Since sampling is not required in qualitative researches, I want to know if there is any minimum-maximum criteria for the number of interviews to be done to meet the research objectives in a qualitative research.
There are a number of influences on how many interviews a researcher may conduct:
1) The type of sampling techniques that are employed (cf Patton 2002, pp. 232)
2) Resourcing of the study, can place limitations on what sampling is feasible (Kvale 1996; Seidman 2006). If the interviewer can travel only within a limited area, then there is a geographical limitation on the sample. For example, Stoodley (2009), limited his travel to Southeast Queensland due to resourcing (p. 76). Such a geographical limit imposes a cultural limit on a sample.
3) Sampling continues until the researcher senses she has reached saturation. Saturation is a problematic term (Guest et al. 2006; Mason 2010; Morse 1995). Since the first use of the term “theoretical saturation” by Glaser and Strauss (1967, p. 61), the meaning of saturation has become blurred. Glaser and Strauss intertwined data collection and analysis for one category until saturation, before moving on to collect and analyse data for another category. (In grounded theory, a category is a conceptual element of the grounded theory being discovered (p. 36).) The type of saturation the researcher is aiming for may not be theoretical saturation. “Saturation of knowledge” (Bertaux 1981, p. 37) is a better term. Bertaux describes how the researcher is surprised or learns a great deal from the first few interviews. By (say) the fifteenth interview, the researcher recognises patterns in the interviewees’ experiences. More interviews confirm what the researcher has already sensed.
How saturation of knowledge is reached or passed during sampling is uncertain. According to Mason (2010), it is more likely PhD students using qualitative interviews will stop sampling when the number of samples is a multiple of ten rather than when saturation has occurred. Guest et al. (2006) found that 12 interviews of a homogenous group is all that is needed to reach saturation. Conceptually, saturation may be the desired end point of data collection. Operationally, the decision to stop interviewing is a function of a combination of all or some of the following factors:
• interview structure and content (Guest et al. 2006); the more unstructured and variable the content, the more interviews are required
• heterogeneity of the group (Guest et al. 2006); the more heterogeneous, the more interviews are required
• the number of interviews done already (Ryan and Bernard 2006); the weaker the sense there are enough interviews, the more interviews are required
• the complexity of the interviews (Ryan and Bernard 2006); the greater the complexity the more interviews are required
• the researcher’s experience, fatigue (Ryan and Bernard 2006), and confidence (Mason 2010)
• the number of researchers in the research team (Ryan and Bernard 2006)
• the more interviews, the more defensible the researcher believes the research will be (Mason 2010)
• doing what was stated in a research proposal (Mason 2010)
• the nature of the sample being limited by the sampling technique (Browne and Russell 2003)
• resourcing (Kvale 1996; Seidman 2006)
• the orthodoxy of the method: the number of interviews is expected to fall within a certain range
• meeting all of the purposeful sampling criteria the researcher has determined that are necessary for the study.
This will depend on the size of the pool of potential interviewees, and the particular qualitative interview technique that you are planning to use. We need more information about your study to help you!
Amrit, I think Andrew's advise is excellent.
Sampling is key to good research (whether it is qualitative or quantitative or mixed). Good qualitative research is as rigorous (sometimes even more rigorous) than quantitative approaches, so I would review the design of the study and see whether convenience sampling vs a quota or a snowball sampling strategy would meet your needs. But Andrew is right; there is no clear guideline on this issue.
Luborsky and Rubinstein (1995) argued that sampling for meaning was at that time one of the most appropriate sampling approaches and I would assume this continues to be the case, at least in some fields where qualitative research predominates (read "Sampling in qualitative research on Research on Aging"). They also provided some rough guidelines: between 12 and 26 subjects is about what most authors look for and may also fit well with available budget and staff as Ashwin pointed out.
A good criteria is to questionnaires is to observe if you have too many similar completions/answers, but in Qualitative Research we have to observe who are the critical people to be interviewd or which comments are important. In other researches, a good tip is know how many people compose your main target and get a percentage of... says...20% more or less or until you reach important answers to your problem. Good studies!
Qualitative research is differ than quantitative you cannot calculate sample size before starting. But you can follow the following points to get enough interviews
- repeated responses which mean that you almost cover most.
