Multilevel Mixed Methods Design — Integrating Data Across Levels of Analysis
Multilevel Mixed Methods Research Design · Also known as: multilevel MMR, nested mixed methods, hierarchical mixed methods design, cross-level mixed methods
Multilevel mixed methods design is a research approach that collects and integrates both quantitative and qualitative data at two or more distinct levels of a social or organizational hierarchy — for example, individuals nested within classrooms, classrooms within schools, or patients within healthcare teams. By pairing quantitative measurement of outcomes at one level with qualitative exploration of meaning at another, researchers gain a richer, more complete picture than either strand alone could provide.
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.
+20 more
When to use it
Use multilevel mixed methods design when your research question is inherently nested — that is, when individuals, groups, and/or institutions are all part of the phenomenon and no single level of analysis is sufficient on its own. It is especially well-suited to educational research, health systems research, organizational studies, community-based interventions, and policy evaluation. Do not use this design if your data are collected at a single level of analysis, if you lack access to multiple levels, or if resource constraints make coordinating two or more strands across levels infeasible. A simpler concurrent or sequential design may be preferable when the hierarchical structure is not theoretically central to the research question.
Strengths & limitations
- Captures complexity by pairing the appropriate data type with each level of the hierarchy, yielding more complete understanding than a single-level or single-strand study.
- Particularly powerful for evaluating interventions in nested settings (classrooms, clinics, organizations) where context at higher levels shapes outcomes at lower levels.
- Supports both top-down (policy to practice) and bottom-up (individual experience to systemic pattern) explanations within the same study.
- Flexible in sequencing — strands can run concurrently or one can inform the other, depending on what each level of analysis requires.
- Produces findings that are directly relevant to multiple stakeholder audiences — practitioners, managers, and policymakers — simultaneously.
- Substantially more complex to design, manage, and report than single-level or single-strand designs; requires expertise in both quantitative multilevel analysis and qualitative methods.
- Resource-intensive: multiple sampling frames, data collection instruments, and analytical procedures must be coordinated across levels.
- Integration across levels is challenging — the units of analysis differ, making direct comparison or merging of findings methodologically demanding.
- Access to data at all required levels is not always feasible, and gatekeeping at institutional levels can derail the design mid-study.
- Reporting conventions are not fully standardized, which can complicate peer review and publication in journals that favor single-paradigm designs.
Frequently asked
Is multilevel mixed methods design the same as hierarchical linear modelling (HLM)?
No. HLM is a purely quantitative statistical technique for analyzing nested data. Multilevel mixed methods design is a broader research design framework that integrates both quantitative and qualitative data at multiple levels of analysis. HLM may be one of the quantitative tools used within the design, but the design itself involves qualitative strands and integration across levels that HLM alone cannot provide.
How many levels do I need for this design to be appropriate?
A minimum of two distinct levels is required — for example, individuals nested within groups, or groups nested within institutions. Three or more levels are common in educational and health systems research. The key criterion is that the research question is genuinely nested: outcomes or meanings at one level are shaped by or embedded in structures at another level.
Can I use a multilevel mixed methods design with a small sample?
At the quantitative level, small samples limit statistical power and the feasibility of multilevel modelling. However, the qualitative strand at another level can operate with a smaller purposive sample. The design is feasible with modest overall samples if the quantitative strand is descriptive rather than inferential, but researchers should be transparent about the limitations this places on generalizability of the quantitative findings.
How do I integrate findings across levels that use different units of analysis?
Integration is typically achieved through joint displays — visual matrices or tables that place quantitative and qualitative findings side by side for comparison — and through narrative integration that explicitly discusses convergence, divergence, and complementarity. Some researchers develop explanatory models in which findings at one level explain patterns observed at another. The integration step should be planned before data collection, not treated as an afterthought.
When should I prefer a simpler mixed methods design over a multilevel one?
Choose a simpler concurrent or sequential design when your research question does not require data at multiple levels of a hierarchy, when you lack access to multiple levels, or when resource and time constraints make a multilevel design unworkable. The added complexity of multilevel design is only justified when the nesting structure is theoretically central to understanding the phenomenon.
Sources
- Creswell, J. W., & Plano Clark, V. L. (2018). Designing and Conducting Mixed Methods Research (3rd ed.). Sage Publications. ISBN: 978-1483357829
- Teddlie, C., & Tashakkori, A. (2009). Foundations of Mixed Methods Research: Integrating Quantitative and Qualitative Approaches in the Social and Behavioral Sciences. Sage Publications. ISBN: 978-0761930129
How to cite this page
ScholarGate. (2026, June 3). Multilevel Mixed Methods Research Design. ScholarGate. https://scholargate.app/en/research-design/multilevel-mixed-methods-design
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.
- Concurrent Embedded Mixed Methods DesignResearch Design↔ compare
- Concurrent Triangulation Mixed Methods DesignResearch Design↔ compare
- Explanatory Sequential Mixed Methods DesignResearch Design↔ compare
- Exploratory Sequential Mixed Methods DesignResearch Design↔ compare
- Multiphase Mixed Methods DesignResearch Design↔ compare