Data Collection in Qualitative Research
In qualitative research & qualitative studies, the main instrument is the researcher him or herself. The researcher observes, takes notes, talks to people, etc. All of these are skills that need to be learned. As Delamont (2004) wrote: “The biggest problem novices find when preparing for ethnographic field work is that the methods books are not explicit enough about what to observe, how to observe and what to write down” (p. 225). The same applies to carrying out interviews. Just the fact that we talk to people in our everyday life, listen, ask questions and communicate does not make us naturally a good interviewer. According to Helfferich (2009), a good interviewer needs the following skills: Technical competence, Interactive competence: attention and steering, competencies in communication theory and knowing how to deal with previous knowledge and personal bias.
Technical competence is needed in organizing interviews. You need to find participants, make arrangements for the interview, explain matters of confidentiality, prepare consent forms, make your interview participant feel comfortable, and find the right words to open the interview situation.
Interactive competence refers to paying attention to your interview partner and steering the interview into the desired direction. Your job is to open up the stage so that the interviewee feels comfortable talking. The roles need to be clarified: you ask a few questions, but mostly, you listen and the interviewee talks. This violates the rules of everyday talk as the balance shifts. You need to be an active listener, showing interest and encourage the interviewee to speak. You need to find the right moment to ask the next question, to find the right way and form in asking it and you need to keep the conversation going. This entails dealing with silence, reading non-verbal signals and sending appropriate signals. You need to be self-reflexive all the time, controlling your reactions and showing the right level of empathy. This requires some practice, but we all need to start somewhere. When you have the chance, take part in interview training. A good way to check how good or bad you were as an interviewer is when you transcribe the data.
Knowledge about communication theory helps you in recognizing certain dialog signals and strategies of talk. Further, it will help you in mediating roles to generate the right power balance and level of cooperation in the interview. Within the communication literature, you will also find some advice on how to deal with difficult interview situations and participants.
In dealing with previous knowledge and personal bias resulting from it, you need to train holding back or working with your own thoughts, feelings, convictions and expectations. A prerequisite is that you are aware of your biases and that you can explicit them. This is a requirement for overcoming selective attention. During the process of conducting the study, you need to be prepared for the effects that this may have on you. You need to be prepared for having to revise your previous knowledge.
Understanding your research participant is another issue. Based on your personal bias, you may not understand what he or she is talking about or do not find it logical or meaningful. Within seconds during the interview, you have to decide whether it is ok to ask a clarifying question or leaving it to the interpretation phase to gain a better comprehension. The following is a quote from an interview with a person addicted to shopping:
Interviewer: How would you feel if you were to go out without any money?
“I just cannot get out of the door. I can’t get out of the door. I’ve tried it. I had my purse, I lost my purse or I had it stolen I don’t remember and I reported my credit cards missing. This is probably one of the worst things I can remember. And it would take some time to have them replaced, the credit cards. So this particular three days I had no income, no money at all. And, erm, I thought I’ll just stay in for the three days and every time I came to the front door to go to pick my daughter up or take her to school, I couldn’t do it and for the three days.”
I did not interrupt to clarify what she meant by equating credit cards with income. It may also sound puzzling that it is impossible for someone to leave the house because the purse was lost. I continued to listen and let her tell me her story.
In planning an interview study, the first consideration should be what type of interview to conduct. There are a number of different forms and they yield different kinds of data (see for example Helfferich, 2009).
The interview form should fit your research goals. A dialogue produces other forms of data than a monologue. Depending on your subject of research, you may want to find out about subjective concepts or unconscious motives; or you may be interested in biographical self-description or simply in information from an expert. Interviews differ in the degree of steering and structure; you may go into an interview knowing already a lot about the subject matter or you go into it as a stranger; the interview may take place as part of an everyday activity as in an ethnographic setting or in a more artificial context. The focus may be on listening to a long narrative or on working towards mutual understanding and everything in between.
