The controversy that has surrounded the value of quantitative research methods as opposed to qualitative approaches as a means to increasing the knowledge and understanding of human behaviour in health and illness, has been contested by nurse scholars for several decades. This paper continues debate around this issue and provides a critique of the problems associated with these competing paradigms. It challenges the convention that all nursing research must be objective and value free in order to be scientific, and provides an overview of the processes that should be considered by researchers utilising qualitative methods of inquiry.
Key Words: Research, Qualitative, Quantitative, Credibility, Generalisability, Reliability
Research and theory are critical elements that have a direct impact upon the development of professional nursing practice. Pullen (2000) describes the clinical practice of nursing as dealing with "the subjective condition of individual patients" (p. 124). The value of qualitative research in nursing is that it can help to explain the complexity and meaning of human behaviour, by addressing questions such as why some patients require more postoperative pain relief than others and why nurses respond differently to verbal aggression. Thus, nursing research often needs to capture the individual's interpretation of an experience and the meaning attributed to that experience in order to interpret and synthesise data that are not always responsive to quantitative research methods (Munhall, 2001; Pullen, 2000).
Being so closely associated with medicine, the dominant paradigm for past nursing research and theory development has traditionally been that of positivism, imported directly from modern Western medicine (Pullen, 2000). A result of this was a lack of research tradition unique to nursing (Donaldson & Crowley, 1977; Gorenberg, 1983; Morse, 1994; Munhall, 2001). Even in the mid 1960's, when nursing scholars looked to develop research that would enable them to investigate the social world of nursing practice (Street, 1990), they continued to embrace the more positivistic approaches within other disciplines of the social sciences as the normative paradigm for conducting nursing research (Carr, 1994; Gushing, 1994; Playle, 1995). They considered its scientific method to be superior to any other and therefore, the most legitimate approach to use. Accordingly, the blind acceptance of research methods associated with the quantitative tradition of positivism has had a significant impact upon scholarship in nursing research (Morse, 1994; Munhall, 2001; Thompson, 1985). Although this has been severely limiting in some respects, the adoption of these scientific methods from other disciplines has at least assisted nursing to identify, establish and develop itself as a science (Munhall, 2001; Parse, Coyne, & Smith, 1985).
During the past two decades nurse researchers have begun to use a far wider range of methods such as ethnography, phenomenology and symbolic interactionism, and in so doing have become embroiled in an ongoing and often acrimonious dispute between advocates of quantitative or qualitative methods (Dootson, 1995). Thompson (1985, cited in Street, 1990, p. 5) has argued that many nurse scholars continue to remain unequivocal in their support of quantitative research, because it is perceived as the 'proper' scientific paradigm in which nursing research should be situated. Positivists often perceive qualitative approaches as unscientific, soft scholarship, exploratory, overly subjective and biased, indeed, as an assault on the scientific method that undermines the "crowning achievements of Western civilisation" (Denzin & Lincoln, 1994, p. 4). The main challenge to this dominating stance is a critique of the way nursing research has clung to, and become preoccupied with, semantic analysis of terms and concepts, deducing and testing hypotheses through quantitative techniques that depend on the operationalisation of all variables. Whilst this approach undoubtedly contributes towards the knowledge base of nursing practice, in some ways it fails to provide a holistic view of the complexity of human behaviour, human experience, or the health-illness continuum, let alone acknowledging the relationships between them. Schön (1987) also tends to agree that some complex areas of investigation are not necessarily amenable to the 'technical rationality' of positivist research. In short, the adequacy of quantitative methods in terms of the interpretation of data relating to human subjects is increasingly being questioned. But it is difficult to turn the tide, because as Leininger (1985, p. 4) states, the "norms of nursing are to follow the 'scientific' quantitative method unquestioningly."
Despite the dominant orientation to positivism with its allegiance to a putative objective reality in past nursing research, there has been a growing realisation that qualitative research can provide fresh insights, particularly through the study of people's unique lived experiences (Gorenberg, 1983). Leading nurse scholars (Benner, 1984; Leininger, 1985; Munhall, 2001; Oiler, 1982; Omery, 1983; Stern, 1989) have mounted a challenge to the normative paradigm of positivism, advocating a shift towards a qualitative research perspective as a means of investigating the nature of human reality, particularly in terms of understanding the subjectivity of phenomena.
