Embedded Pragmatic Mixed Methods — Nested Pragmatist Mixed Design
Embedded Pragmatic Mixed Methods Design · Also known as: embedded pragmatic MMR, pragmatic nested mixed methods, embedded pragmatist MMD, nested pragmatic mixed design
Embedded pragmatic mixed methods is a mixed methods design in which one data strand (typically qualitative) is nested within a larger, dominant strand (typically quantitative), and the entire study is guided by a pragmatist philosophical stance — selecting methods for what works best to answer the research question rather than adhering to a fixed paradigmatic commitment. The nested strand enriches or contextualises the dominant strand without standing on its own.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use embedded pragmatic mixed methods when a primary design (e.g., an RCT, a survey, or a case study) does not fully address an important secondary question, and embedding a smaller strand is more efficient than running two separate studies. It is well suited to applied and evaluation research where practical constraints limit resources and the pragmatist question is 'what combination of data will best answer this problem?' Do not use it when the secondary question is as important as the primary one — in that case, a concurrent triangulation or parallel design is more appropriate. Avoid it when the nested strand requires a sample size or timeline incompatible with the primary design.
Strengths & limitations
- Efficient: one study generates both quantitative and qualitative evidence without the overhead of two separate projects.
- Pragmatically justified: method selection is driven by the research problem, making it easy to defend design choices to funders and reviewers.
- Adds explanatory depth to dominant-strand findings (e.g., explaining survey outliers through interviews).
- Flexible: the nested strand can be collected concurrently with or sequentially after the primary strand depending on what is needed.
- Well suited to applied, evaluation, and intervention research where mixed evidence is expected by stakeholders.
- The nested strand is constrained by the primary design — sample size, timing, and scope are not optimised for the secondary questions.
- Integration can be complex when the two strands produce discrepant findings that the pragmatic framing does not fully resolve.
- The subordinate strand may be under-powered for its own analytic purposes if the primary design drives sample size decisions.
- Reporting the two strands within a single manuscript can be structurally difficult; journals oriented toward one paradigm may resist the integrated format.
Frequently asked
What makes this design 'pragmatic' rather than just 'embedded'?
The embedded structure refers to the nesting of one strand within another. The pragmatic label signals the philosophical orientation: design decisions — including why to embed, which strand is dominant, and how integration is handled — are justified by what best answers the research problem rather than by allegiance to quantitative or qualitative paradigms. Pragmatism is a substantive philosophical position traceable to Peirce, James, Dewey, and, in mixed methods, to Morgan and Tashakkori and Teddlie.
How do I decide which strand should be dominant?
The dominant strand should correspond to the primary research question. If the study's main aim is to measure the effect of an intervention, the quantitative strand is dominant. If the primary aim is to understand a social process in depth, the qualitative strand dominates. The nested strand addresses a secondary question that the primary strand cannot fully answer on its own.
How large does the nested sample need to be?
There is no fixed rule, but the nested sample must be large enough to answer its specific secondary question meaningfully. For qualitative nesting within a quantitative primary strand, 8–20 participants is commonly sufficient for thematic analysis; fewer than six makes it difficult to identify patterns. The nested sample is typically a purposeful sub-sample of primary-strand participants, not the full sample.
When should I use a concurrent triangulation design instead?
Use concurrent triangulation when both quantitative and qualitative strands are equally important to the main research question and you want to compare their findings directly. Use embedded pragmatic design when one strand is clearly subordinate — its purpose is to enrich or explain the dominant strand rather than to stand as an equal source of evidence.
How is integration reported?
Integration should be explicit: report where in the analysis the two strands were brought together (e.g., after quantitative results, before final conclusions), what procedure was used (e.g., joint display, following a significant quantitative result with qualitative illustration), and what the combined evidence reveals that neither strand could show alone. A joint display table or figure is a practical tool for demonstrating integration to reviewers.
Sources
- Creswell, J. W., & Plano Clark, V. L. (2018). Designing and Conducting Mixed Methods Research (3rd ed.). Sage. ISBN: 978-1483344379
- Morgan, D. L. (2014). Pragmatism as a paradigm for social research. Qualitative Inquiry, 20(8), 1045–1053. DOI: 10.1177/1077800413513733 ↗
How to cite this page
ScholarGate. (2026, June 3). Embedded Pragmatic Mixed Methods Design. ScholarGate. https://scholargate.app/en/research-design/embedded-pragmatic-mixed-methods
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Embedded Multiphase Mixed MethodsResearch Design↔ compare
- Embedded Transformative Mixed MethodsResearch Design↔ compare
- Explanatory Sequential Mixed Methods DesignResearch Design↔ compare
- Multilevel Mixed Methods DesignResearch Design↔ compare
- Pragmatic Mixed Methods DesignResearch Design↔ compare