Examples of various interview forms that can be arranged on a continuum of the above mentioned dimensions are ethnographic interviews, narrative interviews, guided interviews, biographic interviews, problem-centered interviews, episodic interviews, in-depth interviews, semi-structured open ended interviews, group interview or focus groups. In recent years, online interviews have also become a possibility. In order to conduct such interviews, you need to be comfortable with the technical requirements. You may only hear but not see the person. Thus, non-verbal signals are lost and important context information may be missing. Advantages are that it is easier to overcome space, location and time constraints. When you have a small budget, you cannot travel all over the place, but you may be able to reach the persons you want to talk with online.
In sum, there are a large number of options to choose from. Your task is to make an informed decision and to be able to explain your choice in research methodology chapter of your report. You find a list of interview forms including references at the end of this chapter for further information.
When considering observation as your mode of data collection, similar issues need to be considered. What is the best form of observation in relation to your research question? Do you want to be an external, a passive, a balanced, an active or a total participant? There are advantages and limitations for each of these observation types (see for example Creswell, 2009). As mentioned above, writing field notes is also a skill that needs to be learned.
Other forms of data to be considered for a qualitative research project include printed documents, online documents, web pages, images, audio and video materials or geographic data. This also needs to be taken into account when designing your research project.
Creswell, John W. (2009). Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. London: Sage. Second edition available online: http://isites.harvard.edu/fs/docs/icb.topic1334586.files/2003_Creswell_A%20Framework%20for%20Design.pdf
Delamont, Sara (2004). Ethnography and participant observation, in: Seale, Clive; Gobo, Giampietro; Gubrium, Jaber F. and Silvermann, David (eds). Qualitative Research Pratice. London: Sage.
Helfferich, Cornelia (2009, 3rded). Die QualitätqualitativerDaten. Manual für die Durchführung qualitative Interviews. Wiesbaden: VS Verlag für Sozialwissenschaften.
Further readings – types of interviews
Ethnographic interview: Spradley, J.P. (1979) The Ethnographic Interview. New York: Holt, Rinehart & Winston.
Narrative Interview: Lucius-Hoene, Gabriele &Deppermann, Arnulf (2002). Rekonstruktion narrativer Identität. Ein Arbeitsbuch zur Analyse narrativer Interviews. Opladen: Leske + Budrich.
Biographic interview: Rosenthal, Gabriele (2004). Bibliographic research. In Seale, Clive; Gobo, Giampietro; Gubrium, Jaber F. and Silvermann, David (eds). Qualitative Research Practice. London: Sage.
Episodic interview: Flick, Uwe (2006, 3rd ed / 2007). An Introduction to Qualitative Research.London: Sage. Chapter 13. / Qualitative Sozialforschung: Eine Einführung. Reinbek bei Hamburg: rowohlt. Kapitel 13.
Focused interviews: Merton, R.K. & Kendall, P.L. (1956).The Focused Interview.Glencoe, Ill. / Merton, R.K. & Kendall, P.L. (1975). Das fokussierte Interview. In Hopf, Christel & Weingarten, E. (eds). Qualitative Sozialforschung. Stuttgart, Klett-Cotta, pp. 171-204.
In-depth Interview: Kvale, Steinar (1996). An Introduction to Qualitative Research Interviewing.Thousand Oak, CA: Sage. / Lamnek, Siegfried (2005, 4. ed). Qualitative Sozialforschung: Lehrbuch. Weinheim,Basel: Beltz Verlag. Chapter 8.
Group Interviews: Flick, Uwe (2006, 3rd ed / 2007). An Introduction to Qualitative Research.London: Sage. Chapter 15. / Qualitative Sozialforschung: Eine Einführung. Reinbek bei Hamburg: Rowohlt, chapter 15.
Focus groups: Morgan, David L. (1988). Focus Groups as Qualitative Research.Thousand Oaks, CA: Sage. / Morgan, David L. and Krueger, R.A. (1998). The Focus Group Kit.TousandOask, CA: Sage.