According to Omery (1983), the use of traditional scientific methods has served nursing well in the past, but concedes that the inherent nature of positivist research, which reduces humans to 'quantitative units', has prompted nurse researchers to consider alternative methods of studying the human dimension of nursing and its interventions. Parker (1987) also contends that past nursing practice has been constrained through its organisation around an 'old science' world view, based on knowledge that is compartmentalised, fragmented, controlled by experts and the domain of other disciplines.
The adequacy of utilising quantitative research methods and its language to investigate nursing practice issues holistically is increasingly being questioned (Munhall, 2001). There is growing recognition among nurse researchers that not all research interests can be accommodated by a quantitative perspective, therefore, qualitative approaches can and do offer a legitimate means of studying the nature and essence of phenomena, especially from the perspective of the individual's own lived experiences. However, Munhall (2001) acknowledges that nurse scholars are increasingly embracing the merits of both traditions as this provides a much broader repertoire of available research methods.
Leininger's (1985, p. 3) assertion that for over thirty years, nurse researchers have been "strongly socialised" and have "blindly accepted" that quantitative methods have greater validity, and, therefore, legitimacy in nursing research may not be as true today as it was when the statement originated. The emphasis upon quantitative research methods and statistical techniques continues to be viewed as the 'proper' scientific method and is often the only paradigm taught in undergraduate nursing curricula. This is despite the increasing number of published qualitative research studies that have been pushing out the frontiers of nursing knowledge and understanding of professional nursing practice. Leininger (1985) states that:
It has been difficult to 'turn the tide' of thinking because so few nurses have critically examined where qualitative and quantitative methods lead to in knowledge discovery and verification. (p. 4)
Consequently, nurse educators unquestioningly adopt the normative stance of accepting, teaching and supporting the methodological correctness of quantitative research. This legacy of orthodoxy also appears to have pervaded graduate studies and may stifle interpretative inquiry when graduates are pressured by faculty to employ only the methods in which they were themselves trained.
One of the key issues facing the credibility of contemporary nursing is its relationship with the discipline of medicine. In its endeavour to compete with medical research, nursing has adopted traditional scientific approaches without fully questioning the applicability of those methods in the discovery of nursing practice.
Oiler (1982) has been critical of the way that qualitative research in nursing is portrayed as the 'handmaiden' to scientific methods. One of the reasons for this is that past health research has been dominated by medical researchers using the traditional scientific methods of positivism. In other words, medical researchers have not been well acquainted with, and have had difficulty in accepting the research methodologies of the social sciences in which the generation of hypotheses replaces the testing of hypotheses, where explication is more important than measurement, and where there is emphasis upon understanding rather than generalisation. Swanson-Kauffman (1986) insists that:
Nurses must explore how best to study their own concerns. We need to allow ourselves the comfort to recognise that those values and methods that hold up well elsewhere should not dictate how we choose to go about answering questions that arise from our science of humanistic practice. (p. 59)
This view articulates with Parker's (1987) interpretation of Capra's (1982) 'new world' perspective, which is based on holism and the interrelatedness and independence of all phenomena.
As the discipline of nursing matures, it needs to move beyond the epistemological arguments that pitch qualitative against quantitative approaches. Instead, the benefits and disadvantages of each approach should be considered when selecting a methodology that will be the most appropriate for investigating the research problem. It is interesting to note that although this may seem a very obviously useful exercise, whilst it is standard practice for a qualitative researcher to explore and justify appropriateness of the methods they choose to use, very few quantitative researchers do the same. One hopes that this will not prove to be an enduring difference between these paradigms.
