Education Qualitative, Quantitative, and Mixed Methods Research Sampling Strategies
by
Timothy C. Guetterman
  • LAST MODIFIED: 26 February 2020
  • DOI: 10.1093/obo/9780199756810-0241

Introduction

Sampling is a critical, often overlooked aspect of the research process. The importance of sampling extends to the ability to draw accurate inferences, and it is an integral part of qualitative guidelines across research methods. Sampling considerations are important in quantitative and qualitative research when considering a target population and when drawing a sample that will either allow us to generalize (i.e., quantitatively) or go into sufficient depth (i.e., qualitatively). While quantitative research is generally concerned with probability-based approaches, qualitative research typically uses nonprobability purposeful sampling approaches. Scholars generally focus on two major sampling topics: sampling strategies and sample sizes. Or simply, researchers should think about who to include and how many; both of these concerns are key. Mixed methods studies have both qualitative and quantitative sampling considerations. However, mixed methods studies also have unique considerations based on the relationship of quantitative and qualitative research within the study.

Sampling in Qualitative Research

Sampling in qualitative research may be divided into two major areas: overall sampling strategies and issues around sample size. Sampling strategies refers to the process of sampling and how to design a sampling. Qualitative sampling typically follows a nonprobability-based approach, such as purposive or purposeful sampling where participants or other units of analysis are selected intentionally for their ability to provide information to address research questions. Sample size refers to how many participants or other units are needed to address research questions. The methodological literature about sampling tends to fall into these two broad categories, though some articles, chapters, and books cover both concepts. Others have connected sampling to the type of qualitative design that is employed. Additionally, researchers might consider discipline specific sampling issues as much research does tend to operate within disciplinary views and constraints. Scholars in many disciplines have examined sampling around specific topics, research problems, or disciplines and provide guidance to making sampling decisions, such as appropriate strategies and sample size.

Sampling Strategies

Patton 2015 gives an overview of different sampling strategies and options, which provides a good starting point on purposive sampling. To learn more about strategies, an entire book, Emmel 2013, provides a comprehensive treatment of sampling strategies and how to think about qualitative sampling using a realist approach. The sampling frame is also important. Devers and Frankel 2000 describes purposive sampling strategies and provides advice on deciding a sampling frame. In addition, Tuckett 2004 considers the sampling frame, selecting and gaining access to sites, selecting participants, and saturation. Another consideration is the type of qualitative data collected. Specific to interview studies, Robinson 2014 provides practical guidance that researchers can use to develop a sampling plan. Singh, et al. 2011 discusses large-scale qualitative data, (e.g., customer reviews), but may be increasingly relevant as researchers mine large text data sources, such as social media data. Two specialized sampling strategies are experience sampling, as described in Koro-Ljungberg, et al. 2008, and snowball sampling, covered in Noy 2008.

  • Devers, K. J., and R. M. Frankel. 2000. Study design in qualitative research–2: Sampling and data collection strategies. Education for Health 13.2: 263–271.

    DOI: 10.1080/13576280050074543Save Citation »Export Citation »

    This article, part of a series on qualitative research design, reviews how to determine a sampling frame, guided by purposive sampling strategies. They discuss considerations in gaining access to sites and individuals.

    Find this resource:

    • Emmel, N. 2013. Sampling and choosing cases in qualitative research: A realist approach. London: SAGE.

      DOI: 10.4135/9781473913882Save Citation »Export Citation »

      This book provides one of the most comprehensive discussions of sampling strategies used in qualitative research. Situated within a critical realist paradigm, the principles may be widely applicable. He discusses four approaches: (1) theoretical—driven by theory development; (2) purposeful—driven by the pragmatic need for information rich cases; (3) theoretical and purposive—driven by analytical induction and the researcher; and (4) realist—guided by realism and drawing from the other approaches.

      Find this resource:

      • Koro-Ljungberg, M., R. Bussing, P. Williamson, and F. M’Cormack-Hale. 2008. Reflecting on the experience sampling method in the qualitative research context: Focus on knowledge production and power during the data-collection process. Field Methods 20.4: 338–355.

        DOI: 10.1177/1525822X08320201Save Citation »Export Citation »

        This article discusses the experience sampling method and how it can be adapted for qualitative research. Experience sampling methods include a variety of approaches, such as diary methods and ecological momentary assessment and share common features in that participants report in their natural setting, in real time, and on multiple occasions. They acknowledge the tension between the researcher-prompted recording of data with the technique and qualitative participant-driven data production in addition to adjustments needed.

