Sampling in Qualitative Research

Qualitative sampling is a purposeful sampling technique in which the researcher sets a criteria in selecting individuals and sites. The major criterion used in selecting respondents or sites is the richness of information that can be drawn out from them. There are several strategies under this sampling technique.

Extreme or deviant case sampling focuses at highly unusual manifestations of the phenomenon of interest. This strategy tries to select particular cases that would gather the most information about a given research topic. For example, in a group of patients in a psychiatric ward, a researcher might choose to include in his/her samples those who have extreme cases of schizophrenia or those who have suicidal tendencies other than. those who have common problems of depression or anxiety.

Intensity sampling involves information-rich cases that manifest the phenomenon intensely, but not extremely (e.g., good students and poor students, above average and below average). The strategy is also similar to extreme or deviant case sampling because it uses the same logic. The difference is that the cases selected are not as extreme. This type of sampling requires prior information on the variation of the phenomena under study so that intense, although not extreme, examples are chosen. If one were studying a particular emotion, say envy, the respondent to be selected should have had an intense experience with this particular emotion; a mild or extreme experience would not likely explain the phenomena in the same way as an intense experience would.

Maximum variation sampling selects a wide range of variation on dimensions of interest. The purpose is to discover or uncover central themes, core elements, and/or shared dimensions that cut across a diverse sample. It also provides an opportunity to document unique or diverse variations. For example, to implement this sampling, a matrix (of communities, people, etc.) is created where each item on the matrix is as different (on relevant dimensions) as possible from all other items.

Homogeneous sampling brings together people of similar backgrounds and experiences. It reduces variation, simplifies analysis, and facilitates group interview. This strategy is used most often when conducting focus groups. For example, in a study about a parenting program, all single-parent, female head of households are selected.

Typical case sampling focuses on what is typical, normal, and/or average. This strategy may be adopted when one needs to present a qualitative profile of one or more typical cases. In this sampling, a broad consensus is required. about what is “average.” For example, in a study that involves development projects in the Third World countries, a typical case sampling of “average” villages may be conducted. Such study would uncover critical issues to be addressed for most villages by looking at the samples.

Critical case sampling looks at cases that will produce critical information. In order to use this method, what constitutes a critical case must be known. This method permits logical generalization and maximum application of information to other cases because what is true to one case should also be true to all other cases. For example, if people’s understanding of a particular set of federal regulations is desired, the regulations may be presented to a group of highly educated people (“If they cannot understand them, then most people probably cannot”) and/or to a group of undereducated people (“If they can understand them, then most people probably can”).

Snowball or chain sampling is done by asking relevant people if they know someone or somebody fitted or is willing to participate in a study. For instance, a researcher will use the social media to ask people if they can refer persons who meet the criteria to become the respondents for the research study. From these nominations, the researcher would select participants to be included as members of the sample.

Criterion sampling selects all cases that meet some predetermined criterion. This strategy is typically applied when considering quality assurance issues. In essence, cases which are information-rich and which might reveal a major system weakness that could be improved may be chosen. For example, the average length of stay of a patient who have undergone a certain surgical procedure is three days. Any patient who have undergone the same surgical procedure and whose stay exceeded three days may be set as a criterion of becoming a sample in the study. Interviewing these cases may offer information related to aspects of the process/system that could be improved.

Operational construct or theoretical sampling identifies manifestations of a theoretical construct of interest to elaborate and examine the construct. This strategy is used in grounded theory studies in which people or incidents are sampled based on whether or not they manifest an important theoretical or operational construct. For example, if the theory of resiliency in adults who were physically abused as children is studied, people who meet the theory-driven criteria for resiliency are selected.

Confirming and disconfirming sampling seek cases that are both “expected” and the “exception” to what is expected. This strategy deepens initial analysis, seeks exceptions, and tests variation. Both confirming cases (those that add depth, richness, credibility), as well as disconfirming cases (those that do not fit and are the source of rival interpretations), must be found. This strategy is typically adopted after an initial fieldwork has established what a confirming case would be. For example, in a study about factors affecting academic performance, the researcher must present cases of positive and negative factors that affect students’ achievement. In this case, good school facilities and low teacher-to-student ratios may be positive factors while low parental involvement and low economic status may be negative factors affecting academic performance.

Stratified purposeful sampling focuses on characteristics and comparisons of particular subgroups of interest. This strategy is similar to stratified random sampling (samples taken within samples), except that the sample size in the former is typically much smaller. In stratified sampling, samples based on a characteristic are stratified. Thus, in conducting a study about academic performance, the samples are clustered into below average, average, and above-average learners. The main goal of this sampling is to capture major variations (although common themes may emerge).

Opportunistic or emergent sampling follows new leads during fieldwork, takes advantage of the unexpected, and is flexible. This strategy takes advantage of whatever is readily available for the researcher and considers other samples that may be useful for the researcher as they come. For example, in studying sixth-grade pupils’ awareness of a topic, it may be advisable to include fifth-grade pupils as well to gain additional understanding.

Purposeful random sampling looks at a random sample and adds credibility to a sample when the potential purposeful sample is larger than one can handle. While this is a type of random sampling that uses small sample sizes, its goal is to increase credibility, not to encourage representativeness or the ability to generalize. For example, if clients at a drug rehabilitation program is studied, 10 out of 300 current cases may be selected. This reduces judgment within a purposeful category because the cases are picked randomly and without regard to the program outcome.

Convenience sampling selects cases based on ease of accessibility. This strategy saves time, money, and effort, however, it has the wefkest rationale and the lowest credibility. This strategy may yield information-poor cases because they are picked simply because they are accessible, rather than based on a specific strategy/rationale. Sampling co-workers, family members, or neighbors simply because they are easily accessed by the researcher is an example of conve
nience sampling.

The combination or mixed purposeful sampling combines two or more sampling techniques. Basically, using more than one strategy is considered combination or mixed purposeful sampling. This type of sampling meets multiple interests and needs. For example, snowball or chain sampling may be used to identify extreme or deviant cases. That is, people may be asked to identify cases that would be considered extreme/deviant and do this until a consensus on a set of cases needed is attained.

How do we determine the number of samples needed in the study? When determining the sample size for qualitative research, it is important to remember that there are no hard and fast rules. There are, however, at least two considerations on sample size in a qualitative investigation:

  1. What sample size will reach saturation or redundancy? That is, how large does the sample need to be allowed for the identification of consistent patterns? Some researchers say the size of the sample should be large enough to leave the researcher with “nothing left to learn.” In other words, one might conduct interviews, and after the tenth one, realize that there are no new concepts emerging. That is, the concepts, themes, etc. begin to be redundant.
  2. How large a sample is needed to represent the variation within target population? That is, how large must a sample be in order to assess an appropriate amount of diversity or variation represented in the population of interest?

The sample size is estimated based on the approach used in the study or the data collection method employed. However, experts prescribe numbers for sample sizes in some qualitative research studies. Cited from Creswell (2013), one to ten subjects are recommended for phenomenology (Dukes, 1984), 20 to 30 individuals for grounded theory (Charmaz, 2006), four to five respondents for case study, and single culture-sharing groups for ethnography.