An outline of how to carry out phenomenological research, with advice for novice nurse researchers on how to use this approach
Paula Balls, PGCE, BSc, RN, is practice and education facilitator, Norfolk and Norwich University Hospitals NHS Foundation Trust.
Balls, P. (2009) Phenomenology in nursing research: methodology, interviewing and transcribing. Nursing Times; 105: 31, early online publication.
While nurses can relate to the phenomenological approach because they see it as sharing the values of nursing, this may not be sufficient on its own to start conducting this kind of research.
Using examples from my own research experience, this article aims to examine what draws nursing to this method and why nurses think they may be good at it. It also offers practical advice on how to carry out a good interview, how and what to transcribe and how to use quotes to support research.
Keywords: Phenomenology, Research, Interview, Transcription
- This article has been double-blind peer reviewed
Phenomenology supports the re-examination of a taken-for-granted experience and, through examining the qualities of the experience, allows us to identify its essence.
Therefore, the goal of phenomenological research is not to describe a grand theory or develop a model but to describe accurately a person’s ‘lived’ experience in relation to what is being studied. This could be revealed through asking, for example: ‘What does it feel like to be told you have cancer?’ or: ‘What was your experience when you were discharged home after a myocardial infarction?’
The information (data) for analysis is collected primarily through interviews, stories or observations with the relevant people who have had that experience, that is, the research participants.
Interviews will be transcribed and, after data has been collected from several people, interpretation (data analysis) can be carried out. This may be done using a number of approaches, with the aim of identifying meaningful information and categorising this into a set of themes once all texts have been analysed (Byrne, 2001). These themes will be issues that are common between several participants.
Nurses often relate to the phenomenological approach because it values the individual’s experience and they may feel they already have some of the necessary skills, such as interviewing. However, the type of interview skills they possess may not be suitable to this method.
Using examples from my own research experience, this article aims to examine what draws nursing to this method and why nurses think they may be good at it.
For nurses who are new to this type of research, it gives practical points on how to conduct a good interview, how and what to transcribe and how to use quotes to support research.
Why do nurses choose phenomenology as a method?
An increasing number of nurse researchers have been drawn to phenomenology; possibly because, like nursing, it considers the whole person and values their experience.
Nursing is concerned with understanding people, being perceptive and sympathising with them. It recognises the validity of individuals’ experiences and supports them in exercising control over their own health care (Oiler, 1982).
Nurses are taught to respect individuals, listen to them and believe them. They can also relate to the methods used to obtain data. Nurses are taught to be good listeners, to empathise and to create rapport, often in a short period of time. All these are valuable skills in phenomenology and nurses may feel they are sufficient to allow them to conduct a good interview in this type of research.
However, phenomenological research is not so straightforward and one must firstly decide which of two main approaches are to be used. These are:
- Descriptive phenomenology;
- Interpretative phenomenology.
Many authors agree that if researchers are not clear which approach they are using from the start so that they collect data to suit that method, the value of that research is significantly reduced (Annells, 2007; Dowling, 2004; Crist and Tanner, 2003).
Descriptive phenomenology, which is attributed to Husserl (1963; original work 1913), attempted to make phenomenology a rigorous science within the tradition of its time, and used the concept of bracketing to maintain objectivity. Bracketing involves putting aside what the researcher already knows about the experience being investigated and approaching the data with no preconceptions about the phenomenon (Dowling, 2004; Lopez and Willis, 2004).
Subsequently, phenomenologists such as Heidegger (1962) modified and built on Husserl’s theories and developed the interpretative tradition (also known as the hermeneutic tradition).
Principally, interpretative phenomenologists believe it is impossible to rid the mind of preconceptions and approach something in a completely blank or neutral way. They believe instead that we use our own experiences to interpret those of others.
We may also use our experiences to guide our research questions.
As researchers, we are interpreting something in which we ourselves exist; therefore we have no detached standpoint (Koch, 1995).