        Find this resource:

        • Noy, C. 2008. Sampling knowledge: The hermeneutics of snowball sampling in qualitative research. International Journal of Social Research Methodology 11.4: 327–344.

          DOI: 10.1080/13645570701401305Save Citation »Export Citation »

          This article reviews snowball sampling, which is a method of sampling in which contact information about potential research participants is provided by other participants. Noy discusses two qualitative studies relying on snowball sampling to reach and keep track of participants who were frequently moving about (e.g., backpackers, drivers). Unique implications of snowball sampling include power relations, social networks, and social capital.

          Find this resource:

          • Patton, M. Q. 2015. Qualitative research and evaluation methods: Integrating theory and practice. 4th ed. Thousand Oaks, CA: SAGE.

            Save Citation »Export Citation »

            As with many popular qualitative textbooks, Patton has devoted a portion of a chapter to sampling. However, Patton’s book is noteworthy because the majority of Chapter 5, “Designing Qualitative Studies,” is about sampling. Through multiple sections within the chapter, he gives substantive treatment to purposeful sampling, various sampling strategies with their options, and sample size.

            Find this resource:

            • Robinson, O. C. 2014. Sampling in interview-based qualitative research: A theoretical and practical guide. Qualitative Research in Psychology 11.1: 25–41.

              DOI: 10.1080/14780887.2013.801543Save Citation »Export Citation »

              In this article, Robinson presents four procedures for sampling in qualitative research that uses interviews as the data collection method. The four points are (1) defining the sample universe that includes who could potentially be included and who would be excluded; (2) determining a sample size, which may be a range; (3) deciding a sampling strategy, such as probability, convenience, or a specific type of purposive sampling; and (4) sourcing the sample, which involves recruiting participants.

              Find this resource:

              • Singh, S. N., S. Hillmer, and Z. Wang. 2011. Efficient methods for sampling responses from large-scale qualitative data. Marketing Science 30.3: 532–549.

                DOI: 10.1287/mksc.1100.0632Save Citation »Export Citation »

                This article discusses large-scale qualitative data (e.g., online customer reviews, message boards, blogs, discussion boards) that are often overlooked in the sampling literature. The authors present four sampling options: simple random sampling, stratified random sampling, an approach based on the largest number of information units, and a sequential random sampling. They evaluate the four options using simulations, discuss selecting a method, and apply the approach to large-scale survey data and their dataset of customer reviews.

                Find this resource:

                • Tuckett, A. G. 2004. Qualitative research sampling: The very real complexities. Nurse Researcher 12.1: 47–61.

                  DOI: 10.7748/nr2004.07.12.1.47.c5930Save Citation »Export Citation »

                  Tuckett provides a general discussion of qualitative sampling. He discusses considerations for the sampling frame, selecting and gaining access to sites, selecting participants, and saturation. The article covers both individual and group (e.g., focus group) interviews. It concludes with helpful practical and logistical issues.

                  Find this resource:

                  Sample Size

                  “How many?” is a common question in qualitative research design. Researchers developing proposals wonder how many is enough, and reviewers look for whether the sample size is appropriate. The question, however, is complicated and could be considered from a few different angles including the complexity of the central phenomenon being studied, the concept of saturation, the type of methodology used, and the methods of data collection. Malterud, et al. 2015 provides a multidimensional way to assess information power in qualitative research in order to guide sample size determinations. Saturation is the point at which researchers are not adding new understanding or complexity to findings as they continue to collect new data. Fusch and Ness 2015 provides a concise introduction to saturation. Connecting saturation to sample size, Mason 2010 grounds its recommendations in an empirical examination of sampling in PhD studies. Currently, the preponderance of guidance about qualitative sample size is about the use of interviews as the primary method of data collection. A helpful starting point for considering sample size in interviews is Guest, et al. 2006, a highly cited article that provides an empirical rationale for how many interviews are needed to reach saturation. More recently, Guest, et al. 2017 turns to studies employing focus groups and discusses sample size considerations.

                  • Fusch, P. I., and L. R. Ness. 2015. Are we there yet? Data saturation in qualitative research. Qualitative Report 20.9: 1408–1416.

                    Save Citation »Export Citation »

                    Fusch and Ness provide a brief overview of data saturation, including its relationship with triangulation. They discuss different methods of data collection and stress collecting rich and thick data. The article concludes with two example studies along with a critical appraisal of data saturation in each.

                    Find this resource:

                    • Guest, G., A. Bunce, and L. Johnson. 2006. How many interviews are enough? An experiment with data saturation and variability. Field Methods 18.1: 59–82.

