A critique of concept analysis

Authors


I argue that there is no place for concept analysis in the scholarship of modern nursing. Although it may have played a role in the intellectual history of our profession, concept analysis is methodologically weak and philosophically dubious. Furthermore, there is no evidence that it has made a contribution to patient care.

Concept analysis looked promising in the turbulent days of the 1980s. Nursing practice was being transformed by the introduction of the nursing process and primary nursing. Many changes were driven by an infusion of intellectual energy from newly formed university departments of nursing. Nursing research began to flourish and nursing theories and models were discussed passionately. We were confident that – within a decade or two – nursing would have its unique knowledge base, the ultimate mark of a true profession. Our knowledge base would be built ‘from the ground up’. Concept analysis would help us to describe the key elements of nursing practice and we would explore the relationships between concepts through programmes of empirical research. In time, we would build theories and conceptual models and these would feedback to improve practice in a virtuous circle of professional development.

Some elements of this vision have, indeed, flourished. In many countries, nursing education is embedded in universities. Nursing research is vibrant, and nurses have played a leading part in the development of evidence-based practice. Other aspects of the prospectus have ‘withered on the vine’. Few now discuss nursing models and theories, and the notion that a discipline can lay claim to a unique knowledge base has been thoroughly undermined by the globalization of knowledge. All knowledge belongs, and is available, to everyone. And yet in nursing alone, and despite the evidence, the idea persists that there is merit in concept analysis.

Concept analysis is ‘cottage industry’ for nursing journals. When I recently searched CINAHL for articles with the term concept analysis in the title, I found over 800 hits. Collectively, these papers addressed approximately 300 separate concepts. To scan this list is to hear the faint ringing of alarm bells. How can it be imagined that a worldview as rich and diverse as ‘feminism’ can be pinned down through concept analytical procedures? And at the other end of the scale, what could concept analysis possibly add to our understanding of ‘waiting’? The subjects of concept analysis veer between the pretentious – surely, it would take at least one large book adequately to address a concept like ‘power’ – and the trivial.

I then searched the databases of our cognate professions including medicine, clinical psychology and social science. Having stripped out the ‘hits’ that were published in nursing journals, I was unable to identify a single case of a medic, psychologist or social scientist undertaking concept analysis. There is no evidence that a similar programme of work exists in other disciplines. This could be evidence of the uniqueness of nursing scholarship, but it seems unlikely that scholars from these disciplines would neglect a fruitful approach. After all, nurses who write on research methodology are often cited in the wider literature. It is more likely that concept analysis makes sense only as part of a ‘closed circuit’ of nursing knowledge that is of no interest to the wider intellectual and professional world.

Partly, the problem lies with the assumption that concepts can be clarified in principle, and partly, it lies in the detail of the methods that are used to analyse nursing concepts. The difficulty of analysing concepts in principle can be illustrated by taking the concept ‘chair’ as an example. The concept of ‘chair’ is a mental category that enables us to think and talk about chairs in general, but it is impossible further to ‘clarify’ the concept of ‘chair’ by ‘analysing’ it, if the goal of that process is to produce a list of ‘defining attributes’ as recommended by Walker and Avant (1995). This is because there is no list of necessary attributes that characterizes chairs. A chair is not simply something to sit on, for we also sit on sofas, stools and the floor. Chairs do not all have legs – some have wheels. Indeed, in a university, the word chair may not refer to an item of furniture at all. Wittgenstein (1958) said that the meaning of a word is its use, and paradoxically, we find the concept ‘chair’ useful even though we cannot list its definitive attributes.

This brings us to a second problem. As we have seen, a key part of concept analysis methodology is to determine the defining attributes of the concept in question. When I examined a subset of ten published concept analyses, I found that the source of these defining attributes is invariably ‘the literature’. There is now a scholarly consensus about the standards for reporting literature reviews and particular importance is attached to transparency about inclusion and exclusion criteria, for these are the cornerstone of quality appraisal. Only two of my subset of papers disclosed their inclusion and exclusion criteria, and none described a data extraction process. It is, therefore, tempting to characterize many concept analyses as low-grade literature reviews, for most of them would certainly not meet the criteria for publication as systematic reviews.

These criticisms concern the internal logic of concept analysis as a method. Further criticisms can be made of the external impact – or lack of it – on nursing scholarship or practice. Although as a profession we continue to generate concept analyses, I have been unable to find evidence that they feed in to coherent and sustainable programmes of research. The ultimate test of nursing scholarship is its impact on nursing practice. Having scoured the websites of organizations whose mission is to sift the evidence and bring the best to bear on practice by producing systematic reviews and guidelines (organizations such as the Centre for Reviews and Dissemination, and the Cochrane Collaboration), I found no evidence of even the potential to influence clinical practice.

The fact that concept analysis is unique to the nursing literature whilst being absent from that of any related discipline suggests that it is an intellectual ‘dead-end’. It is arguable that the dominant methods of concept analysis are based on a mistaken understanding of the nature of concepts. The nursing profession should think again about concept analysis. The approach may have looked like a good idea in the 1980s, but it has no contribution to make to the scholarship of contemporary practice.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest

No conflict of interest has been declared by the authors.

Author contributions

The author agreed on the final version and meets at least one of the following criteria [recommended by the ICMJE (http://www.icmje.org/ethical_1author.html)]:

  • substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data;
  • drafting the article or revising it critically for important intellectual content.

Ancillary