|Year : 2016 | Volume
| Issue : 2 | Page : 41-47
Sampling and methods of data collection in qualitative research
Professor, College of Nursing, CMC, Vellore, India
|Date of Web Publication||9-Jun-2020|
Source of Support: None, Conflict of Interest: None
Qualitative research deals with data collected in narrative form which requires an in-depth knowledge in selecting appropriate sampling techniques and data collection methods. Unlike quantitative studies, non- probability sampling techniques are used in qualitative studies to enhance information richness. Sample size is not determined using statistical formula, but is based on the principle of data saturation and the type of qualitative study methodology. Qualitative studies adopt flexible data collection plans which may evolve as the study progresses. Observation, in-depth interviews and focus group discussions are the primary data collection methods in qualitative studies.
Keywords: qualitative methods, purposive sampling, snowball sampling, theoretical sampling, sample size, observation, in-depth interview, focus groups.
|How to cite this article:|
Seetharaman B. Sampling and methods of data collection in qualitative research. Indian J Cont Nsg Edn 2016;17:41-7
| Introduction|| |
In nursing, the two widely used approaches to carry out studies in education, service, and administration are quantitative and qualitative in nature. The steps in doing research with these two approaches are different from each other in terms of sequence and methodology. Generally, researchers believe that the data collection in qualitative approach is quite exhaustive and time consuming compared to the quantitative approach. Due to this reason, researchers may not be inclined to adopt qualitative methods. However, qualitative studies play a vital role in generating new concepts and developing nursing theories (Polit & Beck, 2014). This article discusses the various sampling techniques and data collection methods adopted in qualitative studies.
| Sampling in Qualitative Research|| |
The main goal of qualitative research is to understand a phenomenon, not to represent a population or make generalizations from study samples to population. The participants are selected using non- probability sampling techniques which will allow the researchers to select people who will give a wealth of information (Polit & Beck, 2014).
| Types of Qualitative Sampling|| |
The non random sampling techniques used in qualitative studies are classified below:
- Volunteer sampling
- Snowball sampling
- Purposive sampling
- Theoretical sampling
A. Volunteer Sampling
Volunteer sampling is one method of convenience sampling where the researcher posts a request for volunteers who had the experience being studied or who can talk about the phenomenon of interest, to participate in the study. Volunteer sampling is helpful when the study topic or the experience is difficult to identify through routine survey or when the study topic is sensitive in nature. For example, if a researcher wants to explore the experiences of women with urge incontinence, volunteer sampling method will be a good choice. Although this approach is easy and economical, it is not a preferred method always, because of the fear that the volunteers may not be those who can provide sufficient information (Burns & Grove, 2004; Polit & Beck, 2014).
B. Snowball or Chain Sampling
In snow ball sampling, study participants recommend others who have the same experiences or have similar characteristics which are of interest to the researcher (Burns & Groove, 2005). The early sample members are asked to refer others who meet the eligibility criteria. This method is more cost-effective compared to convenience sampling. However, this strategy can cause ethical problems if the study deals with personal or sensitive areas such as sexuality, drug use and so on because confidential information about participants is revealed to the researcher before getting their consent (Polit & Beck, 2014). [Figure 1] depicts graphically, the result of one snowball sample. The first participant aged 27years, from UK who referred the researcher to the second person from Italy and goes on.
C. Purposive or Purposeful Sampling
It refers to selecting participants who will most benefit the study. It also involves inviting participants who can give the most information on the experience that is studied. A wide array of purposive sampling methods are described by Patton (2001) and Patton (2002).
i. Criterion Sampling
Criterion sampling involves selecting people who meet some predetermined criterion of importance to the study. This sampling can be used following a survey study if the researcher wants an in- depth analysis from the particular participants identified, e.g., A researcher wants to explore patients’ satisfaction with nursing care. If the level of satisfaction is assessed using a 5point Likert scale with 1 = ‘not at all satisfied’ and 5 = ‘extremely satisfied’ a criterion that may be used for qualitative sampling is: “Interview with patients who score only 1 and 2”.
ii. Maximum variation sampling
This strategy involves cases representing variations on specified aspects of a phenomenon of interest e.g. geographical variations, education etc. This type of sampling helps in capturing and describing central themes that cut across variation (common patterns among variations) resulting in high quality, detailed description of each case, e.g., Evaluating the essential newborn care programme in districts of India. Sample could be mothers with children less than 5 years. Participants need to be purposefully selected representing geographical variations e.g. sites in rural, urban, semi-urban areas.
