Full text

Abstract

Qualitative sampling methods differ from quantitative sampling methods. It is important that one understands those differences, as well as, appropriate qualitative sampling techniques. Appropriate sampling choices enhance the rigor of qualitative research studies. These types of sampling strategies are presented, along with the pros and cons of each. Sample size and data saturation are discussed.

Qualitative research findings are useful to inform practice and policy. In the last issue of the Journal of Human Lactation, Berndt (2020) discussed sampling methods for quantitative studies. In this issue, I discuss sampling methods for qualitative research. Qualitative methods differ from quantitative methods in many ways including philosophical underpinnings, use of theory, research designs, and much more. Sampling strategies are also different.

The purpose of qualitative research is to discover meaning grounded in human experience (Sandelowski, 2004). A number of qualitative research methods exist, for example, phenomenology, ethnography, grounded theory, or narrative inquiry. Sampling strategies and sample sizes may differ based on which qualitative approach is utilized.

Sampling methods, while rigorous, are different from those used in quantitative research. All qualitative samples are non-random; only non-probability sampling methods are used. The qualitative researcher interviews people (participants) who can help the researcher understand the study phenomenon. They may observe situations or settings, and/or review documents, artifacts, photographs and/or drawings as possible data sources.

The qualitative researcher identifies participants who can provide information (data) to answer the research question. Participants who have “direct and personal knowledge” (Sandelowski, 1995, p. 180) of the study topic and are able to share and reflect on the experience of interest are recruited. Additionally, participants must be willing to spend the time necessary to share their experiences (Morse, 1998).

The process of sampling may change as the research process is iterative. Participants are identified, data are collected, and the data are analyzed. The sampling approach is refined as data collection and analysis proceed (Farrugia, 2019).

Types of Sampling

Common qualitative sampling methods are convenience, also called volunteer sampling, snowball, purposive, and theoretical sampling. Qualitative researchers may use more than one sampling approach in their study. Table 1 presents common sampling strategies, definitions, and pros and cons for each strategy.

Table 1 Sampling Methods With Definitions. 

Note. Adapted from “Nursing research: Generating and assessing evidence for nursing practice” (10th ed.), by Polit and Beck (2017).

Sample Size

Qualitative researchers must pre-determine the sample size for their study to satisfy human subjects’ review or ethics committees, grant proposals, and/or funding agencies (Young & Casey, 2019). While there are no specific “rules” for determining sample size, researchers must collect enough quality data to answer the research question. Often, the pre-determined sample size is not feasible or the area of study requires additional participants; it is understood that the desired sample size may change as the research progresses.

Generalizability, important for quantitative research, is not a goal of qualitative research; therefore sample sizes are much smaller than those needed for quantitative designs. Qualitative researchers typically use large samples, determined by a power analysis, while qualitative samples are smaller in order to examine a phenomenon in depth.

The goal of qualitative sampling is to recruit enough participants and/or observations that provide rich, in-depth data, in order to understand the phenomenon studied (Hennink et al., 2019). “Sample size in qualitative research may refer to numbers of persons, but also to numbers of interviews and observations conducted or numbers of events sampled” (Sandelowski, 1995, p. 180). Sample size must be sufficient to generate quality data that provides a rich understanding of the experience (Sandelowski, 1995). The goal of qualitative sampling is to choose enough participants and/or observations that provide rich data in order to understand the phenomenon studied (Hennink et al., 2019).

Determining an adequate sample size in qualitative research is ultimately a matter of judgement and experience in evaluating the quality of the information collected against the uses to which it will be put, the particular research method and sampling strategy employed, and the research product intended. (Sandelowski, 1995, p. 183)

Qualitative researchers may find that few guidelines exist to determine how best to ascertain an adequate sample size, although recently some authors have posited strategies to determine sample size a priori (Sim et al., 2018). Vasileiou et al., 2018 systematically analyzed interview-based qualitative studies over 15 years reported in three journals: British Medical Journal (BMJ), British Journal of Health Psychology (BJHP), and Sociology of Health & Illness (SHI), to determine how sample sizes were characterized and justified by the study authors. Twenty-one articles from BMJ, 53 articles from BJHP, and 140 articles from SHI were included in their analysis. They found that sample size sufficiency was not reported in the majority of the articles and that sample size justification was not based on the number of interviews. Most often, authors justified sample size using the principle of data saturation.

Data Saturation

Data saturation is linked to sampling strategies and sample size; sufficient sample sizes are necessary for quality data. Qualitative researchers often report that they have collected enough data to “reach saturation,” although confusion about the meaning of saturation is common. Data saturation occurs when no new information is obtained from interviews and/or observations (Morse, 1995).

Estimating the number of participants in a study required to reach saturation depends on a number of factors, including the quality of data, the scope of the study, the nature of the topic, the amount of useful information obtained from each participant, the number of interviews per participant, the use of shadowed data and the qualitative method and study design used. (Morse, 2000, p. 3)

Studies with a broad scope may require more participants or observations; clear topics require fewer participants. Some participants are able to share eloquently about the phenomenon under study while others are not. When participants provide quality data, fewer participants are needed. Few researchers provide details about how they reached data saturation.

Conclusion

Authors of qualitative studies must convince the reader that the sample was appropriate and sufficient to justify the findings. Minimally, the author should report how the sample was selected, the number of participants and/or observations with a rationale for that number, and the participant characteristics. If the author reports data saturation, they must provide evidence as to how they determined data saturation was achieved.

THE END

来源 | 健康界

版权归原作者所有，若有违规、侵权请联系我们