For nurses conducting research using descriptive phenomenology, one would expect to see some discussion around how they bracket their preconceptions and ensure a neutral approach to the topic. This may even mean not conducting a literature search before carrying out the research to avoid contaminating or influencing the interviews.
On the other hand, those conducting interpretative phenomenological research will need to show how their own experiences have shaped the choice of research topic, the questions and their interpretations. They may even write down in an appendix to the work what they expect to uncover.
In short, nurses should make sure they read around the different types of phenomenology and use language consistent with the approach chosen. They should not refer to bracketing if using interpretative phenomenology and should not explain how they stand within a hermeneutic circle if using descriptive phenomenology.
There has been some debate about whether nursing research that purports to use phenomenology conforms to standard methods and whether the results really illuminate human experience.
Some authors have considered whether nurses have a particular way of using phenomenology that works for the profession, a ‘nursing phenomenology’ (Annells, 2007; Crotty, 1996). While this is being debated, I do not suggest that nurses invent their own phenomenology, especially if the research is for an academic award. Nurse researchers should follow the advice in Box 1.
Box 1. Advice before starting
Before nurse researchers start they should:
- Make sure phenomenology is the right method, consider other methods and be sure about the aims. For example, will the results illuminate or explain human experience and its meaning, or will they describe it?
- Ensure they have the time to interview and re-interview to really understand something in depth;
- Be clear which phenomenological approach they are using - usually descriptive or interpretative;
- Ensure they use language consistent with that approach throughout and choose a method of data analysis which is also consistent;
- Acknowledge any ways in which they could not follow the approach as a study limitation, such as fewer or shorter interviews.
The interview process
As previously discussed, there are several ways of collecting data for phenomenological research, including storytelling, tape-recording, documentary analysis and focus groups. The most common method is the unstructured or semi-structured interview (Wimpenny and Gass, 2000; Colaizzi, 1978).
The researcher’s role has been described as that of a facilitator to help respondents talk freely. Therefore, the only interview questions should be those that seek clarification, illustration or further exploration (Parahoo, 2006).
While the unstructured interview may be the gold standard in phenomenological research, nurses should not be surprised if the ethics committee insists on the presentation of an interview schedule as part of the application.
Having prepared a schedule – or at least having an opening question – researchers need to consider some practicalities.
An interview area that is quiet and undisturbed is the ideal location. However, for nurses who have their own private office, this is not necessarily the best location. It is important to minimise any power differential between researcher and interview participants. It is also advisable for nurses to promote their role as a researcher, rather than their role in the organisation, which may be senior to that of the person being interviewed.
It is more acceptable to conduct the interview in a neutral location where participants are relaxed, or in a location familiar to them.
Wherever researchers choose, they should ensure they put a sign on the door stating an interview is in progress, divert any telephones and make every effort to avoid disturbances.
Consideration also needs to be given to the equipment used for recording. A tape recorder with a built-in microphone is generally not suitable as it needs to be placed too close to participants to allow a relaxed conversation and may pick up too many external noises.
A better option is a recorder with a lapel or remote microphone where the recorder itself can be placed on the floor in a less intrusive position. While these can be relatively inexpensive, it is worth asking colleagues and any research areas, audio-visual departments or teaching centres to borrow one.
Whatever equipment the use, researchers should try to make the recording process as unobtrusive as possible. Always conduct a sound check before starting to record and, although it sounds obvious, ensure the tape-recorder is switched on.
While not ideal, it may be necessary to ask participants to speak up. I interviewed one person who, whenever she said anything significant, dropped her voice and, during transcription later on, I missed some parts of what she said.
Several authors have suggested carrying out an in-depth interview on a neutral topic with a volunteer before doing the first bona-fide research interview and that this is also transcribed (Green and Thorogood, 2004; Streubert-Speziale and Carpenter, 2003). I would certainly support this as valuable practice and worth the time.
