Sampling strategies in mixed methods research

Mixed-methods sampling involves integrating qualitative and quantitative sampling techniques to address the diverse requirements of a mixed methods research design. Read this article to explore various types of mixed methods sampling, including basic, sequential, concurrent, multilevel, and combinations of these strategies.

Introduction

In qualitative and quantitative methods, sampling strategies are crafted to align with the specific goals of the research. Quantitative research typically emphasizes generalizability through techniques like random sampling, which ensures that findings represent a broader population. On the other hand, qualitative research focuses on in-depth exploration and employs purposeful techniques such as maximum variation sampling to capture diverse perspectives and generate meaningful insights during qualitative analysis. The distinction between qualitative and quantitative approaches highlights their complementary strengths. Quantitative studies prioritize statistical precision and large sample sizes to test hypotheses, while qualitative methods offer nuanced understandings through smaller, intentional samples. Mixed methods studies integrate these two approaches, blending their strengths to address complex research questions. In mixed methods studies, sampling strategies are particularly important, as they must harmonize the objectives of both qualitative and quantitative components, whether in sequential or concurrent designs.

Quantitative and qualitative sampling

The fundamental distinction between quantitative and qualitative sampling lies in their objectives. Quantitative research seeks generalizability through large, random samples, while qualitative research prioritizes in-depth understanding through smaller, purposefully selected samples.

Quantitative sampling aims to achieve generalizability by selecting samples that represent the broader population. This approach primarily relies on probability sampling techniques, including random, stratified, and cluster sampling. In experimental designs, sample size is often determined through power analysis to ensure sufficient statistical power for detecting significant effects. For survey research, standardized sample size tables are commonly used to guide representativeness and reliability in the collected data. This structured methodology enhances external validity, allowing researchers to make broader inferences from the data.

Quantitative studies often aim for statistical significance, whereas qualitative studies aim for thematic saturation—the point at which all dimensions of the findings are sufficiently understood and no new information emerges from the data. Qualitative sampling focuses on depth rather than breadth. It employs purposive sampling strategies to select participants who can offer rich, detailed insights relevant to the research question. The choice of sample size and participants varies based on the research design, such as narrative, phenomenological, grounded theory, ethnographic, or case study approaches. Unlike quantitative research, qualitative sampling is not about representativeness but about selecting information-rich cases that help uncover nuanced understandings of a phenomenon.

Types of mixed methods sampling

The choice of sampling strategy depends on the type of mixed methods design adopted—whether sequential, concurrent, or multilevel—and the specific objectives of the study. These strategies allow researchers to achieve representativeness, depth, and contextual understanding, often required for complex, multi-faceted research problems.

Sequential mixed methods sampling

This type of sampling involves sequential phases where the findings of one phase guide the sampling decisions of the next. It can take two forms:

Explanatory sequential design: Quantitative data is collected first. The results inform the selection of a smaller, purposeful sample for qualitative exploration. As an example, after conducting a large-scale survey to measure teacher job satisfaction, the researcher might select specific teachers (e.g., high and low scorers) for in-depth interviews to understand underlying reasons.

Exploratory sequential design: Qualitative data is collected first. Insights from this phase are used to develop or refine quantitative sampling. For example, conducting focus groups to explore barriers to vaccination, then designing a survey based on themes identified and sampling households for a quantitative study.

Concurrent mixed methods sampling

Quantitative and qualitative data are collected simultaneously, often using distinct samples for each method. The integration of the two datasets happens during or after analysis. This approach is efficient as it saves time by collecting both types of data simultaneously but it requires careful planning to ensure alignment. For example, a researcher studying community health might simultaneously administer a survey to a large sample of households and conduct interviews with health workers using purposive sampling.

Multilevel mixed methods sampling

This involves sampling at multiple levels of analysis, such as individuals, groups, organizations, or geographical regions. It is particularly useful for complex research questions that span various levels of social or institutional systems. For example, a study on educational disparities might use random sampling to select schools across districts (quantitative) and use purposive sampling to interview teachers and administrators within those schools (qualitative).

Guidelines for mixed methods research sampling

According to Teddlie and Yu (2007), there are some guidelines researchers can consider when they are planning sampling strategies for mixed methods research:

• Align sampling with research questions: The choice of sampling techniques should directly reflect the research objectives. Quantitative methods may prioritize generalizability, while qualitative methods aim for a deeper understanding of a phenomenon.
• Balance representativeness and depth: Mixed methods sampling must address the trade-off between generalizability (quantitative) and saturation or in-depth insights (qualitative). A clear plan to achieve both goals is critical.
• Integrate sampling decisions across methods: Ensure that the quantitative and qualitative sampling strategies are linked and complement one another. For example, a quantitative survey can identify patterns or trends, while qualitative sampling can explore the reasons behind those patterns.
• Specify the purpose of mixing: Researchers must explicitly state why they are combining sampling techniques. Common reasons include triangulation of findings, explaining unexpected results, and exploring new themes to generate hypotheses.
• Account for contextual factors: The research context, such as cultural, geographical, or institutional settings, should influence sampling decisions. This ensures the study remains relevant and practical.
• Use resources efficiently and adapt to emergent findings: Mixed methods sampling can be resource-intensive, requiring careful planning to manage time, budget, and participant recruitment effectively. Sampling in mixed methods research should remain flexible to accommodate new insights that arise during the study. For example, initial qualitative findings might reveal additional variables to measure quantitatively.
• Evaluate and document sampling decisions: Continuously assess and refine sampling strategies throughout the research process. Periodic reviews ensure the sampling remains aligned with the study's goals and emerging findings. Clearly describe the rationale, methods, and outcomes of sampling choices in the research report. This transparency enhances the study’s credibility and allows others to replicate or build upon it.

Conclusion

In mixed methods research, sampling strategies are pivotal in bridging the distinct goals of quantitative and qualitative approaches. By integrating probability-based techniques for generalizability with purposive methods for in-depth exploration, mixed methods sampling enhances the breadth and depth of research findings. Whether employing sequential, concurrent, multilevel, or combined strategies, effective sampling ensures that the diverse components of a study are aligned with its overarching objectives. Careful planning, flexibility, and clear documentation of sampling decisions not only strengthen the quality of the research but also contribute to its overall rigor. As mixed methods research continues to evolve, thoughtful integration of sampling strategies will remain central to addressing complex research questions across disciplines.

References

1. Creswell, J. (2023). Sampling and integration of mixed methods data [Video]. Michigan Medicine. YouTube. https://www.youtube.com/watch?v=tlFRTQpZi1s
2. Teddlie, C., & Yu, F. (2007). Mixed methods sampling: A typology with examples. Journal of Mixed Methods Research, 1(1), 77-100. DOI https://doi.org/10.1177/1558689806292430