iii. Extreme or Deviant Case Sampling
This sampling method is sometimes called as outlier sampling. It includes selection of participants who give information which is very different from the majority of others (unusual or special in some way) or at either end of a continuum instead of selecting typical cases or a randomly selected one, e.g., outstanding successes andnotable failures.
iv. Homogeneous Sampling
This method includes selection of similar cases to describe a particular group in-depth. It is often used in focus group discussions, e.g., Selecting participants who have similar demographic characteristics such as age, gender, education, etc.
v. Critical Case Sampling
Critical cases are “information rich” participants (Patton, 2002) who can make great impact on the development of knowledge. When such participants are purposively included as samples it is called as Critical Case Sampling.
D. Theoretical Sampling
According to Patton (2002), theoretical or theory- based sampling is defined as “a strategy involving the selection of incidents, slices of life, time periods or people on the basis of their potential manifestation or representation of important theoretical constructs.” This sampling is used in grounded theory research. It is based on the question, ‘Who will provide the data that will assist in the development of theory that is evolving?' (Glaser, 1978). The data can be collected from any individual or group by the researcher which can be used as a relevant source of information for theory generation.
| Sample Size in Qualitative Research|| |
Unlike quantitative studies, the sample size is not mathematically calculated in qualitative studies. There are no set rules for deciding on the sample number in qualitative studies (Patton, 2002). However, one must keep in mind that the sample should be large enough to make meaningful comparisons in relation to the research questions. The guiding principle used is ‘data saturation’ (Polit & Beck, 2014). Itmeans, sampling to the point at which the researcher recognizes no new information is obtained from the participants and data on all dimensions of an emerging category is generated (Glaser, 1978). In qualitative research, sampling method and sample size may vary and change as the study progresses and findings evolve.
Sample size also varies based on the Qualitative traditions chosen for the study. In Ethnography there is no restriction on number of samples. Phenomenologists who look at lived experiences of people choose less than 10 participants who have experienced the phenomenon. In grounded theory studies 20 to 30 individuals may be included but as data saturation is specifically emphasized the number may vary accordingly (Polit & Beck, 2014).
| Data Collection in Qualitative Research|| |
There are three main data collection methods in qualitative research. They are observation, in-depth interview and focus group discussion. These methods differ with respect to the relationship between the data collector and the participant (Rossman&Rallis, 1998).
1. Observation methods
Observation method involves observing and recording what is seen. This method is used not only in qualitative but also in quantitative research. Observation is also part of the interview and group discussion as the researchers also observe the body language, facial expression and other nonverbal cues while being an interviewer or the moderator for group discussions. The qualitative observations will differ from quantitative approach primarily in the focus on process rather than numbers. The types of observation are discussed below:
a. Participant Observation
As the name implies, the researcher will participate in terms of interaction with participants and their activities. “Participant observation is the process of enabling researchers to learn about the activities of the people under study in the natural setting through observing and participating in those activities” (DeWalt & DeWalt, 2002). Participant observation requires a prolonged period of engagement with the participants in the study in their social and cultural context(Patton, 2002). This method of data collection is used extensively in Ethnography research.
To illustrate this process imagine that a researcher wants to study about care giving experiences of spouses of mentally ill patients. The researcher after getting an informed consent may have to spend the initial days in establishing relationship with the family and then continue observing the care giving activities that are meaningfully related to the experience. The researcher continues to dialogue with the spouses as observations are carried out.
b. Mystery Client Technique
It is a special type of participant observation which may combine quantitative and qualitative data collection and is used in patient-provider studies. This method is used when the situations that are likely to make alterations with respect to the participants’ behaviour in the presence of observer. Participants/patients are trained to enact roles which will reflect their real-life experiences and they will present themselves to actual health care providers in the natural setting. They record their experience on a structured form and report their observations to researchers after each encounter. The key feature of this method is that the provider is not aware of the patients’ real identity. The decision to use this method needs to be taken only after careful reflection on the purpose of the study and ethical implications linked to this, because getting an informed consent is an issue (Ulin, Robinson & Tolley, 2004). For instance, this method can be adopted to study on interactions between patients and health care providers.
c. Non-participant or Non-reactive Observation
In this, the researcher collects data without interacting or reacting visibly to participant’s activity. The researcher observes events as unobtrusively as possible as an outsider. For instance, the researcher may want to study the admission procedure of a woman in pain getting admitted into the labour room. The researcher observes the patient’s experience including the interaction between the participants and health care providers without participating in the event.