Quantitative research derives from the natural sciences and is based on the premise that phenomena can be explained by objective and factual measures that help to keep data free from researcher bias. It makes epistemological assumptions that reality is unitary and objective and can only be 'discovered' through transcending the perspective of the individual, and, therefore, that phenomena must be explained through the analysis of data obtained by objective forms of measurement. In that it attempts to measure the cause-effect relationships between discrete and pre-selected variables by controlling and manipulating data through experimental or quasi-experimental techniques (Leininger, 1985), the perceived knowledge outcome of quantitative inquiry is singular truth.
It is with the above epistemological assumptions that qualitative research takes issue (Wildemuth, 1993). Interpretative approaches such as ethnography, phenomenology and symbolic interactionism attempt to understand the nature of social reality through people's narrated accounts of their subjectively constructed processes and meanings, as opposed to the measurement of quantity, frequency and distribution across a given population. Qualitative research, therefore, is an approach that yields findings by means other than quantifiable statistical procedures (Strauss & Corbin, 1990). That is, it tends to generate verbal data as opposed to numerical (Knafl & Howard, 1984; Polgar & Thomas, 1995) and reflects an epistemology where phenomena have "multiple, socially defined realities" (McMurray, 1994, p. 18). The qualitative tradition assumes that the complexity of phenomena necessitates researcher's utilising methods that will bring them closer to information sources, interacting with participants, interrogating data, drawing upon their own experiences, following-up leads and checking out hunches. These data are usually generated in naturalistic settings and the qualitative researcher avoids controlling or manipulating situations, events or actions.
The manner in which people make sense of their experiences is referred to as subjectivity by quantitative researchers, who recognise that people construct meanings in relation to their environment and previous experiences. Munhall (2001, p. 73) postulates that the objective of quantitative research is to "disclose subjectivity" through exploring and collecting data that describes the experience being researched.
Qualitative research appears to offer a sound methodological framework for developing an understanding of the implicit subjectivity that occurs within professional nursing practice. The process of documenting and interpreting information provides a fundamental basis for describing and interpreting the contextual variability of phenomena that surrounds the lives of human subjects. The utilisation of qualitative research methods provides access to the lived reality of individuals, facilitating the exploration of people's internal construction of their personal worldview.
Such methods also assist in generating data that is rich in the subjectivity of actions, interactions, emotions, culture, symbols and rituals. Interaction between researcher and participant is recognised as a key component of data generation and valued as such, because it is a means of getting close to the experiences of participants so that phenomena can be viewed from their own perspective. Mies and Vandana (1997, p. 34) refers to this as 'conscious partiality', which brings the researcher closer to the actual reality of phenomena.
A qualitative framework also provides a foundation for the production of data through a variety of sources and means. It offers an interpretative flexibility that is both reflexive and reactive to the concepts and theories, which emerge from data. This reflects an epistemological pluralism, as it involves moving away from the constraints of the value-free orientation of positivism, which upholds objectivism through ignoring many differences. Instead, the researcher selects and applies methods that are appropriate to the research question being addressed (Wildemuth, 1993). This leads to a creativity that is guided by the epistemological approach rather than being constrained by it. According to Rolfe (1995) the rigidity of thinking by researchers who avoid stepping outside of the rules, is unlikely to produce the necessary leap into creative discovery. An epistemological pluralism should facilitate the researcher to gain valuable insights to complex phenomena, such as those associated with the varied dimensions of human behaviour occurring in naturalistic settings.
Essentially, the aim of qualitative research in the discipline of nursing is to increase, as fully as possible, practitioners' understanding of the phenomena that occur within the varied context of nursing practice. It should assist in the discovery of unknown phenomena by helping to identify key features and characteristics that may previously have been obscure or unexplained. Such phenomena may be recognised through observed behaviour evident within interpersonal interactions or recalled from previous life experiences. It is through the study of such phenomena that qualitative research can help us to discover, interpret and understand our world. Munhall (2001) states that:
Qualitative research involves broadly stated questions about human experiences and realities, studied through sustained contact with persons in their natural environments, and producing rich, descriptive data that help us to understand those person's experiences. (p. 68)
Its focus, therefore, is to identify, interpret and provide meaning, which helps to explain what makes phenomena what they are. Knowing these dimensions and the contexts, in which they occur, can assist us to learn more about the extent, diversity and variability of professional nursing practice. It provides clues about the different aspects of life events and how individuals may react towards them. Leininger (1985, in citing Pelto and Spradley) confirms that qualitative research is the most appropriate approach to "discover essences, feelings, attributes, values, meanings, characteristics, and teleological or philosophical aspects of certain individuals or group lifeways" (pp. 6-7).
