mixed-methods-design

# Mixed Methods Design (Creswell & Plano Clark) Mixed methods combines **qualitative and quantitative strands** to answer questions neither can fully address alone. Integration is the hard part: designs specify **sequence, priority, and mixing points**. ## Notable design families (overview) - **Convergent parallel:** QUAN + QUAL concurrently; merge at interpretation (joint displays, side-by-side discussion). - **Explanatory sequential:** QUAN → QUAL to explain mechanisms, outliers, or surprising scores. - **Exploratory sequential:** QUAL → QUAN to develop measures, typologies, or survey items. Advanced variants (embedded, transformative, multiphase) add complexity—match to aims and ethics. ## Mixing strategies (integration) Common techniques: - **Connecting:** outputs of one strand feed the other (e.g., interview themes → survey items). - **Merging:** compare/contrast results in discussion (watch for contradiction as insight). - **Embedding:** one strand secondary within a larger project. Document **where** integration happens and **who** decides how conflicts are resolved. ## Philosophical foundations Mixed methods can sit within **pragmatism** (many US textbooks), **critical realism**, or **transformative** frames. Declare how you handle **incommensurability** if paradigms clash—integration is easier when you articulate a coherent stance. ## When to use mixed methods Use when you need **both** pattern/generalization-oriented evidence and meaning/process depth—or when one strand reduces waste in the other (e.g., QUAL discovers constructs before QUAN measures them). ## Grounded theory in mixed methods GT can occupy the qualitative strand, especially in exploratory sequencing. Risks: - QUAN pressures can **force** premature closure. - Integration meetings may privilege “hard numbers.” Mitigate by protecting **iterative coding cycles** and treating integration as **dialogue**, not automatic trump rules. ## Quality criteria Beyond mono-method rigor, assess: - **Integration clarity** (not parallel play). - **Meta-inferences** justified by both strands. - **Discrepancies** analyzed, not hidden. ## Reporting Use diagrams of the design; specify priority, sequence, purpose of each strand, and integration points. Follow **MMARS** or journal-specific mixed-methods reporting guidance when required. ## Checklist - [ ] Design matches question (explore/explain/converge). - [ ] Priority/sequence explicit. - [ ] Integration points and products defined (joint display, instrument, typology). - [ ] Paradigm stance coherent. - [ ] GT strand protected from premature quantification if emergent. ## References (starting points) - Creswell, J. W., & Plano Clark, V. L. *Designing and conducting mixed methods research* (3rd ed.). - Fetters, M. D., Curry, L. A., & Creswell, J. W. (2013). Achieving integration in mixed methods designs. *Annals of Family Medicine*. - Greene, J. C. *Mixed methods in social inquiry*.