The quality of the non- participant observation data relies on the ability of the researchers to watch diligently and listen without interrupting the natural flow of activity. The distortion of observed behaviour must be kept at minimum. For this, the observer needs to be introduced to the participants as someone who will learn about the health care provided in that setting. This will further allow the observer to remain with participants and observe silently from the periphery while providing care. The longer the observation period, the more effective will be the observation as the participants become accustomed to the presence of the observer.
Taking notes on the observation made is very important, and based on the notes the list of things or events to be observed will be revised at the end of each observation period. Note taking needs to be done as unobtrusively as possible because frantic scribbling of verbatim conversations and flipping notebook pages will notify participants that they are being observed which will eventually increase their anxiety. It is mandatory that the researcher summarises the notes, and incorporates the mental notes immediately after each observation period. Interpretations or tentative conclusions need to be drawn to the notes which will enable the researchers to formulate or add new points to be observed in the subsequent sessions (Ulin et al., 2004).
d. Documentary Research
It is a type of non-reactive observation method. In this method, the material which is examined has been already collected and reported by others for other purposes. However, it gives a lot of information about how people think and behave in natural settings with no outsider influence. The examples of documentary sources are hospital and clinic records, newspaper stories, radio and television shows etc. Qualitative researchers also use this method for secondary analysis of transcriptions from interviews or focus group discussion conducted in the past (Ulin et al., 2004).
Ways to Recording Observations
Participant observers are expected to participate, observe and record the observations. Information through observation cannot be trusted to memory. The observer needs to record information as soon as possible after the observation (Polit & Beck, 2014). The commonly used observational records are logs or field diary and field notes.
i) Logs or Field Diary
Log is a record to be used in the field to note the daily events and conversations. This can also be used for planning and making time schedule and for keeping track of expenses.
ii) Field notes
The researcher must have a practice of recording conversation and events once happened without any delay and also to set aside time for making field notes. Each note must start with the date, time, location, purpose of observation and the demographic data of participants being observed. After making the field notes, the researcher needs to develop working hypothesis, plan subsequent steps, revisit and revise the notes. As the observations are being recorded, the researcher should keep asking new questions, interpreting and reinterpreting the information he or she had heard and seen. Summaries of the transcriptions need to be entered into the notes if audiotape is used for collecting the data. When the volume of field notes increases, the researcher should keep organising the notes (Ulin et al., 2004).
There are two types of field notes, one is descriptive notes and the other is reflective notes. Descriptive notes will include objective descriptions of observed events and conversations, dialogue and context. Reflective notes are used to document researcher’s personal experiences, reflections and progress while in the field (Polit & Beck, 2014).
Advantages and Disadvantages of Observation Method
The observation method provides deeper and richer understanding of situations pertaining to human behaviors, hence used widely by researchers whose interest is to develop conceptualization of phenomena. The potential problems associated with observation are observer bias and observer influence (Polit & Beck, 2014).
2. In-depth Interviews
In in-depth interview method, the researcher/ interviewer is guided by a few broad topics. These interviews are basically an exchange between the one interviewer and one respondent. It is otherwise known as “conversational partnership” (Rubin & Rubin, 1995), a “social encounter” (Holstein and Gubrium, 1999) and conversation with a purpose (Burgess, 1984). The participants will be constantly encouraged to take an active role in discussions. The main goal or focus of this method is to collect information-rich data which calls for mental sharpness, sensitivity and practice of the researcher (Ulin et al., 2004).
How to frame Qualitative Questions in in-depth interview?
The questions need to be informal, non-judgemental, and open. The interviewer must speak clearly but casually and avoid any suggestions. The qualitative interviews comprise of three types of questions namely main questions, follow-up questions and probes (Rubin & Rubin, 1995).
The main questions must come from the themes and subthemes of the study topic. The questions should move from easy and least threatening to more complex and interesting issues. The most important questions will be asked from various angles and at different points in time. The follow-up questions will move the discussions into a deeper level. At this point, the participant understands that the interviewer is interested to listen to him and gather more information. A probe is also similar to follow-up questions which will further take the discussions into next level. There must be a balance between too much and too little information. Probing, when insufficient will lead to boredom, but when aggressive leads to intrusion (Rubin & Rubin, 1995).
For example in a study on parenting children with burns (Ravindran, Rempel & Ogilvie, 2013) the researcher started with the main question which was general question on the phenomenon: Can you tell me about your child’s burn injury?.