The connotations and applicability of the term 'credibility' is more acceptable to qualitative researchers than that of validity (Kincheloe & McLaren, 1994), which tends to reflect positivist values of 'rigour' and 'objectivism'. Qualitative studies have a clear need to address issues of credibility (McMurray, 1994; Morse & Field, 1995). Janesick (1994), insists that qualitative researchers have been patiently responding to the issue of credibility, citing Patton (1990, p. 216), who maintains that qualitative studies should address three fundamental questions when writing-up the narrative:
Each of the above questions suggest that the proper application of qualitative research methods requires the researcher to ensure that methods used to generate data are fully described and documented in the study. These documented descriptions "allow the reader to exercise joint responsibility with the researcher in judging the evidence on which claims are based." (Murphy, Dingwall, Greatbatch, Parker, & Watson, 1998). This should leave an 'auditable' (Morse & Field, 1995; Sandelowski, 1986) trail that can be followed by others so that the reader or other researchers can trace the decision making processes which led to data production and analysis. The qualitative researcher should also acknowledge (or bracket) their own expert frames of reference gained through previous knowledge or experience with the subject area, by documenting the insights or biases that may otherwise permeate and influence the interpretation of data. The assumptions that derive from data and which explain the construction of theory and related meanings must also be fully described. Morse and Field (1995, p. 118) cite Lincoln and Guba who, in their model for trustworthiness in qualitative research, termed this "neutrality or confirmability." Further issues of rigour described in this model are those of "applicability" and "consistency", the definitions of which are encapsulated in the term genalisability.
The term generalisability refers to the extent to which findings of a research study can be applied to other groups, situations or settings. Kincheloe and McLaren (1994) refer to this as a measure of external reliability. Although qualitative research has tended to attract criticism levelled at its lack of applicability to situations outside of the setting in which studies have been conducted, Janesick (1994) warns that "for too long we have allowed psychometrics to rule our research and thus to decontextualise individuals" (p. 217). In other words, a preoccupation with employing positivist procedures for defending methods undermines what is arguably the cornerstone of qualitative research - the interpretation of different people's behaviour and situational events.
In view of such possibilities, others such as Wolcott (1990), whilst critical of traditional ways of thinking about generalisability, acknowledges that it should not be dismissed completely when it is still far from clear how the notion will be developed.
The danger of loss of difference appears to be the paramount concern of most qualitative methodologists. Janesick (1994, p. 216) challenges the notion that the "trinity of validity, generalisability and reliability", terms usually synonymous with the quantitative paradigm, should be strictly applied to all research. She argues that traditional thinking about generalisability is not appropriate to qualitative studies, where the researcher is more interested in questioning and understanding the "meaning and interpretation in individual cases" (p. 217). Guba and Lincoln (1994) put forward the argument that:
Precise quantitative approaches that focus on selected subsets of variables necessarily 'strip' from consideration, through appropriate controls or randomisation, other variables that exist in the context that might, if allowed to exert their effects, greatly alter findings. (p.106)
This is defined as context stripping (Guba & Lincoln, 1994, p. 106), and they point to the contradiction that the overly rigid design of some quantitative studies can actually detract from their applicability (and generalisability), because outcomes can only be applied in the same contextual conditions, such as laboratories. The way out of such difficulties appears to be through a lateral paradigm shift: to recognise the practice value of generalisations, but to allow people to generate them for themselves through their own perceptions of data, rather than for the researcher to do so through statistical control of variables thus leading to the development of what Stake and Trumbell (cited in Denzin & Lincoln, 1994, p. 240) terms 'naturalistic generalisation'.
