Qualitative Sample Size And Sampling Strategy Essay
In research, sampling describes the selection of the units, individuals, and settings under investigation. Quantitative studies, primarily, include random sampling. In contrast, qualitative studies strive for purposeful sampling, which involves a sample relevant to a particular research question. If one were studying the adult survivors who faced childhood sexual abuse, for example, examining a random sample of ten people may indicate only one survivor. Consequently, the researchers will have a sample size of one individual and should continue the random sampling until they reach a suitable number of adult survivors. Such a method, however, is time-consuming and ineffective. Thus, the primary difference in the sampling approaches between qualitative and quantitative studies results from their different objectives (Luborsky & Rubinstein, 1995). Typically, qualitative investigations begin with a particular group or event. In the study on the adult survivors, the investigator should choose the sample purposefully and include only the individuals who experienced the sexual abuse. Hence, the qualitative research’s goal involves an in-depth understanding of the settings under study.
Some features of quantitative sampling are sometimes relevant to a qualitative investigator. For example, a qualitative study on schoolchildren’s experiences of the Hurricane Katrina may investigate 3,000 children, who experienced the cyclone. In such a study, one may randomly sample ten of the 3,000 children. Similarly, an ethnographic survey research may require the researcher to obtain sample sizes that are similar to the sizes used in quantitative designs. Therefore, ambiguities may occur when choosing a suitable sample size or sampling strategy. Nevertheless, the research purpose, study period, and available resources should drive the adopted strategy.
Usually, qualitative researchers combine different strategies with the aim of establishing an efficient approach. The investigator, for example, may utilize snowball sampling in the identification of deviant or extreme cases. Such an approach is a case of mixed purposeful sampling. Sampling strategies, therefore, are not mutually exclusive, and a qualitative research design may include a range of strategies.
According to Patton (2001), Deviant or Extreme Case Sampling examines unusual manifestations of a particular phenomenon. Phenomena of interest may include an outstanding success, crises, notable failures, school dropouts, and exotic events. The approach attempts to select specific cases that may gather relevant and adequate information. An example of a deviant or extreme case may involve the sexually abused women who eventually kill their abusers. Patton (2001) explains that Chain or Snowball Sampling, nonetheless, determines cases of interest by interviewing different people. The strategy also identifies reliable participants who can engage in interviews. Hence, this approach locates the information-rich cases. During Snowball sampling, the researcher may request for the identification of individuals proficient in a particular issue (Patton, 2001). For example, the investigator may request for nominations that may result in the repeated appearance of key names. A combination of Snowball Sampling and Extreme Case Sampling is an efficient sampling strategy for an investigation of the survivors of childhood abuse.
Marshall (1996) argues that certain issues may affect the selection of sample sizes. Such factors include sampling errors, bias, inefficiency of investigators in the observation and interpretation of behavior, and statistical normal distribution. Moreover, variability in the information derived from potential test cases may influence the choice of sample sizes.
In the determination of sample size, the qualitative investigator can make two considerations. First, the sampling expert should identify and use a sample size that can reach redundancy or saturation. Thus, the need for consistent patterns determines the required sample size. Secondly, the investigator should consider the sample size needed to represent the variations in a particular population adequately. Hence, the appropriate sample size should allow the assessment of a suitable and representative amount of variation.
Marshall (1996) urges the use of small sample sizes in the studies that involve qualitative research plans. While citing different researchers, Marshall (1996) states that qualitative researchers sometimes select large sample sizes. He observes that such a selection is inappropriate because it considers generalization as the goal of excellent research (Marshall (1996). Such misapprehension is the primary cause for the utilization of incorrect sampling strategies. According to Marshall (1996), the enhanced understanding of different and complex human issues is better than the generalization of results. Therefore, a suitable sample size for a qualitative investigation should provide an adequate answer to a given research question. Small samples may be appropriate for detailed studies or simple questions. In contrast, complex issues may require relatively large samples, as well as several sampling techniques. Nevertheless, the number of the required subjects often becomes apparent as a qualitative study progresses. In addition, as new categories and explanations cease emerging from a particular data, a state of data saturation is achieved (Marshall, 1996). Such a situation further clarifies the required sample size.
The selection of sample sizes utilized in qualitative studies is in stark contrast with the stepwise designs, as well as the sampling strategies, used in various quantitative research studies. According to Creswell (2009), some of the approaches for selecting suitable sample sizes for qualitative investigations include the use of convenience samples, theoretical samples (Denzin & Lincoln, 2000), and judgment samples. Moreover, the choice of qualitative sample sizes is iterative and may require flexibility. Usually, qualitative studies begin with a general idea regarding a particular population and some solicitations for the test cases. The subsequent test cases are then selected using the previous case selections. Samples can be adjusted as the investigation progresses because it gives the researcher a better conceptualizations of the research objectives. The sampling process then continues until a state of saturation occurs. Landreneau (2012) observes that the final sample may allow the confirmation or disconfirmation of specified cases.
A sample size selection can also occur based on the study approach employed. A research approach involving a biography or case study may require the selection of a particular case or person. However, an ethnography, grounded theory, or action research may require the assessment of 20-30 individuals. Typically, such a sample size is enough to reach a saturation level. Data collection techniques can also influence the selection of sample sizes. For example, interviews involving key informants may require a sample size of about five people. However, in-depth interviews often utilize about thirty informants. Researchers sometimes carry out ethnographic surveys while collecting data. In such investigations, one should choose a large representative sample. In addition, the research purpose should influence the selection of the sample size. Regardless of the approach that one adopts or the sample size that one uses, a rationale for the selected choices requires an enunciation of the expected weaknesses and benefits. Moreover, a significant aspect of qualitative research designs is flexibility. Consequently, if one’s research program includes a qualitative research design, one should have tolerance for ambiguity. The qualitative research plan investigating the survivors of childhood sexual assault may require in-depth interviews. Hence, a sample of thirty respondents would be appropriate for such a study.
The analysis of a research process requires the use of representative samples. If performed correctly, the samples may promote the efficiency of a research program. Samples, therefore, are very critical to study programs such as the qualitative investigations. Although the researcher may have an excellent research question, the sample dictates the nature of the inferences obtained from the study. A biased sample, for example, may cause invalid results. Therefore, several considerations are crucial to the determination of an adequate sampling strategy before commencing a research study. In addition, an appropriate sample size allows the elimination of biases in the sample. Thus, an effective sampling strategy and a suitable sample size enable the researcher to make accurate generalizations.
Creswell, J.W. (2009). Research design: Qualitative, quantitative, and mixed methods approaches (3rd ed.). Thousand Oaks, CA: Sage Publications.
Denzin, N.K., & Lincoln, Y.S. (2000). Handbook of qualitative research. Thousand Oaks, CA: Sage Publications.
Landreneau, K.J. (2012). Sampling strategies. San Francisco, CA: Natco.
Luborsky, M. R., & Rubinstein, R. L. (1995). Sampling in Qualitative Research: Rationale, Issues, and Methods. Research on Aging, 17(1), 89–113. Doi: 10.1177/0164027595171005
Marshall, M.N. (1996). Sampling for qualitative research. Family Practice, 13(6), 522-525. New York, NY: Oxford University Press.
Patton, M. Q. (2001). Qualitative evaluation and research methods (3rd ed.). Newbury Park, CA: Sage Publications.