Evaluation-Focused Concurrent Embedded Mixed Methods
Evaluation-Focused Concurrent Embedded Mixed Methods Design · Also known as: concurrent embedded evaluation design, embedded mixed methods evaluation, nested concurrent evaluation design, mixed methods program evaluation
Evaluation-focused concurrent embedded mixed methods is a research design in which both quantitative and qualitative data are collected simultaneously within a program evaluation context, with one strand nested inside and playing a supporting role to the dominant strand. The design produces outcome evidence alongside embedded process or contextual evidence from the same evaluation cycle, without extending the timeline.
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 this design when you are conducting a formal program evaluation and need outcome data from a large sample alongside contextual or process data from a subset — within a single data-collection period. It is well-suited when funders require measurable outcomes but stakeholders also need to understand mechanism and participant experience. The design is not appropriate when the two research questions are of equal priority (use convergent parallel design instead), when the timeline allows sequential follow-up (use explanatory sequential design instead), or when evaluators lack the resources to run two concurrent data-collection streams.
Strengths & limitations
- Delivers both outcome evidence and contextual explanation within a single evaluation cycle, saving time relative to sequential designs.
- The nested structure keeps the dominant strand methodologically uncompromised while adding qualitative depth.
- Particularly persuasive to mixed audiences: funders see numbers; practitioners and policymakers see participant narratives.
- Allows evaluation questions about mechanism and fidelity to be answered alongside impact questions using the same participants and setting.
- The asymmetric priority structure clarifies resource allocation decisions from the outset.
- Managing two concurrent data streams in the field requires careful coordination and can strain evaluation budgets and staff.
- The secondary strand, by design, has lower priority and a smaller sample; it may not achieve theoretical saturation.
- Integration at interpretation — not at analysis — can leave evaluators uncertain about how to handle divergent findings.
- The design demands dual methodological expertise; an evaluator strong in only one tradition may inadvertently weaken the secondary strand.
Frequently asked
How is this design different from a convergent parallel design?
In a convergent parallel design both strands carry equal weight and are merged to produce a single unified interpretation. In the concurrent embedded design one strand is explicitly dominant and the other is nested within it to serve a supplementary purpose — contextualising, explaining, or illustrating the primary findings. The asymmetry of priority and sample size is the defining structural difference.
Does the embedded strand need to reach saturation?
Not necessarily. Because the embedded strand plays a supporting role, its sample size is intentionally smaller and purposive. The goal is not theoretical saturation but sufficient depth to contextualise the dominant findings. Typically 10–20 participants for the embedded qualitative component is defensible, depending on the scope of the evaluation.
When should I choose this design over an explanatory sequential design?
Choose concurrent embedded when the evaluation window is fixed (e.g., a single program cycle) and there is no opportunity for a follow-up qualitative phase. Choose explanatory sequential when you can afford a two-phase timeline and you want the dominant quantitative results to directly shape the questions asked in the qualitative phase.
How do I handle it when the embedded qualitative findings contradict the quantitative outcomes?
Divergence is informative, not a flaw. Report both strands honestly and use the divergence as an analytic starting point: do the qualitative accounts suggest implementation failures, subgroup differences, or measurement limitations that the quantitative strand could not detect? Forced reconciliation obscures genuine complexity; transparent discussion of divergence strengthens evaluation credibility.
Can the dominant strand be qualitative rather than quantitative?
Yes. A QUAL+quan embedding is legitimate when the evaluation is primarily interpretive — for example, a large ethnographic study of program culture with a nested survey of participant attitudes. In practice, evaluation-focused embedded designs more often have a quantitative dominant strand because funders typically require measurable outcomes.
Sources
- Creswell, J. W., & Plano Clark, V. L. (2018). Designing and Conducting Mixed Methods Research (3rd ed.). Sage. ISBN: 978-1483344379
- Greene, J. C. (2007). Mixed Methods in Social Inquiry. Jossey-Bass. ISBN: 978-0787983826
How to cite this page
ScholarGate. (2026, June 3). Evaluation-Focused Concurrent Embedded Mixed Methods Design. ScholarGate. https://scholargate.app/en/research-design/evaluation-focused-concurrent-embedded-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 Case StudyQualitative↔ compare
- Program EvaluationField Methods↔ compare