Qualitative research can attain an appropriate level of external reliability by documenting the succession of moves through the stages of data production, analysis and interpretation. This can be achieved by explaining the methodological framework and the range of strategies that have been used within the study. The rationale for the way in which participants were selected to take part should also be described, as should the researcher's role and their perceived relationship to those participants. It will be necessary to document analytic constructs and meanings, which derive from data, alongside the methodological approach and procedures that were used for producing data. This would include providing descriptions of phenomena with appropriate narrative of the social context in which they occurred, particularly in terms of persons, places and events. Theoretical propositions also need to be fully explained in terms of how constructs have been formed through detailed analytic procedures.
Through the overall structure, approach and the methods described in qualitative research, it should be possible to replicate the study, though not in the positivist sense of the term. Perhaps the most that can be said is that, given sufficiently similar conditions and comparable population sample, data production procedures and method of analysis, another researcher should generate theoretical constructs that share distinct similarities. An exact replication of theory would not be expected because "theory is derived from the researcher's best analysis which includes the researcher's skill, creativity, time, resources, and analytic ability" (Chenitz & Swanson, 1986, p. 13). In other words, with even the most rigorous protocols, the variability of the researcher's own resourcefulness, ability and individuality will influence the way in which theory is generated. In terms of the development of theory, therefore, the test for reliability cannot be a specific reproduction, nor even a similar outcome, but rather by the extent to which differences and contradictions can be explained by a combination of findings of different studies. The other major test for the reliability of a theory is, of course, prediction in practice: how well does this understanding of phenomena enable us to predict other occurrences of the same phenomena.
It has been argued that notions of generalisability, validity, and reliability cannot be strictly applied to the qualitative paradigm, particularly where the researcher is more interested in questioning and understanding the meaning and interpretation of phenomena. However, qualitative researchers must be prepared to document and justify their methodological approach, and to describe, in detail, the critical processes and procedures that have helped them to construct, shape and connect meanings associated with those phenomena.
Benner, P. (1984). From novice to expert: Excellence and power in clinical nursing practice. Menlo Park, CA: Addison-Wesley Publishing Company.
Capra, F. (1982). The turning point: Science, society, and the rising culture. New York: Simon and Schuster.
Carr, L.T. (1994). The strengths and weaknesses of quantitative and qualitative research: What method for nursing? Journal of Advanced Nursing, 20, 716-721.
Chenitz, W. C., & Swanson, J. M. (1986). From practice to grounded theory: Qualitative research in nursing. Menlo Park, CA: Addsion-Wesley.
Denzin, N. K., & Lincoln, Y. S. (Eds.). (1994). Handbook of qualitative research. Thousand Oaks, CA: Sage.
Donaldson, S. K., & Crowley, D. M. (1978). The discipline of nursing. Nursing Outlook, 26, 113-20.
Dootson, S. (1995). An in-depth study of triangulation. Journal of Advanced Nursing, 22, 183-187.
Gorenberg, B. E. (1983). The research tradition of nursing: an emerging issue. Nursing Research, 32, 347-349.
Guba, E., & Lincoln, Y. (1994). Competing paradigms in qualitative research. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 105-118). Thousand Oaks, CA: Sage.
Gushing A. (1994). Historical and epistemological perspectives on research and nursing. Journal of Advanced Nursing, 20, 406-411.
Janesick, V. J. (1994). The dance of qualitative research design: Metaphor, methodolatry, and meaning. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 209-235). Thousand Oaks, CA: Sage.
Kincheloe, J. L., & McLaren, P. L. (1994). Rethinking critical theory and qualitative research. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 138-157). Thousand Oaks, CA: Sage.
Knaffl, K. A., & Howard, M. J. (1984). Interpreting and reporting qualitative research. Research in Nursing and Health, 7, 17-24.
Leininger, M. M. (Ed.). (1985). Qualitative research methods in nursing. Orlando, FL: Grune & Stratton.
McMurray, A. (1994). Researching rural health: The qualitative approach. The Australian Journal of Rural Health, 2(4), 17-24.
Mies, M. (1983). Towards a methodology for feminist research. In G. Bowles & R. D. Klein (Eds.), Theories of women's studies (pp. 25-46). London: Routledge & Kegan Paul.