                      DOI: 10.1177/1525822X05279903Save Citation »Export Citation »

                      This highly cited article reports a methodological study examining the point of data saturation in qualitative interview-based research. In their experiment, they found saturation of themes was achieved with n = 12 semistructured interviews though metathemes were saturated as early as six interviews. Much of the article focuses on determining data saturation using purposive sampling, and it explains important considerations of interview structure, narrowness of the content, homogeneity of the sample, and the extent to which themes are finely grained.

                      Find this resource:

                      • Guest, G., E. Namey, and K. McKenna. 2017. How many focus groups are enough? Building an evidence base for nonprobability sample sizes. Field Methods 29.1: 3–22.

                        DOI: 10.1177/1525822X16639015Save Citation »Export Citation »

                        Guest and colleagues (2017) conducted a similar methodological experiment about sample size in research relying on focus groups. Using thematic analysis of forty focus groups, they achieved saturation in identifying 80 percent of themes in two to three focus groups and 90 percent within three to six focus groups. They discuss five important considerations: (1) focus group instrument structure [e.g., semistructured], (2) sample homogeneity, (3) topic complexity, (4) purpose and granularity of themes sought, and (5) granularity in coding.

                        Find this resource:

                        • Malterud, K., V. D. Siersma, and A. D. Guassora. 2015. Sample size in qualitative interview studies: Guided by information power. Qualitative Health Research 26.13: 1753–1760.

                          DOI: 10.1177/1049732315617444Save Citation »Export Citation »

                          This article discusses the concept of information power as a way to determine an adequate sample size in qualitative studies using interviews. As information power increases, the sample size required decreases. Conversely, larger samples are needed when information power is weaker. Assessment of information power of a sample has five dimensions: (1) the aim of the study, (2) sample specificity, (3) use of established theory, (4) quality of dialogue, and (5) analysis strategy.

                          Find this resource:

                          • Mason, M. 2010. Sample size and saturation in PhD studies using qualitative interviews. Forum: Qualitative Social Research 11.3.

                            DOI: 10.17169/fqs-11.3.1428Save Citation »Export Citation »

                            Mason’s highly cited article examined 561 qualitative PhD studies using interviews and found the mean number of interviews was thirty-one, though many were considerably larger. He concluded that many PhD researchers are not using the concept of saturation skillfully, which inflates samples. Mason’s results further suggested that some may conduct large numbers of interviews to ensure the PhD is defensible. He concludes with practical recommendations, returning to the principles of qualitative research and saturation.

                            Find this resource:

                            Qualitative Design Considerations

                            Sampling and sample size has been linked to whatever type of qualitative design a researcher uses. Both Creswell and Poth 2018 and Morse 1994 provide general recommendations for sample size based on design. Notably, Morse 2000 later clarifies with more factors to consider. Other authors have provided detailed discussions of sampling strategies and sample size based on literature. In particular, Gentles, et al. 2015 provides an overview of sampling strategies for three popular qualitative designs. Guetterman 2015 accounts for the methodology used and provides recommendations for sample size based on an empirical study of published articles that used five different qualitative designs after broadly looking at top cited articles in health and educational sciences.

                            • Creswell, J. W., and C. N. Poth. 2018. Qualitative inquiry and research design: Choosing among five approaches. 4th ed. Thousand Oaks, CA: SAGE.

                              Save Citation »Export Citation »

                              This popular textbook on quality inquiry includes a brief discussion of sample size. However, it is often cited in determining sample size based on the type of qualitative design: grounded theory, phenomenology, case study, ethnography, and narrative research. The recommendations are general guidelines, often presenting a range.

                              Find this resource:

                              • Gentles, S. J., C. Charles, J. Ploeg, and K. A. McKibbon. 2015. Sampling in qualitative research: Insights from an overview of the methods literature. Qualitative Report 20.11: 1772–1789.

                                Save Citation »Export Citation »

                                This article summarizes qualitative sampling in the research methods literature. What is very helpful is the authors’ comparison and discussion of sampling for three major qualitative designs: grounded theory, phenomenology, and case study. For each design, they break down sampling discussions by the overall definition, sampling strategies (e.g., purposeful, theoretical), sampling units, saturation, sample size, and timing sampling decisions.

                                Find this resource:

                                • Guetterman, T. C. 2015. Descriptions of sampling practices within five approaches to qualitative research in education and the health sciences. Forum: Qualitative Social Research 16.2.