- investigate different groups till you fulfil role one (gender, urban and rural, education levels, economical levels, different political parties, different age groups ....etc)
-your fund run out :(
sample size might be useful in generalisation which is not the point in qualitative research
You could check the book "Your Education Research Project Handbook by Anthony Coles & Jim McGrath" ... hopes this helps
Actually the beauty of qualitative research is to find an answer for 'WHY' & on the other hand quantitative study would respond to 'WHAT', although views varies from researcher to researcher still at least 10 respondents in each category/sample is needed for qualitative study.
While conducting the study you will realize that the information from the respondents are repeating that means nothing new is coming out and that is the end point of the study.
However, sample size can be determined before the initiation of qualitative research and purposive sampling is recommended to avoid frustration.
There is no rule as such. However, if your qualitative method is designed to meet rigour and trustworthiness, thick, rich data is important. To achieve these principles you would need at least 12 interviews, ensuring your participants are the holders of knowledge in the area you intend to investigate. In grounded theory you could start with 12 and interview more if your data is not rich enough.
In IPA the norm tends to be 6 interviews.
I don't think there is a specific number. It really depends on what you are trying to achieve, and the role of the qualitative interviews. In some cases one interview might be enough, and in others a much larger number would be preferable.
Let's revisit your question. How do you mean "are needed in qualitative research?" Left to readers, your question can mean how many interviews are needed to be "portrayed" to "reviewers" in order to increase the likelihood of a paper based on the interviews to be "accepted". If that is the question's intent, then my observation is that most disciplines have learnt not to tolerate any less than a dozen interviews in any methodological context. But this is invariably irrelevant if your question is intended in the context of your gaining useful knowledge of whatever domain you are trying to learn about. You simply do as many as you can under the constraints of your situation. There's always more to learn by talking to people.
Turns out to be a fun discussion, so, as a poor physicist hard-pressed by dark matter and even darker energy out there, I stick to ELEVEN, the greatest number of dimensions for a supersymmetric theory. Finally, interviews are somewhat symmetric regarding interviewer-interviewee interactions.
“The phenomenon dictates the method (not vice-versa) including even the type of participants” (Hycner, 1999, p. 156). The number of participants should be determined based on the purpose of the research (Babbie, 1995; Greig & Taylor, 1999; and Schwandt, 1997). Seidman (2006) recommend sufficiency and saturation as the two criteria for deciding the number of participants. Sufficiency refers to the amount and range of participants needed to reflect the population, while saturation of information refers to the point where the data collection no longer reveals new information. Boyd (2001) suggests that saturation can often be reached after interviewing two to ten participants. But two to ten is often too few to meet the sufficiency standard, depending on the context.
@Ashwin: I am not sure that I understand your question, but I will give it a shot.
I do think that there are some qualitative research contexts where arguments that saturation has been reached are quite convincing. Let me give a silly example that might demonstrate the point. Let's say I want to know what someone ate for lunch (maybe I am doing some epidemiology work). If I ask him, he might say pizza and a coke. If I ask his wife to confirm that statement, she might say, "yes, I was there; he had pizza and a coke." How many more people would I have to ask in order to feel confident that I have attained saturation and sufficiency? For me, the answer is zero.
Now if the question is something like "what should be the United States foreign policy toward emerging democracies?" or even "which football team is the best?", interviewing only two people would be clearly insufficient. The universe of probable answers should be taken into account.
Also, you can make a pretty good case that you have met the saturation point when the modal response is unlikely to change even with the introduction of new data (especially if you are using a truly random sample).
Sorry to be long-winded! :-)
I actually believe that we are not in disagreement. With your research (conceptualization of hope), the universe of potential responses is infinite. By infinite I mean that the limits of the set are unknown. In this case, you would be unlikely to reach saturation after 10 to 15 interviews. You might not even reach saturation after 50 or 100 interviews, maybe even more. So in your context, it might be an unreasonable burden to require saturation. Remember that saturation is just one way that you can argue that you have done enough work. :-)
By " the modal response is unlikely to change even with the introduction of new data," I mean that when the one or more themes is so prevalent in the data that the introduction of new data is unlikely to "topple" those themes. That is as clear as mud, I fear, so let's look at an example. If I want to see which nation people think will win the word cup, I might interview 400 people randomly distributed around the world. I look at the results and see the following distribution of responses:
Spain - 100
Germany - 90
England - 80
Brazil - 40
Italy - 20
France - 20
Australia - 5
Other - 45
If I interview 50 more people, my top three vote-getters will not change, nor is it likely that even the top six would change. I would feel pretty secure that I have collected enough data at that point.