The follow up question included questions such as how they managed the wound at home and the probe was by asking what they meant by saying it was difficult. Probes are primarily used to get rich and in-depth responses which will bring forth implicit meanings related to an experience (Patton, 2002).
Stages of In-depth Interview
Rubin and Rubin (1995) described the following stages of in-depth interview in sequence:
1. Creating natural involvement: The researcher will begin the interview by starting an informal chat with the participants in order to set the stage for the conducive environment. Informed consent will be obtained after explaining about all the details of the study including the introduction of the researcher.
2. Encouraging conversational competence: The interview will start with easy and non-threatening questions which will allow the participants to feel secure and enable them to give information freely. The researcher should use the language in which participants find it easy to share their experiences as a partner rather than an interviewer.
3. Showing understanding: The researcher will encourage the participants to give more in-depth details by letting them know that he or she understands and feels empathetic with the responses of the participants. The main focus is to put the participants at ease so as to allow them to share their experiences without fear.
4. Getting facts and basic description: As the conversation between the researcher (partner) and the participants progresses, the researcher slowly gets into the heart of the interview. During this stage, the participants will be asked to describe an incident with respect to the topic.
5. Asking difficult questions: Once the researcher is sure of the trusting relationship with participants, he can ask the participants regarding most difficult questions. The difficult questions can be repeated at different time points also which will offer opportunities for the participants to answer from various angles. Before asking the difficult questions, the researcher can remind the participants of the fact that the interview data will be kept confidential.
6. Toning down the emotional level: The researcher must help in restoring the sense of privacy among the participants particularly if the conversation is on sensitive issues. The researcher will ask the participants whether they want to ask anything to the researcher.
7. Closing while maintaining contact:At the end of the interview conversation, the participants will be thanked for spending their time and sharing their valuable experiences. The investigator will also reiterate on confidentiality of the record. Although the interview is over, the informal conversations between the researcher and the participants may continue.
These steps are very critical to be followed while collecting qualitative data using the in-depth interview.
Strengths and Limitations of In-depth Interviews
The strengths and weaknesses of Qualitative Interviewing is elaborated by Holloway (2005). In interviews, participants’ own words are captured and the researcher is also able to clarify any doubts immediately. Researchers are also able to get extensive data on the concept that is emerging by asking additional probing questions either in the same interview or in the subsequent interviews. Nonverbal behaviours also may be observed and noted.
The major limitation of interviews is that they are time consuming. Another vital point to be noted is that the narration is a reconstruction of facts rather than the actual behaviour. Interference from noise and other people in the area where the interviews happen also can have an adverse effect on the quality of data.
3. Focus Groups/ Group Discussion
Morgan (1988), defines focus group as “use of group interaction to produce data and insights that would be less accessible without the interaction found in a group.” Focus groups depend on exchange of ideas among participants to specific questions asked by the interviewer.
Like in-depth interviews, in this method also, three types of questions are used to collect data: the main questions, follow-up questions and probe questions. Focus groups will be effective in giving rich information over other methods especially when the researcher wants to do study about topics concerning social norms, expectations, values and beliefs. As far as the size and composition of the focus groups is concerned, it is imperative to include participants with similar characteristics with respect to the study topic. For instance, if the topic is on coping of mothers who had still births, participants need to have the experience of having had a still birth (Ulin et al., 2004).
Recruitment of appropriate number of participants for a focus group is yet another important factor in qualitative studies. Each group must have at least six to ten participants to enhance or facilitate the flow of discussion efficiently by the members and to manage the group by the moderator. Anonymity of the participants is important which will encourage them to speak without inhibition ( Morgan, 1995).
The number of focus groups required depends upon the number of defining demographic variables. For each variable, two focus groups can be conducted. In a day, one group with two-hour discussion including transcription can be managed which will be of around 25 to 40 pages of transcript (Ulin et al., 2004).
Procedure to Conduct Focus Groups
- The moderator welcomes the participants and introduces herself or himself and the note taker and then explains about the role of each and the purposes of focus group.
- The moderator assures participants that anonymity and confidentiality of information shared will be maintained.
- Informed consent will be obtained.
- The ground rules to be followed during the focus groups will be explained to the participants e.g., speaking one at a time, not interrupting each other.
- The participants are encouraged to speak freely.
- The first topic will be introduced slowly.
- The moderator must be in a position to create a group of conversational partners and he/she should have good interpersonal skills.
- The moderator should have good listening skills to listen to the conversation among participants carefully and non-judgementally while still keeping the discussion moving and focused. Audio recording or videotaping of discussions will ensure accurate data.