                                  DOI: 10.17169/fqs-16.2.2290Save Citation »Export Citation »

                                  Guetterman reviewed fifty-one highly cited articles using five major approaches to qualitative research (case study, grounded theory, ethnography, phenomenology, narrative research) in health sciences and education, fields where qualitative research is popular. He reviewed methodological details and how authors approached sampling. When planning a study, researchers should determine an appropriate sample size both from a research quality perspective and a time and resource perspective. Adequacy and representativeness of the sample are important considerations. Available online.

                                  Find this resource:

                                  • Morse, J. M. 1994. Designing qualitative research. In Handbook of qualitative inquiry. Edited by N. K. Denzin and Y. S. Lincoln, 220–235. Thousand Oaks, CA: SAGE.

                                    Save Citation »Export Citation »

                                    In this chapter, Morse provides general recommendations for the minimum sample size for phenomenology, ethnography, grounded theory, and ethnoscience. The book is often cited, but Morse 2000 clarified her position on sample size and the multitude of characteristics.

                                    Find this resource:

                                    • Morse, J. M. 2000. Determining sample size. Qualitative Health Research 10.1: 3–5.

                                      DOI: 10.1177/104973200129118183Save Citation »Export Citation »

                                      In a follow-up brief editorial, Morse’s Qualitative Health Research offered very practical advice on sampling that clarified important factors to consider in addition to the design. She stresses considering the scope, the nature of the topic, data quality, and the use of shadowed data in addition to the study design.

                                      Find this resource:

                                      Discipline Specific and Special Considerations

                                      Researchers in different fields and disciplines have written about sampling considerations in studies involving qualitative and quantitative research approaches. Although much of methodology is discipline agnostic, and researchers can learn from reading outside of their discipline, sometimes there is utility to having resources to cite within one’s discipline. In a few instances, disciplines may have unique considerations, such as organizational research that might sample at the level of the organization and individuals within the organization. Also, research is often shaped by our discipline. Disciplines may form expectations, such as what constitutes an adequate sample size, and those expectations permeate journals in that field. Therefore, it is important to become aware of discipline specific sampling considerations when designing and writing qualitative research. In the geography of health or healthcare, Curtis, et al. 2000 examines sampling considerations. Higginbottom 2004 discusses sampling issues for nurse researchers, though the concepts are broadly applicable to other disciplines. Marshall, et al. 2013 examines sampling, focusing on sample size considerations in information systems research using interviews. Koerber and McMichael 2008 reviews sampling strategies for technical communicators, and turns to other disciplines, namely health, to guide readers. Finally, when conducting research that involves friends as participants, Brewis 2014 is a helpful resource.

                                      • Brewis, J. 2014. The ethics of researching friends: On convenience sampling in qualitative management and organization studies. British Journal of Management 25.4: 849–862.

                                        DOI: 10.1111/1467-8551.12064Save Citation »Export Citation »

                                        This article discusses the ethical implication of convenience sampling among friends in management and organizational studies research. The article argues that existing relationships with friend-respondents can increase depth but also brings ethical challenges and new responsibilities.

                                        Find this resource:

                                        • Curtis, S., W. Gesler, G. Smith, and S. Washburn. 2000. Approaches to sampling and case selection in qualitative research: Examples in the geography of health. Social Science in Medicine 50.7–8: 1001–1014.

                                          DOI: 10.1016/S0277-9536(99)00350-0Save Citation »Export Citation »

                                          This article applies the Miles and Huberman’s (Qualitative data analysis: An expanded sourcebook [2nd ed.]. Thousand Oaks, CA: Sage, 1994.) sampling framework to three studies focused on the geography of health or healthcare. After a review of principles of qualitative sampling, the authors provide a critical discussion of the sampling in the studies using criteria of relevance to conceptual framework, potential to generate rich information, analytic generalizability, potential for believable explanations, ethical issues, and feasibility.

                                          Find this resource:

                                          • Higginbottom, G. M. A. 2004. Sampling issues in qualitative research. Nurse Researcher 12.1: 7–19.

                                            DOI: 10.7748/nr2004.07.12.1.7.c5927Save Citation »Export Citation »

                                            This article about sampling practices is written for nurse researchers although the discussion can apply broadly across disciplines. It provides a brief overview of grounded theory, ethnography, and phenomenology along with their specific sampling considerations. To encourage transparency and rigor in qualitative research, the article covers qualitative sampling strategies (e.g., purposive, theoretical, maximum variation) along with their advantages and weaknesses.

                                            Find this resource:

                                            • Koerber, A., and L. McMichael. 2008. Qualitative sampling methods: A primer for technical communicators. Journal of Business and Technical Communication 22.4: 454–473.