Make sense?
Cheers,
Steve
Whoops, *World Cup, not word cup.
And I apologize in advance if I missed your favorite team.
As a young academic just foraying into serious academic life, I found the contributions of Ashwin and Steven very engaging and enjoyable. They made some things clearer to me. Thanks both
From the perspective of qualitative research, the need is not a representative sampleit can be snapped from the case study: an interview with one person representative of a social group. When working with a general population, I usually use an interview age group, and if financially you can do, an interview by sex in each group. When finished, you can have several outcomes: can infer the results for age group, or all age groups by sex, age group for sex, or choose between adolescents, young adults or seniors.
I enjoyed the discussion!
I am glad I could help. Let me know if I can be of service in the future.
Thanks all of the colleagues and scholars who contributed immensely to add up into my question, it's been a vivid discussion................
My review of literature on number of participants for interviews in qualitative research returned a resounding 30
My suggestion to students is saturation plus one or two if no new ideas/themes codes then you have enough. Saturation basically means that no new themes emerge from the data, then do one or two more and again no new themes or ideas this your data is saturated.
Dear Amrit
I totally agree with other colleagues and researchers that there is no definite rule about the number of interviews needed in a qualitative research. In my opinion not less than 7 or 8 and not more than 20 interviews should probably be a standard and a quick answer to your question. Since "saturation" occurs, which means that themes tend to be repeated, you shouldn't need to take more interviews. However, time limits given for the research, the research design and the population which you investigate, the research inquiries you pose, the theoretical framework and the qualitative approach being used should be factors to be considered during the whole process. Moreover, keep in mind that in qualitative research generalization is not the issue as it is in quantitative methods. According to Peavy (1998) and Atkinson (1998) the aim of a qualitative research is to describe the multiple ways in which participants generate meaning to their experiences and not hypotheses testing. A qualitative researcher always tries to figure out the thought and action patterns of the participants but these patterns could be triangulated to expand and clarify the data in a mixed methods design. I suppose that for additional information and tips you would find helpful the books mentioned below: Creswell, J. W. (2009). Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. Thousand Oaks: Sage, and Creswell, J. W. (2007). Qualitative inquiry and research design: Choosing among five approaches (2nd ed.). Thousand Oaks, CA: Sage.
Best wishes for your research!
Andronikos Kaliris, PhD student, Department of Psychology, University of Athens
I am currently undertaking a set of qualitative interviews and when you are doing them you are looking for different things than a quantitative interview. I am looking at how a group have seen change and development for their social group, and what life was like for them before and after.
Quantitative interviews look to find a generalisation. Like all men with white hair are old or all seniors living alone after 80 years of age have lower social interactions than those who are in senior residences. In those examples you want to go beyond your sample so you need to get a mathematical number to say that you have a general trend. In qualitative interviews you are looking for the back story. A deeper meaning is sought and a person is not just a dot on a graph, but a whole graph on their own. The limitation is that you can’t say what happens outside your sample but for those inside your sample you can get a really good look at the process and the lived reality of being that person.
There are plenty of books on qualitative methods, but the thing to remember is that you need to do enough that you have reached saturation (enough data) but it depends on the individual project. For me I am looking at 12-15 interviews, but a colleague of mine did 22 for her project. It is really based on when you have enough anecdotes that they start to become data. Check with your supervisor or a T.A. that might be able to have a better idea of scope for you. You also have to remember that these interviews are generally transcribed and that takes time. Ask anyone who has done it. 1 hour interview is about 3 hours of work give or take. If you need any other help let me know!
I have asked myself the same question when trying to use qualitative research for quantitative data gathering. For architectural research on man-environment relationships, at least 12 ethnographies may provide semantic differential data for 400 questionnaires:
www.arquitectonics.com
Most qualitative methodology articles and books suggest the reaching to saturation point and richness of the gathered data. I have collected my field data and conducted intreviews with 10- 16 interviewees.
I advise 12 ethnographies which does not mean just making 12 inteviews.