- There should be two moderators for efficient focus groups, out of whom, one will guide the discussion and the other will monitor the recording and note taking.
- The moderator asks the participants to summarise the points what they shared or the moderator himself can summarize in order to clarify issues before ending the discussion.
- The participants are given chance to make sure that the shared information has been understood correctly and recorded accurately.
- A feedback from participants may be collected if preferred.
Apart from all the steps mentioned above, the researcher should see that brief background information about the participants is collected to help the researcher to describe the participants, interpret what the participants have said, and evaluate the emerging themes. The background information should be relevant and brief and to be kept confidentially with coding (Basch, 1987; Ulin et al., 2004)
Advantages and Limitations of Focus Group Interviews
Focus group interviews are cost- effective and need less time to generate data as many participants’ views on a single topic can be gathered in a few meetings. As participants interact within the group the quality and richness of data is enhanced. Further, differences and similarities in participants’ views can be immediately assessed as the interview is conducted. Participants may also feel motivated by the interactions. It provides real life data including cultural beliefs (Patton, 2002; Manoranjitham, 2007).
The major drawback is the need for expertise in conducting the focus group discussions effectively. If the researcher is not proficient enough to moderate the discussion and keep the conversation flowing, in-depth data may not be generated and therefore time will be wasted. Each individual may not get adequate time to express all their views as every participant should be given time to share their own. Keeping confidentiality can be difficult and sensitive issues may not be discussed openly. Participants may fail to appear for the interviews and if enough number of participants are not available a focus group interview may not be possible (Patton, 2002). The researcher sometimes may directly get into the topic without really checking about the interest of the participants (Kitzinger, 1995).
| Conclusion|| |
There are some similarities with regard to sampling techniques used among qualitative traditions such as use of small sample size, and non-random sampling. However, sampling may vary depending upon the study methodology. The researchers’ knowledge and experience play a major role in determining the sample size. Data collection strategies in qualitative research are flexible in nature. Qualitative researchers should focus on the quality data rather than quantity. Developing and maintaining a trustworthy relationships with participants is essential in qualitative data generation.
Conflicts of Interest: The author has declared no conflicts of interest.
| References|| |
Basch, C. E. (1987). Focus group interview: An underutilized research technique for improving theory and practice in health education. Health Education Quarterly
Burgess, R. G (1984). In the field: An introduction to field research.
London: Unwin Hyman.
Burns, N., & Grove, S. K. (2004). The practice of nursing research: Conduct, critique, and utilization
(5th ed.). Philadelphia: Saunders.
DeWalt, K. M., & DeWalt, B. R. (2002). Participant observation: A guide for field workers
. Walnut Creek, CA: AltaMira Press.
Glaser, B. G. (1978). Theoretical sensitivity: Advances in the methodology of grounded theory.
Mill Valley, CA: Sociology Press.
Holloway, I. (2005). Qualitative research in health care
. Berkshire: England, Open University press.
Holstein, J. A., & Gubrium, J. F. (1997). Active interviewing in D. Silverman (Ed.), Qualitative research: Theory, method and practice
. London: Sage.
Kitzinger, J. (1995). Introducing focus group. British Medical Journal, 311
(7000), 299 - 302.
Manoranjitham, S., & Jacob, K. S. (2007). Focus group dicussion. The Nursing Journal of India, XCVIII(6)
, 125 -127.
Morgan, D. (1988).. Focus groups as qualitative research.
Patton, M. Q. (2001). Qualitative research and evaluation methods
(2nd ed.). Thousand Oaks, CA: Sage.
Patton, M. Q. (2002). Qualitative research and evaluation methods
(3rd ed.). Thousand Oaks, CA: Sage.
Polit, D. F., & Beck, C. T. (2014). Nursing research: Generating and assessing evidence for nursing practice
(9th ed.). Philadelphia: Wolters Kluwer, Lippincott Williams & Wilkins.
Rossman, G. B., & Rallis, S. F. (1998). Learning in the field: An introduction to qualitative research
(3rd ed.). Thousand Oaks, CA: Sage.
Rubin, H. J., & Rubin, I. S. (1995). Qualitative Interviewing: The art of hearing data
. Thousand Oaks, CA: Sage.
Ravindran, V., Rempel, G., & Ogilvie, L. (2013). Parenting burn injured children in India: A grounded theory study. International Journal of Nursing Studies
, 50(6), 786- 796. doi: 10.1016/j .ijnurstu.2012.06.011