Choosing a volunteer or colleague who can be trusted to be honest will yield valuable feedback on performance without prejudice to the research being undertaken. It will also help researchers to develop interview and transcription skills, and allows a realistic test of the equipment so that any problems can be identified before data collection starts.
Although nurses use interview skills in their practice, these may not be the right kind of skills for this type of research.
Nurses may feel they are good listeners, but when do we really explore, illuminate and gently probe someone’s experiences? As Colaizzi (1978) put it: ‘The researcher must realise that his participant is more than merely a source of data… he must listen with the totality of his being and the entirety of his personality.’
An overeagerness and an urge to problem-solve can get in the way of a good interview, where one needs to sit back and quietly encourage the participant to continue and open up about a past experience.
The initial question in this type of research tends to be open, for example: ‘Can you describe what it is like…’ or: ‘Tell me about….’ From this opening, we hope sufficient information will come forward that allows the interview to proceed smoothly from topic to topic while we, as interviewers, have a limited role in guiding responses.
However, this is not necessarily the case. Some participants may have said all they wish to say in the first five minutes – so it is a good idea to have some questions or prompts as back-up to encourage conversation.
In addition, nurses may have selected a research topic of interest to them but not to anyone else, which means no one has much to say on it. It is a good idea to pause and reflect after the first or second interview to ensure enough data will be gathered. It is demoralising to try to keep an interview limping along while trying not to look at your watch. Pausing and reflecting at this stage may cause researchers to re-shape their interview focus, which can help in generating more useful data for later analysis.
In a similar vein, it is possible to conduct a wonderful interview but then find it did not cover the points the researcher wanted it to, but strayed somewhere and was not pulled back into focus.
I find that having the research question written on a piece of paper in front of me and glancing at it occasionally during the interview to make sure the data collected is relevant helps me to keep on track. Gently guiding the interview back to the topic being investigated may be necessary.
At the end of the interview it is important to thank participants and explain any further interviews necessary. Researchers may also like to offer them a summary of the results to thank them for taking part. It is also advisable to ensure participants have a copy of their signed consent form.
For novice researchers, if an experienced researcher offers to let them listen to an interview they have conducted, or will sit in on a pilot interview and offer feedback, this could be a valuable learning experience. Although it may cause the novice researcher initial embarrassment it will pay dividends later. Many writers agree that involving a more experienced phenomenological researcher can help in developing interviewing skills (Streubert-Speziale and Carpenter, 2003; Drauker, 1999).
In my research, interviewing skills improved over time and with the aid of reflection and the use of a research diary (a journal in which to note my own thoughts).
For tips on interviewing skills, see Box 2.
Box 2. Advice on interviewing skills
Nurse researchers should:
- Use the right equipment, preferably not a recorder with a built-in microphone;
- Conduct a pilot interview with a trusted colleague for feedback;
- Find a quiet venue where they will not be disturbed and where participants feel comfortable;
- Remember they are supposed to be listening, not speaking or solving problems;
- Try to choose a topic about which there is plenty to say and which is of interest to others to generate enough data;
- Ensure the interview stays on track by reminding themselves regularly of the interview question and the type of data being sought.
How and what to transcribe
Hearing and transcribing one’s own voice can highlight shortcomings in interview technique which researchers would rather not have to commit to paper. Nevertheless, it is important for the integrity of the research that the spoken word is correctly typed, with all its faults.
One of the first steps to becoming a researcher may be defined as the development of an ‘acute self-awareness’ of one’s own personal values, perspectives and any biases (Field and Morse, 1992).
A research diary can be an aid to this and I used a journal to reflect on my own interview skills, my opinion of the interviews, including establishing rapport and whether my own perceptions had influenced the course and content of them.
I took no notes during the interview itself as this could have interfered with establishing a good rapport, but used my journal once again to note non-verbal aspects, such as a preoccupation with the tape-recording process, nervousness and other impressions which may not be evident on a later transcript.