To me the main distinction between qualitative and quantitative research is numbers. I had similar problems when I was doing qualitative research for the first time. Coming from an engineering background and doing qualitative research in construction I just couldn't help keep numbers out of my mind. But once I learned to convince myself of not thinking about numbers and imesersed myself in data i.e. just jumped in straight into asking questions and looking at the respondents responses made me realise that in qualitative research numbers really are insignificant. It is best not to think of numbers i.e. maximum or minimum number of interviews but rather to keep interviews going till one reach a point of saturation i.e. when you feel that there is no new more information coming out from the interviews. Once you reach the point of saturation that is your final number of interviews. I hope this information is of use to you please do let me know if it was helpful. Thanks
it depends on what is your aim. I would say you should know how diverse is observation case and you should include one informant from each category. in Qualitative methodology one stops gathering date when it does not yield any significant new information. You should look when your information is saturated and stop thereafter.
If you are intersted in conducting research in the realm of quality, the term "qualitative research" may be too broad for practical reasons. You may need to pick up a specific methodology (phenomenology, hermenuetics, grounded theory, ethnographic, etc) and that will determine your sample size. For example in phenomenological research the number of particpiants may be as small as one; yes, one. (Morse, 2000). Once you determine your methodology, you may google it with the size request and you'll find a number of publications addressing the issue.
Best!
There are not suggestions like a quantitative research. The saturation indicator depends on the specificity in the responses. The most important is the selection of informants.
There is no fix number for interviewing qualitative research. Your questions should be related to the topic for which you have design the set of questions for interview. It is a cross check to the answers provided in written questionnaire.
Regarding rigour, richness and trustworthiness of data and saturation issues, there is need to cater for variability in the respondents. One could create homogeneous units (similar categories) and pick respondents from within the categories. This helps in making sure that different perceptions are pooled. However, this division depends on ones research objectives.
With my present study, I did not think too much about the number of respondents.I had a total of 5 interviewees. Using a matrix method comprising of awareness, problems identified and suggestions derived from the interviews, i was able to cross check the answers and managed to reduce the data to a point of saturation. Check out the Miles and Huberman semi-structured interview method.
.you can apply precision test for the questionnaires on it numeric answers than use it on the highest value . That should be your appropriate size
There is no rule but appropriate number of samples are essential...again remember there is no definition of appropriate
Basically Numbers of sample depends on
1) Size of universe
2) Nature of problem you chosen for investigation
3) Objectives of your research
4) and Statement/principle/theory that you want to generalize ?
**** if you realize that 10 or 20 or 30 or ............1000 samples are enough to draw conclusion that is ok to conduct interviews. ******
As per my knowledge, every time sample size is not important. what feedback getting from them this is important.
From the experience I got on the field, the qualitative research tool that is used the most on the field is interview (whatever type of interview, it is up to you). I think it is more easier to interview people to capture their opinions on a certain topic but again maybe it depends on the topic and communities or informants you are working with.
Once I asked the same question to one of researchers who widely used qualitative method. He said that it depends on number of questions and the saturation level of the answers. When you feel that people are repeating the answers previous interviewees, than the number is okay. So it is more up to the reply of the people. In some cases you might not have enough people to interview or you might have a tight budget and time. These things affect the number.....
There is no set numbers. Just use common sense you do not want use very large numbers or too small sample.
As described by other friends, there is no exact measure of the respondents in the qualitative research. Only researchers who know, whether research goal has been reached. If it is, that's the limit of the number of respondents who had interviewed
Mason, M. (2010), ‘Sample Size and Saturation in PhD Studies Using Qualitative Interviews’, Forum: Qualitative Social Research, 11:3, http://www.qualitative-research.net.
This article would be helpful.
Keep interviewing participants until you reach data saturation. Researchers said that you can reach the point of saturation between 12-20 participants, or from 20-25 participants.
Hi. Observing the different PhD candidates using qualitative research, the minimum number of interviews matches the largest Focus group size ~ 12 persons and it could grow up to 30 persons for a good comparative analysis.
I would suggest 10 depending on the topic under consideration. It is important to remember that deviant or stigmatised populations are much more difficult to recruit. So, in my experience your topic and sample may largely dictate the amount of interviewees.