Qualitative online interviews: Mann and Stewart (2005) Internet Communication and Qualitative Research: A Handbook for Researching Online London: Sage. / Rezabek, Roger (2000, January). Online focus groups: Electronic discussions for research [67 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research [On-line Journal], 1(1). http://www.qualitative-research.net/fqs-texte/1-00/1-00rezabek-e.htm
Sign Up for our Newsletter
Let’s assume, you have chosen a topic where gaining access to the field is possible. If so, a number of selecting strategies are available.
You can go for maximal variation in your data or look at a homogeneous group. Another option is to focus on specific cases like typical, extreme, deviant, positive or negative cases. Based on previous research, you may find suggestions in the literature as well. I call this theory based sampling. This needs to be distinguished from the idea of theoretical sampling as propagated by grounded theory. Corbin and Straus define theoretical sampling as “…a method of data collection based on concepts/themes derived from data” (2008, p. 143).
Theoretical sampling is not something you can determine up front before you begin to collect data. It rather refers to the dialectic process of data collection and data analysis. The purpose of a grounded theory analysis is to build theory and while you are working on creating the building blocks of your theory, you may find that some of your categories are rather “thin” and that you need to collect more data on a particular issue. Or you formulate some hypotheses based on your data, and need some more data material to test these hypotheses. Then it is also time for collecting some more data. Theoretical sampling thus is not suggested by existing literature or theory but by your own data.
You may also select your participants based on criteria, e.g. you want to talk to three ethnic groups with both males and females from two generations. When planning for three interviews in each category, this means 3 x (3 x 2 x 2) = 36 interviews.
If it is only important that the person knows something about the question you want to study, then convenience sampling or snow balling may be sufficient. For a convenience sample you just talk to everyone that comes your way and can tell you something about the issues you are interested in. When using snow balling, you start with only a few people that you select yourself and then ask them whether they know of someone that might be interested in taking part in the research, i.e. throwing the ball to the next person.
In summary, here is a list of the various sampling strategies. A combination of two of them like combining typical and deviant cases is also possible. References are provided at the end of this chapter.
- Maximal variation
- Homogenous group
- Positive / negative cases
- Typical cases
- Extreme or deviant cases
- Criteria based
Corbin, JulieteandStrauss, Anselm L. (2008, 3rded). Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. London: Sage.
Flick, Uwe (2006, 3rd ed / 2007). An Introduction to Qualitative Research.London: Sage. Chapter 11. / Qualitative Sozialforschung: Eine Einführung. Reinbek bei Hamburg: Rowohlt. Chapter 11. / Helfferich, Cornelia (2009, 3rded). Die Qualität Qualitativer Daten: Manual für die Durchführung von qualitativen Interviews. Wiesbaden: VS Verlag. Chapter 5.2. / Silvermann, David (2000). Doing Qualitative Research: A Practical Handbook. London: Sage. Chapter 8. / Patton, M.W. (1990, 2ed ed). Qualitative Evaluation and Research Methods. London: Sage. (pp. 100 – 175).
Very often data collecting takes place in a defined time period and qualitative data analysis is viewed as a next step in the research process after all data have been collected and transcribed. Researchers forego many opportunities and miss out on some of the great advantages of qualitative research – flexibility and adaptation – when they handle their project like this. Sometimes there is no way around it as organizational matters or cooperation within a team won’t allow for a process where data collection and analysis go hand in hand. But when you are free to plan your project and your time frame is sufficient; then I recommend the grounded theory method of data collection. This does not mean that you conduct a grounded theory study; you just take advantage of one of the general procedures of research that they and others before them have described (see for example Blumer, 1969). Strauss (1987/1991) writes about the triad of the research process: data collection followed by coding and memo writing. Both, codes and memos guide the search for new data and can lead to more coding and more memo writing. In later phases of the research project, it is not unusual that the researcher goes back to already analyzed data, revises coding and refines memos. Coding and memo writing continues until the end. When the researcher has the feeling that the achieved level of data integration is not sufficient and there are gaps in the data, new data may even be collected in the process of report writing.