                                              DOI: 10.1177/1050651908320362Save Citation »Export Citation »

                                              Though written as informational for technical and scientific communication scholars, this article provides an excellent accessible review of qualitative sampling methods. The authors succinctly summarize purposeful and theoretical sampling as presented by major texts in the field. Cognizant of gaps in their own field, they turn to health research and address four questions: what qualitative sampling is, what techniques are available, how to select a technique, and how to determine sample size.

                                              Find this resource:

                                              • Marshall, B., P. Cardon, A. Poddar, and R. Fontenot. 2013. Does sample size matter in qualitative research? A review of qualitative interviews in IS research. Journal of Computer Information Systems 54.1: 11–22.

                                                DOI: 10.1080/08874417.2013.11645667Save Citation »Export Citation »

                                                This article provides a discussion of sampling size in qualitative research in information systems. Authors reviewed actual research articles conducted in the field and describe sampling practices. In brief, they identified problems with sample size justification and urge researchers to include rigorous justifications.

                                                Find this resource:

                                                Sampling in Quantitative Research

                                                Sampling in quantitative research also involves the topics of sampling strategies, sample size, and other specific considerations. For the purpose of this article, sampling strategies refers to overall approaches to sampling. In quantitative research or quantitative strands of mixed methods studies, sampling typically involves probabilistic strategies, such as simple random sampling, stratified random sampling, or cluster sampling. When designing a sampling plan, researchers consider the overall strategy and a specific sample size calculation for the type of analysis planned. A few textbook-sized resources are dedicated to sampling strategies or sample size. Additional literature is available in articles or individual chapters within research methods and statistics books.

                                                Sampling Strategies

                                                For a helpful and complete reference on quantitative sampling strategies, Cochran 2007 is a textbook that provides an excellent reference on different strategies and power calculations. Lohr 2009 is also a detailed textbook on samples in qualitative research and is very helpful when designing a study.

                                                • Cochran, W. G. 2007. Sampling techniques. 3d ed. New York: John Wiley & Sons.

                                                  Save Citation »Export Citation »

                                                  For a thorough reference on sampling, Cochran provides a comprehensive treatment. Although the reference is over 10 years old, the text is very useful, and most concepts are relevant today. The text covers major sampling strategies including sample size estimation. Of particular note, it also provides two chapters on cluster sampling.

                                                  Find this resource:

                                                  • Lohr, S. L. 2009. Sampling: Design and analysis. 2d ed. Boston: Cengage.

                                                    Save Citation »Export Citation »

                                                    This textbook is a complete reference on designing a sampling plan. The book covers probability sampling aspects in detail, stratified sampling, cluster sampling, complex survey sampling, two-phase sampling, rare populations, and how to deal with nonresponse. Additional chapters are devoted to ratio estimation, analysis of surveys, estimation of population size, and survey design. The text is focused on survey research specifically.

                                                    Find this resource:

                                                    Sample Size

                                                    Sample size for quantitative research or the quantitative strand of a mixed methods study is determined by a careful sample size calculation and power analysis. Power refers broadly to the ability to detect a statistical significance of a specified effect size when an effect indeed exists. For an introduction to sample size, Lenth 2001 describes the relationship between sample size and power. Cohen 1992 provides a clear and easily readable discussion of statistical power that is helpful in determine sample size. For a comprehensive reference on power analysis, Cohen 1988 is an entire book devoted to calculating power and sample size. Sample sizes and power calculation depend on what type of analysis is being conducted. While the aforementioned resources cover most statistical tests, subsequent articles have focused on specific and advanced analyses, including sample sizes in mediation analysis (Fritz and MacKinnon 2007); structural equation modeling (MacCallum, et al. 1996); and factor analysis (MacCallum, et al. 1999; MacCallum, et al. 2001).

                                                    • Cohen, J. 1988. Statistical power analysis for the behavioral sciences. 3d ed. New York: Academic Press.

                                                      Save Citation »Export Citation »

                                                      Cohen is one of the most important authors on the topic of determining sample sizes and statistical power calculations. This textbook is a definitive resource on power analysis for various techniques. It provides in-depth guidance about how to conduct a rigorous power analysis—to decide how large a sample size must be, at a given power level, to detect a certain effect size, for a particular statistical test.

                                                      Find this resource:

                                                      • Cohen, J. 1992. A power primer. Psychological Bulletin 112.1: 155–159.

                                                        DOI: 10.1037/0033-2909.112.1.155Save Citation »Export Citation »

                                                        This excellent article provides a concise, practical, and accessible review of statistical power. Cohen reviews key components of a power calculation: significance level, power, sample size, and the effect size. Table 1 summarizes formulae to calculate effect sizes for eight common statistical tests (p. 157) and Table 2 provides sample sizes for each of these tests at various criteria to detect a small, medium, and large effect size. It concludes with example power calculations (p. 158).