There what qualitative researchers know as saturation point. Qualitative is indepth research with a small sample of people. Research can be conducted as one to one interviews or in focus groups.Saturation is a point in qualitative research where the researcher has exhausted all possible themes that the data suggests. It means that any additional interview will not bring in any new data -data saturation
I am currently conducting a qualitative study and have been discovering new things- saturation may happen for some questions in the interview guide but not others. Still, new insights keep coming up for which one has to follow through and I realize that one has to stop collecting information somehow. This therefore means that saturation is not a stand alone criterion for judging when to stop but must be considered alongside objectives of the study, the methodology used, time and resources available, among other things.
there is no thumb rule but you can apply the precision test on the numeric numbers of the test sample then higher value shall be the size of sample. this test is explained in Schindler book and i used this in my research too.
I agree with linda's observations.
Often academic samples for qualitative work are ridiculously small (in qual interviews and focus groups etc) compared to commercial research which is a real pity because academic research should be at least as rigorous.
Build your sample based on anticipated sub samples or differences, use the more academic saturation point to some degree but don't fall into the trap of thinking you have it all from tiny easy samples.
It depends on the depth of the information and the saturation. If you feel you have reached the deepest level of the information you look for, and if there is no more significant insights from all the potential informants, then you can stop the interview. You can identify the depth and the saturation by simultaneously analyze the data during collecting. That's the fun of the qualitative research (:
I guess it will also depend on your focus, how many participants do you have and what is the research for? It is also important that the researcher/ interviewer understands the focus of the study and the reason for using interviews. While I agree on data saturation being one of the rule of thumb, but I always wonder what is data saturation. Data get saturated in some cases among several due to interviewing skills and the mood of the participants hence the uniqueness of qualitative study. I would always say ensure that you have covered your ground and always be on the look out of the "blind spots" and ensure that you cover them before you say you are done.
Answer would vary, and is subject to the number of questions, time, money and so on. I have seen rigorous qual work with a sample of one (case study); as well as samples much larger (multiple focus groups). A lot also depends on your analytical tools available, for example software: Atlas-ti, Folio Views, etc. Last but not least, if this qual work is for a dissertation, the really practical answer is as few as your Committee will allow...the objective is to finish. Good luck!
I agree with what has been said about saturation. Once the interviews or observations are resulting in information already heard based on the questions asked, the researcher has reached a point of satisfactory sample size. This will also vary depending upon the type of qualitative research being done. Grounded theory, for example, generally requires collecting information until theoretical saturation has been reached which takes more data. There are maxi and mini ethnographic studies which require differing amounts of data saturation.
Members of IRBs, particularly those who are more familiar with quantitative methods, often will require a given numbers of subjects to be interviewed or observed. A researcher can estimate the number, saying it is an estimate, by the typical number of individuals interviewed in similar qualitative research projects but also adding the interviews will continue until saturation is reached.
A confounding factor in planning the number of interviews needed in qualitative research is the requirments of the ethics committee or research review board. If this group is populated by psychologists or others with a strong quantitative background they may insist that you show the planned research population to be balanced and sufficient to satisfy QUANTITATIVE standards. This will mean that they will expect between 20 and 30 of each category, irrespective of arguments about saturation. It is best to check the backgrounds of those involved in the review process before making a plan. This is the wrong way round, but if the plan is not approved there is no research project. I wish it were otherwise.
Charles
As a rule of thumb, in the research I do, I tend to conduct interviews until such time as no new themes emerge from 3 consecutive interviews. This assumes, of course, that you have access to an appropriate population for your research question and that the analysis is done soon after interview is complete.
What type of objectives and approach is sole determinants of numbers of questions requirement in questionaire. Decide in the design such as grounded theory, phenomenology building, and case study.
Depend on many factors such as the topic of research , level of experience and knowledge of participants and depend on the objectives of your research. To collect reliable and valid data , it's very imperative to meet data saturation.
Hi Amrit!
I considered this question in my proposal and took a couple of tacks. First, I looked at the literature on data saturation for qualitative studies. Next I performed an analysis of the 10 most recent dissertations that used semistructured interviews, were broadly aligned to my topic, and came from unique institutions. I found the average sample size of these dissertations landed right on the data saturation literature. Third, I built in Francis Buttle's recommendation - for data saturation, that no new themes arose in the final 2 interviews. A solid defense of my choices!
