage. Situational ethics requires consideration of the specific context for the research and how that might or should affect ethical decisions. Relational ethics involve treating others well during the research process: offering respect, extending compassion, keeping promises, and so on. And exiting ethics cover the ways in which researchers present and share findings, as well as aftercare for participants and others involved in the research.
Research that has meaningful coherence effectively does what it sets out to do. It will tell a clear story. That story may include paradox and contradiction, mess and disturbance. Nevertheless, it will bring together theory, literature, data and analysis in an interconnected and comprehensible way.
These criteria are not an unarguable rubric to which every qualitative researcher must adhere. Indeed there are times when they will conflict in practice. For example, you may have a delightfully resonant vignette, but be unable to use it because it would identify the participant concerned; participants may not be willing or able to be involved beyond data provision; and all the diligence in the world can’t guarantee a significant contribution. So, as always, researchers need to exercise their powers of thought, creativity, and improvisation in the service of good quality research, and use the criteria flexibly, as guidelines rather than rules. However, what these criteria do offer is a very helpful framework for assessing the likely quality of research at the design stage, and the actual quality of research on completion.
Next week I will post a case study demonstrating how these criteria can be used.