                                                        Find this resource:

                                                        • Fritz, M. S., and D. P. MacKinnon. 2007. Required sample size to detect the mediated effect. Psychological Science 18.3: 233–239.

                                                          DOI: 10.1111/j.1467–9280.2007.01882.xSave Citation »Export Citation »

                                                          It discuss the sample size needed for adequate statistical power to test for mediation. Reviewing six common tests for mediation, they then present their empirical simulation of the sample size needed for each to achieve .8 power. Table 3 succinctly presents their results of estimated sample sizes under different conditions (e.g., various effect sizes of path from independent variable-mediator and from mediator-dependent variable) (p. 237). They conclude with recommendations about tests for mediation.

                                                          Find this resource:

                                                          • Lenth, R. V. 2001. Some practical guidelines for effective sample size determination. American Statistician 55: 187–193.

                                                            DOI: 10.1198/000313001317098149Save Citation »Export Citation »

                                                            It discusses the relationship between sample size and power, including an example. The article covers sample size determination and the importance of knowing residual variance. It provides important cautions, such as considering ethical issues and risk, avoiding broad effect size guidelines without considering the study circumstances, and avoiding retrospective power analysis.

                                                            Find this resource:

                                                            • MacCallum, R. C., M. W. Browne, and H. M. Sugawara. 1996. Power analysis and determination of sample size for covariance structure modeling. Psychological Methods 1.2: 130–149.

                                                              DOI: 10.1037/1082-989X.1.2.130Save Citation »Export Citation »

                                                              This article discusses how to conduct a power analysis and calculate sample size for structural equation modeling. It includes examples, calculation of prospective and retrospective power, and comparisons of other methods for power analysis.

                                                              Find this resource:

                                                              • MacCallum, R. C., K. F. Widaman, K. J. Preacher, and S. Hong. 2001. Sample Size in Factor Analysis: The Role of Model Error. Multivariate Behavioral Research 36.4: 611–637.

                                                                DOI: 10.1207/S15327906MBR3604_06Save Citation »Export Citation »

                                                                MacCallum and colleagues 2001 extended their earlier article to account for model error and sampling error. They empirically tested with both simulated and real data. Their results are primarily applicable to exploratory factor analysis.

                                                                Find this resource:

                                                                • MacCallum, R. C., K. F. Widaman, S. Zhang, and S. Hong. 1999. Sample size in factor analysis. Psychological Methods 4: 84–99.

                                                                  DOI: 10.1037/1082-989X.4.1.84Save Citation »Export Citation »

                                                                  MacCalum and colleagues’ 1999 highly cited article reviews the effects of sample size when conducting a factor analysis. They argue that common rules of thumb for the ratio of sample size to number of variables (e.g., items on an instrument) are not accurate.

                                                                  Find this resource:

                                                                  Discipline Specific and Special Considerations

                                                                  Authors have often written about sampling specific to disciplines. Some of this literature discusses unique considerations. Other articles examine sampling among empirical articles within a specific field to address questions, such as how researchers in a particular field are making sampling decisions, including mass communications research (Erba, et al. 2018), organizational research (Bartlett, et al. 2001), and health research (Lwanga, et al. 1991).

                                                                  • Bartlett J. E., J. W. Kotrlik, and C. C. Higgins. 2001. Organizational research: Determining appropriate sample size in survey research. Information Technology, Learning, and Performance Journal 19.1: 43–50.

                                                                    Save Citation »Export Citation »

                                                                    It provides a thorough overview of sample size determination in organization survey research although the methods apply to a variety of fields. The article summarizes sample size calculation. They apply sample size calculation to continuous and categorical data along with important considerations about variance of variables. Table 1 provides a quick reference for minimum sample sizes of completed surveys for specified population sizes, margins of error, and alpha sizes (p. 48).

                                                                    Find this resource:

                                                                    • Erba, J., B. Ternes, P. Bobkowski, T. Logan, and Y. Liu. 2018. Sampling methods and sample populations in quantitative mass communication research studies: A 15-year census of six journals. Communication Research Reports 35.1: 42–47.

                                                                      DOI: 10.1080/08824096.2017.1362632Save Citation »Export Citation »

                                                                      Erba and colleagues examined 1,173 mass communication research studies published in six journals between 2000 and 2014. Experimental studies tended to use more on nonprobability and student samples than surveys. Funded studies used more probability and nonstudent samples. They conclude with a critical reflection on the sampling practices found and implications. However, throughout the article, they often use the term “sample population,” which makes it difficult to discern whether they are referring to the sample or to the population at each point.

