Comparer des méthodes

	Méthodes mixtes d'intervention basées sur la conception	Inférence méta-mixte basée sur la conception
Domaine	Conception de la recherche	Conception de la recherche
Famille	Process / pipeline	Process / pipeline
Année d'origine	2003–2010s (convergence of DBR and mixed methods traditions)	2003–2009
Auteur d'origine	Design-Based Research Collective; Creswell & Plano Clark (mixed methods framework)	Abbas Tashakkori & Charles Teddlie
Type	Mixed methods research design variant	Mixed methods integration strategy
Source fondatrice	Creswell, J. W., & Plano Clark, V. L. (2018). Designing and Conducting Mixed Methods Research (3rd ed.). Sage Publications. ISBN: 978-1483344452	Teddlie, C., & Tashakkori, A. (2009). Foundations of Mixed Methods Research: Integrating Quantitative and Qualitative Approaches in the Social and Behavioral Sciences. Sage. ISBN: 978-0761930129
Alias	DBR intervention mixed methods, design-based intervention study, design experiment with mixed methods, intervention design-based mixed methods	mixed methods meta-inference, MMR meta-inference, integrated meta-inference, design-based meta-inference
Apparentées	6	0
Résumé	Design-based intervention mixed methods is a research design that embeds both quantitative and qualitative data collection within iterative intervention cycles drawn from design-based research (DBR). The approach systematically tests and refines a practical intervention — typically an educational program, curriculum, or organizational solution — while using qualitative data to explain why and how the intervention works, and quantitative data to assess its measurable impact. Iteration between design, testing, and revision is the hallmark of this approach.	Design-based mixed methods meta-inference is the overarching conclusion drawn by explicitly integrating the separate quantitative and qualitative inferences from a mixed methods study, with the integration logic anchored to the a priori research design. Rather than treating quantitative and qualitative results as parallel outputs, the approach requires the researcher to specify — at the design stage — how and why the two strands will be combined, and then to construct a unified meta-inference that is consistent with that design rationale.