Each interview should be transcribed as soon as practically possible after taking place. It is important not to underestimate how long transcription can take. Each 40-minute interview can take up to three hours to write up, not including re-reading it and ensuring accuracy.
Transcription in phenomenology should not be delegated to someone else. Researchers themselves should transcribe the interviews, thereby living with the data, familiarising themselves with and immersing themselves within it.
The process of transcription may sound straightforward but it is rarely that simple. How do you show pauses in conversation? What about the passages which you cannot hear? How do you write expressions such as ‘erm’, ‘huh’ and various others – or do you bother to write them at all?
Listening to my transcripts, it often sounds like I have asked several questions all together, when in fact the first question elicited a non-verbal query and therefore I asked a second question to clarify.
Sandelowski (1994) recognised the difficulties inherent in transcription where only certain features of the interaction can be preserved, and alerted us to the potential for misrepresentation. She has written several texts on transcription and data analysis which are essential reading before undertaking this aspect of the study.
For the purposes of my own research, I took the approach that, as long as I explained in my write-up what I had transcribed and why, it should be acceptable. Therefore, while being faithful to the interview and including all content, even if ungrammatical, expressions indicating participants’ thoughtfulness or hesitation were not included. Occasionally laughter was indicated to allow the transcript to better reflect the spirit of the interview.
I typed my interviews very simply in table form with two columns, one for who was speaking and one for what was said. Sandelowski (1995) also described a method of ‘proofing’ the interview, a process which I followed, whereby the transcript is read again alongside the source audiotape.
During this proofing process I highlighted any points that struck me as significant and added a third column to the transcription to record the number on my tape-counter. I also checked back to my research diary for any significant non-verbal features or impressions I had at the time of interview. See Box 3 for advice on transcription.
Box 3. Transcription
Nurse researchers should:
- Transcribe what they hear, even if this sounds incorrect, shows researchers in a poor light or makes punctuation difficult;
- Use a research diary to record non-verbal aspects of the interview including impressions and their own interview skills;
- Type up transcripts themselves and not underestimate how long this will take;
- Decide how they will deal with unintelligible sections of transcripts and non-verbal aspects or expressions difficult to write, and explain these in the write-up.
Quotations and integrity
This article has not addressed the process of data analysis, which can be undertaken using many different methods. The one chosen should be consistent with the type of phenomenological approach – interpretative or descriptive – and researchers should justify their choice.
Suffice to say here is that whichever method is chosen, it involves categorising participants’ experiences into themes using their own words. In phenomenology these are presented, often in relatively extended quotes, to illustrate the themes.
I had never appreciated until I carried out this type of research how much personal integrity is required of phenomenological researchers, especially in presenting results.
In theory, researchers can be extremely selective about what they choose to quote and the omission of certain sections of transcript can alter participants’ meaning. Consequently, they have the ability to manipulate their study data to reflect any underlying agenda they may have and, to an extent, to select the required results.
It is generally agreed that the only legitimate source of phenomenological data is the individual who has lived the reality being investigated. The transcript therefore becomes the vehicle through which we share or borrow that experience through a description of it. Researchers have a responsibility to remain true to participants’ words and meanings and to represent their experiences.
The words of individual participants are extremely powerful and can be very emotive. One good quote can replace several lines of explanatory text and often does not need further clarification.
In phenomenology, several lines of continuous text are used and should be presented with the participant identifier in brackets at the end, usually a letter or a number. In this way, a word picture can be built up of the different participants and their individual experiences.
To attribute false meaning through selecting quotes is a poor return to participants who gave their time and commitment to the project. As registered healthcare professionals, nurses have a responsibility to act with integrity (NMC, 2008) and this should extend to their research activity.
The quality of phenomenological research is judged on its transparency and whether it can be audited. Therefore, a description of each stage of the process and explanation and justification of the actions taken as a researcher is essential, especially if the research is for an academic award.