                                                                      Find this resource:

                                                                      • Lwanga, S. K., S. Lemeshow, and World Health Organization. ‎1991‎. Sample size determination in health studies: A practical manual. Geneva, Switzerland: World Health Organization.

                                                                        Save Citation »Export Citation »

                                                                        For health studies, the World Health Organization (WHO) has published a manual of sample size determination along with helpful tables for a quick reference. The publication is comprehensive and provides guidance for sample size determination in the following types of research designs: one-sample designs (e.g., testing population proportions), two-sample comparisons, case-control studies, cohort studies, lot quality assurance, and incident-rate studies.

                                                                        Find this resource:

                                                                        Sampling in Mixed Methods Research

                                                                        Mixed methods researchers contend with sampling from qualitative and quantitative strands, but also sampling issues unique to mixed methods. Researchers have three broad ways to consider mixed methods samples: (1) the quantitative and qualitative sample drawn from a population are exactly the same individuals; (2) the qualitative and quantitative sample drawn from a population are different individuals; and (3) one sample is drawn from a population for one strand and the other strand is a subset of that larger sample (e.g., select a quantitative sample and then draw a purposeful qualitative subsample among those individuals or cases). The literature on sampling specific to mixed methods primarily provides guidance with the overall strategy or how to think about sampling rather than delving into sample size. To begin reading about mixed methods sampling, the sampling typology in Teddlie and Yu 2007 provides guidelines and examples for mixed methods researchers. A useful resource when designing a mixed methods study is the Onwuegbuzie and Collins 2017 article that presents a sampling framework that spans the research process. For more foundational insight into combining probability and purposeful sampling, see Sandelowski 2000. For more-specialized considerations, Lyons, et al. 2017 reviews strategies for sampling and recruiting hard-to-reach populations in mixed methods studies; Palinkas, et al. 2015 focuses on sampling in implementation research.

                                                                        • Lyons, A., W. Heywood, B. Fileborn, et. al. 2017. The sex, age, and me study: Recruitment and sampling for a large mixed-methods study of sexual health and relationships in an older Australian population. Culture, Health & Sexuality 19.9: 1038–1052.

                                                                          DOI: 10.1080/13691058.2017.1288268Save Citation »Export Citation »

                                                                          This article discusses recruiting to achieve a representative sample of a hard-to-reach population for a large mixed methods study. Their challenge was recruiting a large sample, and they succeeded with over 2,100 complete surveys. They discussed their six different recruitment strategies and their practice of recruiting interview participants from the surveys with a question asking if they would like to participate. They attributed their success in recruitment to careful advertising, and in allowing interviewees to choose how to participate.

                                                                          Find this resource:

                                                                          • Onwuegbuzie, A., and K. M. T. Collins. 2017. The role of sampling in mixed methods-research. Kölner Zeitschrift für Soziologie & Sozialpsychologie 69.2 (October): 133–156.

                                                                            DOI: 10.1007/s11577-017-0455-0Save Citation »Export Citation »

                                                                            It emphasize sampling in mixed methods studies as critical for ensuring high quality inferences. After reviewing sampling considerations for qualitative and quantitative research processes separately, they provide a framework for considering sampling at all stages of the mixed methods research process. The article details six considerations: emtic orientation, probabilistic orientation, abductive orientation, intrinsic versus instrumental orientation, particularistic versus universalistic orientation, and philosophical clarity. The article closes with guidance on writing and reporting.

                                                                            Find this resource:

                                                                            • Palinkas, L. A., S. M. Horwitz, C. A. Green, J. P. Wisdom, N. Duan, and K. Hoagwood. 2015. Purposeful sampling for qualitative data collection and analysis in mixed method implementation research. Administration and Policy in Mental Health and Mental Health Services Research 42.5: 533–544.

                                                                              DOI: 10.1007/s10488–013–0528-ySave Citation »Export Citation »

                                                                              This article focus on sampling for the qualitative strand of mixed methods implementation research studies. They summarize different types of purposeful sampling. Table 1 gives a helpful reference on sampling strategies along with each strategy’s objective, considerations, and an example from implementation research (p. 535). Finally, the article covers multistage purposeful sampling strategies in which the emphasis on variation (i.e., breadth) and similarity (i.e., depth) may change.

                                                                              Find this resource:

                                                                              • Sandelowski, M. 2000. Combining qualitative and quantitative sampling, data collection, and analysis techniques in mixed-method studies. Research in Nursing & Health 23.3: 246–255.