Mies, M., & Vandana, S. (1997). Ecofeminism.. Melbourne: Spinifex Press.
Morse, J. (1994). Critical issues in qualitative research methods. Thousand Oaks, CA: Sage.
Morse, J. M., & Field, P. A.. (1995). Qualitative research methods for health professionals (2nd ed.). Thousand Oaks, CA: Sage.
Munhall, P. L. (2001). Nursing research: A qualitative perspective. Sudbury, MA: Jones & Barlett.
Murphy, E., Dingwall, R., Greatbatch, D., Parker, S., & Watson, P. (1998). Qualitative research methods in health technology assessment: A review of the literature. Health Technology Assessment, 2(16). Retrieved July 4, 2002, from http://www.hta.nhsweb.nhs.uk
Oiler, C. J. (1982). The phenomenological approach in nursing research. Nursing Research, 31(3), 178-181.
Omery, A. (1983). Phenomenology: A method for nursing research. Advances in Nursing Science, 5(2), 49-63.
Parker, J. M. (1987). The challenge of change: Nursing and the impact of technology. Paper presented at the Northern Territory State Committee College of Nursing, Alice Springs.
Parse, R. R., Coyne, A. B., & Smith, M. (1985). Nursing research: Qualitative methods. Bowie, MD: Prentice Hall.
Patton, M. Q. (1990). Qualitative evaluation and research methods (2nd ed.). Newbury Park, CA: Sage.
Playle, J. (1995). Humanism and positivism in nursing: Contradictions and conflicts. Journal of Advanced Nursing, 22, 979-984.
Polgar, S., & Thomas, S. (1995). Introduction to research in the health sciences (3rd ed.). Melbourne: Churchill-Livingstone.
Pullen, M. (2000). Understanding research paradigms in nursing research. ADF Health, 1, 124-128.
Rolfe, G. (1995). Playing at research: Methodological pluralism and the creative researcher. Journal of Psychiatric and Mental Health Nursing, 2(2), 105-109.
Sandelowski, M. (1986). The problem of rigour in qualitative research. Advances in Nursing Science, 8, 27-37.
Schön, D. (1987). Educating the reflective practitioner. San Francisco: Jossey-Bass.
Stern, P. N. (1989). Are counting and coping a cappella appropriate in qualitative research? In J. M. Morse (Ed.), Qualitative nursing research: A contemporary dialogue (pp. 124-128). Rockville, MD: Aspen.
Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage.
Street, A. (1990). Nursing practice: High, hard ground, messy swamps and the pathways in between. Victoria: Deakin University Press.
Swanson-Kauffman, K. M. (1986). A combined qualitative methodology for nursing research. Advances in Nursing Science, 8(3), 58-69.
Thompson, J. (1985). Practical discourse in nursing: going beyond empiricism and historicism. Advances in Nursing Science, 7(4), 59-72.
Wildemuth, B. M. (1993). Post-positivist research: Two examples of methodological pluralism. Library Quarterly, 63(4), 450-68.
Wolcott, H. F. (1990). 'n seeking and rejecting validity in qualitative research. In E. W. Eisner & A. Peshkin (Eds.), Qualitative inquiry in education: The continuing debate (pp. 121-152). New York: Teachers College Press.
Vicki B. Drury, M.Cl.Nsg, PGCert.(Psych. Nsg), BHSc (Nsg), B.Ed, RN, RMHN has extensive experience as a nurse and educator within Australia with a particular interest in psychiatric nursing issues. She is currently employed as a lecturer at Edith Cowan University's South West campus and is responsible for coordinating undergraduate clinical nursing practice. She can be contacted at firstname.lastname@example.org.
Morgan, A. K., & Drury, V. B. (2003, March). Legitimising the subjectivity of human reality through qualitative research method. The Qualitative Report, 8(1). Retrieved [Insert date here], from http://www.nova.edu/ssss/QR/QR8-1/morgan.html
Adrian K. Morgan and Vicki B. Drury
Return to the top of the paper.
Return to the Table of Contents.