Hope this helps.
Gail.
The reasons you want to predetermine how many interviews to do will influence your answers. IRBs tend to want specifics about number of interviews as well as number of subjects. A good estimate is helpful for them with the caveat that as a researcher you will need to follow the data that emerges and adjust the interviews thusly.
As much as we try to predict the direction of qualitative research we are usually in uncharted waters. Sometimes there are no new themes coming from the data but there might be a new direction for the research. It becomes a choice for the researcher to pursue that new direction or not. I believe it is important for qualitative researchers to be open to the unexpected and not rigidly adhere to predetermined structure - not that anyone here is suggesting that.
My first qualitative study was about the knowledge base used by critical care nurses when taking care of patients. The interviews with the nurses were coming along as predicted until I noticed something odd about the transcriptions. Because of my limited budget, I had told the typist to leave medical terms she wasn't familiar with blank so not to spend time looking them up. I would fill those in later. Some transcriptions came back full of blank spaces and others hardly any. What I realized after analyzing these transcriptions is there were 4 types of patient needs addressed by the nurses - technical, physiological, physical and psychological. The nurses would generally see two needs as exceptionally important with minor attention given to the third and unaware of the 4th need. A physiologically oriented nurse would focus on data coming from all the equipment in the ICUs, often not concerned about back rubs, lotions, skin care, for example. A physical care nurse was more focused on back rubs, lotions, referring to the physiological needs of the patients as doctor's work.
This finding was not expected and unplanned for yet important for training programs for critical care nurses.
Qualitative researchers need to have the freedom and mindset to interview more or less as the data shows them the direction not just more data related to a theme. That direction is usually not predictable before time.
This is a typical question, as our professors would say, and the typical answer is: it depends. And really it does mean that depending on the research 10 would sound more than enough when for others, a hundred would be too little. To take my experience, I had a qualitative research question to answer about Western view on Chinese gardens for my MA, and after obtaining 14 answers for my questionnaire (I know it is not an interview but I believe it is similar for the sake of the argument) my supervisor was quite content, because these included the most important targets in my research (aka world professors specialised in Chinese gardens).
So I do believe that you will have to decide on the way how much is necessary, and it might also be that you revise your opinion after starting. As long as you do not modify too much your questions and your methods from one participant to the other, it should be fine - but then again it depends on the context.
Good luck
There is no such rules however, it depends on the research sample size. To verify the information you have collected through questionnaire, interview of the persons who have filled in questionnaire is essential .
Perhaps data saturation in qualitative research may determine how many interviews you do. Then it becomes essential to know when saturation is achieved
BW
Grounded theory method present qualitative researchers with the oppoturnity to collect data until they reach theoretical saturation. Therefore, the method you are using becomes critical to answer your question.
BW
My son, 16, will do some interviews with 16+, kind of cool hunting, for the Berne city's youth animators work (running out of money). I found the above discussion lively at first, but then disappointing, because there is really no trace of a "psychosocio-thermodynamic" approach - i.e. nobody seems to ask the question, what the basic invariant properties of the individuals and their distribution "are" by some rational model - & how this presumably will feed back on the way they experience a problem, and give an answer to this.
Now psychological personology has long come up with answers to this. Julius Kuhl's model gives each of us a triple on the dimensions dominance-performance-affiliation. (I like to give orders & not take them, perform, and / or groom).
Temperament is used since Plato's state and assigns (according to my theory) an habitual corridor on a Thought/Action-plane, a trajectory with variable ratios of "T" and "A".
Sociologist usually are uncertainty-oriented (a term from Richard Sorrentino's "The uncertain mind" - he discovered that certainty orientation is the highest level of motivation; Calgary univ.). They believe that everyone is like them: but it's only a third; another third are "neither fish nor flesh" (Kant); the last third are colerics: they know thinking is bad for mood.
I hope the above engineer will pick up on this - else we won't know what taciturn colerics think and do for another century. ;-)
It all depends on what you are doing, there is NO rule. Some interviews may give you very little data, while other may provide a rich insight. So I would tend to ignore any kind of 'theory' and get enough data that you feel that is enough to work with. More interviews does not mean that you have a more rigorous study.
- Badges
- Science method
- Similar topics
- Methodology
- Research Methodology
- Qualitative Research