                                                                                DOI: 10.1002/1098-240X(200006)23:3<246::AID-NUR9>3.0.CO;2-HSave Citation »Export Citation »

                                                                                In addition to discussing integrating qualitative and quantitative data collection and analysis methods, it discusses combining purposeful and probability sampling techniques in a mixed methods study. She discusses three combinations: criterion sampling that uses quantitative results to inform a subsequent purposeful criterion qualitative sample, random purposeful sampling that uses an initial purposeful criterion to identify groups and then randomly selects within groups, and stratified purposeful sampling that seeks participants that vary on parameters.

                                                                                Find this resource:

                                                                                • Teddlie, C., and F. Yu. 2007. Mixed methods sampling: A typology with examples. Journal of Mixed Methods Research 1: 77–100.

                                                                                  DOI: 10.1177/1558689806292430Save Citation »Export Citation »

                                                                                  This classic article provides a typology of sampling strategies for mixed methods studies with examples and sampling guidelines for mixed methods researchers. In addition, they provide a comparison of purposive and probability sampling techniques (see Table 1, [p. 84]). The article also reviews characteristics of mixed methods sampling. Finally, they present types of mixed methods sampling strategies: basic, sequential, concurrent, multilevel, and use of multiple strategies.

                                                                                  Find this resource:

                                                                                  Sampling Strategies Unique to Mixed Methods Designs

                                                                                  Other considerations in mixed methods sampling include the design used and how qualitative and quantitative research interact. Coleman, et al. 1996 discusses sampling in explanatory sequential designs. Sharp, et al. 2012 provides guidance for sampling in multisite mixed methods case studies. Grounding its arguments in and examination of 121 studies across the social and health sciences, Collins, et al. 2007 emphasizes the timing and the relationship between quantitative and qualitative samples. A more recent webinar (Collins 2018) updates the discussion.

                                                                                  • Coleman, T., M. Williams, and A. Wilson. 1996. Sampling for qualitative research using quantitative methods. 1. Measuring GPs’ attitudes towards discussing smoking with patients. Family Practice 13.6: 526–530.

                                                                                    DOI: 10.1093/fampra/13.6.526Save Citation »Export Citation »

                                                                                    In an explanatory sequential type of design, investigators begin with a quantitative phase, which informs a subsequent qualitative phase. Although they did not explicitly label their study as mixed methods, Coleman and colleagues discuss their methods of using an initial quantitative survey to select participants with diverse attitudes for further qualitative exploration.

                                                                                    Find this resource:

                                                                                    • Collins, K. M., A. J. Onwuegbuzie, and Q. G. Jiao. 2007. A mixed methods investigation of mixed methods sampling designs in social and health science research. Journal of Mixed Methods Research 1.3: 267–294.

                                                                                      DOI: 10.1177/1558689807299526Save Citation »Export Citation »

                                                                                      This article examined 121 studies across the social or health sciences and classified the studies in terms of the timing and the relationship between the qualitative and quantitative sample. Timing can be either concurrent or sequential and the relationship has four possible types—identical, nested, parallel, multilevel—yielding eight possible sampling designs. They close with practical advice and urge use of their model to guide sampling decisions.

                                                                                      Find this resource:

                                                                                      • Collins, K. T. 2018. Sampling designs in mixed research (MR). Webinar presented for the International Institute of Qualitative Methodology Mixed Methods Webinar Series.

                                                                                        Save Citation »Export Citation »

                                                                                        This webinar includes both a recorded presentation and slides by Collins on “mixed research,” another term for mixed methods. She presents a typology and concerns unique to mixed methods. A particularly useful section covers unique challenges-representation, legitimation, integration, politics, and ethics—to sampling in mixed methods research.

                                                                                        Find this resource:

                                                                                        • Sharp, J. L., C. Mobley, C. Hammond, et al., 2012. A mixed methods sampling methodology for a multisite case study. Journal of Mixed Methods Research 6: 34–54.

                                                                                          DOI: 10.1177/1558689811417133Save Citation »Export Citation »

                                                                                          This article presents options for using mixed methods to select sites for multisite case studies. Using their research on school to work and college programs, it describe four stages of site selection: identifying diverse sites using existing data, selecting clusters based on hierarchical cluster analysis of measures, rank ordering the sites based on policy implementation levels, and conducting site visits to validate data collected in the three previous stages.

                                                                                          Find this resource:

                                                                                          back to top

                                                                                          Article

                                                                                          Up

                                                                                          Down