Jasper (1994) said the question of generalisability (being able to apply the research to other settings) is inappropriate in phenomenology because the aim is not to produce a theory of general application. However, if researchers provide sufficient detail about the study setting, participants and process, not only can the research be audited but it may also mean it can be repeated in other settings.
One way to demonstrate that the results accurately represent participants’ experience is to arrange a second interview after producing a summary of results and associated quotes, and share these with participants.
In some types of research this is referred to as member checking, although the term is not always associated with phenomenology.
If participants can recognise their own experience within the results and can perhaps see their own words used, this lends credibility to the results.
Nurses should not assume that phenomenological research is an easy option or that they already posess the necessary skills to do it. An understanding of the different approaches in phenomenology is necessary to direct and inform the research process at every stage.
Novice researchers should seek guidance and support but should not be deterred by the language or complexity of phenomenology or by the process of interviewing and transcription.
Carrying out research should be seen as a learning process and skills will develop over time and with experience. What the majority of nurses do have is a level-headed practicality and reflective approach which, combined with a good sense of humour and ability to admit to mistakes, will help them to develop research skills and knowledge that will benefit patients, colleagues and the wider nursing community.
Annells, M. (2007) What’s common with qualitative nursing research these days? Journal of Clinical Nursing; 16: 2, 223–224.
Byrne, M. (2001) Hermeneutics as a methodology for textual analysis. AORN; 73: 5, 968–970.
Colaizzi, P. (1978) Psychological research as the phenomenologist views it. In: Valle, R., King, M. (eds) Existential PhenomenologicalAlternatives for Psychology. New York, NY: Oxford University Press.
Crist, J.D., Tanner, A. (2003) Interpretation/analysis in hermeneutic interpretive phenomenology. Nursing Research; 52: 3, 202–205.
Crotty, M. (1996) Phenomenology and Nursing Research. South Melbourne: Churchill Livingstone.
Dowling, M. (2004) Hermeneutics: an exploration. Nurse Researcher; 11: 4, 30–41.
Drauker, C.B. (1999) The critique of Heideggerian hermeneutic research. Journal of Advanced Nursing; 30, 36–73.
Field, P.A., Morse, J. (1992) Nursing Research – the Application of Qualitative Approaches. London: Chapman Hall.
Green, J., Thorogood, N. (2004) Qualitative Methods for Health Research. London: Sage Publications.
Heidegger, M. (1962) Being and Time. New York, NY: Harper and Row.
Husserl, E. (1963) Ideas: a General Introduction to Pure Phenomenology. Translated by Boyce Gibson, W.R. New York, NY: Collier Books. From the German original of 1913.
Koch, T. (1995) Interpretive approaches in nursing research: the influence of Husserl and Heidegger. Journal of Advanced Nursing; 21: 5, 827–836.
Jasper, M. (1994) Issues in phenomenology for researchers of nursing. Journal of Advanced Nursing; 19: 309–314.
Lopez, K., Willis, D. (2004) Descriptive versus interpretive phenomenology: their contributions to nursing knowledge. Qualitative Health Research; 14: 5, 726–735.
NMC (2008) The Code: Standards of Conduct, Performance and Ethics for Nurses and Midwives. London: NMC.
Oiler, C. (1982) The phenomenological approach to nursing research. Nursing Research; 31: 3, 178–181.
Parahoo, K. (2006) Nursing Research: Principles, Process and Issues. Hampshire: Palgrave Macmillan.
Sandelowski, M. (1995) Qualitative research: what it is and how to begin. Research in Nursing and Health; 18: 371–375.
Sandelowski, M. (1994) The use of quotes in qualitative research. Research in Nursing and Health; 17: 479–82.
Streubert-Speziale, H.J., Carpenter, D.R. (2003) Qualitative Research in Nursing: Advancing the Human Imperative. London: Lippincott Williams and Wilkins.
Wimpenny, P., Gass, G. (2000) Interviewing in phenomenology and grounded theory: is there a difference? Journal of Advanced Nursing; 31: 6, 1485–1492.