Embedded Intervention Mixed Methods — Mixed Methods Design with a Nested Qualitative Strand
Embedded Intervention Mixed Methods Design · Also known as: nested intervention mixed methods, embedded experimental mixed methods, mixed methods intervention design with embedded strand, EIMM
Embedded intervention mixed methods is a research design in which a qualitative strand is nested within a primary quantitative intervention (experimental or quasi-experimental) study. The quantitative component tests whether an intervention works; the embedded qualitative component illuminates how and why it works — or does not work — for participants. The two strands run concurrently or sequentially within a single overarching study framework, with integration occurring during interpretation.
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 intervention mixed methods when you are conducting an experimental or quasi-experimental study and need to understand process, mechanism, or participant experience alongside the outcome — and when resources or participant access limit a fully separate qualitative study. It is appropriate in health interventions, educational program evaluations, social service trials, and community-based research where stakeholder voices matter alongside efficacy data. Do not use it when the qualitative question is equally important to the quantitative question (use a concurrent triangulation or parallel design instead), when the intervention and qualitative strands require separate, independent samples that cannot be merged, or when budget and timeline do not permit the additional data collection burden within an already complex trial.
Strengths & limitations
- Generates both outcome evidence (did it work?) and process evidence (how and why?) within a single study, maximising explanatory value.
- More resource-efficient than running separate quantitative and qualitative studies on the same topic.
- The qualitative strand can identify implementation fidelity issues, unanticipated effects, and context factors that modify the intervention's impact.
- Particularly valuable for health, education, and policy research where funders require rigorous outcome data but practitioners need actionable process insight.
- Supports equity-oriented inquiry by capturing the perspectives of subgroups who may respond differently to the intervention.
- The qualitative strand is by definition secondary; complex qualitative questions may be under-resourced or truncated to fit the trial timeline.
- Managing two methodologically distinct strands within one study places high demands on researcher skill and team coordination.
- Integration at the interpretation stage is conceptually demanding; surface-level juxtaposition of findings does not constitute genuine integration.
- The embedded strand's smaller, purposive sample limits transferability of the qualitative findings beyond the intervention context.
Frequently asked
How is this different from the concurrent embedded mixed methods design?
The concurrent embedded design is a general design category in which a secondary strand is nested within a primary strand running at the same time — the primary strand can be either quantitative or qualitative. The embedded intervention design is a specific application of that logic where the primary strand is always a quantitative intervention study (experiment or quasi-experiment) and the secondary qualitative strand is specifically designed to explain intervention processes or outcomes. The intervention design adds the experimental logic and the process-evaluation purpose.
Should the qualitative sample be drawn from the same participants as the trial?
Usually yes — a purposive sub-sample of trial participants is most common and provides the tightest integration between quantitative and qualitative data. However, the qualitative strand can also include practitioners delivering the intervention or stakeholders not enrolled in the trial, particularly when the process question concerns implementation rather than participant experience. Sampling decisions should be driven by the specific embedded research question.
When should integration happen — during or after data collection?
In most embedded intervention designs, full integration occurs at the interpretation stage, after both strands have been analyzed separately. However, preliminary qualitative findings from an early wave of interviews can inform refinements to intervention delivery (adaptive trial designs) or flag emerging issues mid-trial. Any mid-study integration must be pre-specified in the protocol to avoid bias.
How do I report an embedded intervention mixed methods study?
Report the quantitative and qualitative components using the conventions of each tradition (e.g., CONSORT for the trial, COREQ or SRQR for the qualitative strand), and then add a mixed methods integration section or use a joint display table to show how the two strands address the overarching research question together. Journals in health and education increasingly expect this layered reporting structure.
Can I use this design in a quasi-experimental rather than randomized study?
Yes. The embedded intervention design applies to any quantitative intervention study — randomized controlled trials, cluster randomized trials, stepped-wedge designs, pre-post quasi-experiments, and interrupted time series. The requirement is that there is a clearly defined quantitative intervention component; the embedded qualitative strand logic remains the same regardless of whether randomization is present.
Sources
- Creswell, J. W., & Plano Clark, V. L. (2018). Designing and Conducting Mixed Methods Research (3rd ed.). SAGE Publications. ISBN: 978-1483344379
- Mertens, D. M. (2009). Transformative Research and Evaluation. Guilford Press. ISBN: 978-1593856205
How to cite this page
ScholarGate. (2026, June 3). Embedded Intervention Mixed Methods Design. ScholarGate. https://scholargate.app/en/research-design/embedded-intervention-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 Exploratory Sequential Mixed MethodsResearch Design↔ compare
- Embedded Multiphase Mixed MethodsResearch Design↔ compare
- Explanatory Sequential Mixed Methods DesignResearch Design↔ compare
- Intervention Mixed Methods DesignResearch Design↔ compare