Transcribing and analyzing data early has the advantage of being able to adjust interview questions, asking about new and different aspects that first have come up in the interviews; questions that are truly grounded in the field and not based on your desktop research. Especially novices learn a lot when first transcribing data. Then they realize what went well in the interview and what didn’t work out. They get a better picture of themselves as interviewer and can also improve their own interview skills. Taking it a step further beginning with coding, discovering first preliminary linkages in the data adds further information that supports and helps you in the continuing data collection process.
Above I mentioned the time frame. When you only have three months to complete your research project from start to finish, it may be difficult to implement the dialectic process of data collection and analysis. But you could at least try to fit in the transcription during the phase of data collection. Analysis begins while you transcribe and this is at least something you can take as input with you into the next interview.
A few words about technology:
Digital recorders are wonderful and the quality of the recording is much better than in the earlier days when we only had analogue recorders. But still you need to figure out how to handle them. Run a few trial recordings before your first “real” interview. Test the batteries and just in case take along a new set of batteries. Pay attention to the kinds of buttons you press on your recorder. The recorded interview is a data file and if you press the wrong button, it is deleted in less than a second. As soon as you are at home, transfer the file(s) to your computer and create a backup copy. It may also be a good idea to change the name of the files. The file name that the recorder produces is something like this: WS320122.WMA.
Another option is to use your smartphone or tablet to record data. You can for instance use the ATLAS.ti mobile apps, which you can download free of charge from the apple or google stores.
After the interview:
Plan for having a bit of time after each interview session for writing notes. Then the interview is still fresh in your mind and you can write down your first impressions. It is also a good time for checking your recording. If something went wrong for whatever reason, you can write down what you remember from the interview. This is not as good as having the original wording, but the best you can do when the recording failed. You can still use the interview in your analysis, but need to indicate this in your method section. In most cases, it is possible to clarify some issue that you do not remember correctly via email or by calling the person you interviewed. The notes on the interview – also referred to as interview protocol or postscript (Helfferich, 2009; Cicourel, 1974; Witzel, 2000) – may cover the following issues:
- How did the interviewee appear to me?
- Atmosphere / location
- Disposition to talk / motivation to take part on the interview
- Gestures, non-verbal signals, eye contact
- Interaction during the interview
- Difficult phases
- Duration of the interview
- Specifics of the interview or the interview situation
- The (three) main points that you remember from the interview
It is best to prepare an outline beforehand that fill in after conducting each interview so that you can later compare the notes across all interviewees (see example below).
Another issue you need to take into consideration is research ethics. Depending on which country you live in, your research proposal must pass a human subject board. In other countries this is not necessary; nonetheless, there probably are data protection laws to observe. When you conduct interviews, it is mandatory that there is informed consent between you and your interviewee. This can be an oral agreement or in form of a signed document. You need to explain the purpose of your research, what you intent do with the data, who has access to the data and how long the data will be stored, and in which form the results are used and presented.
A common procedure is to make the data anonymous, i.e. to replace all identifying information like names or persons, location and places, professional status, etc. with pseudonyms or abstract characters like A001, A002 and so on. In preparing a written consent form, pay attention to the respective data protection laws and include the legal regulation and consequences in the formulation of your text. You find examples of such forms online and also in some method books (e.g., Helfferich, 2009).
Blumer, Herbert (1969). Symbolic Interactionism: Perspective and Methods. Englewood Cliffs, NJ. (Introduction)
Cicourel, Aron V. (1974). Theory and method in a study of Argentine fertility. New York: Wiley
Helfferich, Cornelia (2009, 3rd ed). Die QualitätqualitativerDaten. Manual für die Durchführung qualitative Interviews. Wiesbaden: VS Verlag für Sozialwissenschaften.
Strauss, Anselm L. (1987). Qualitative Analysis for Social Scientists.Cambridge: Cambridge University Press. (dt. 1991, Grundlagen qualitativer Sozialforschung. München: Fink).
Witzel, Andreas (2000). Das problemzentrierte Interview [25 Absätze]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 1(1), Art. 22, http://nbn-resolving.de/urn:nbn:de:0114-fqs0001228.