Mixed Methods Research Design - how Not to do it!
Based on Qualitative Researcher Dr Kriukow's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.
Mixed methods requires a planned relationship between qualitative and quantitative components, not just the presence of both.
Briefing
Mixed methods research only counts as “mixed” when qualitative and quantitative components are deliberately connected—running them side by side or simply adding both kinds of data does not automatically create a mixed methods design. The core mistake highlighted is treating qualitative interviews and quantitative surveys as independent add-ons, rather than as parts of a single design where one component informs the other.
To make the distinction concrete, the discussion uses an example study about jaywalking: people repeatedly cross illegally at a specific spot despite the presence of a crosswalk, creating traffic problems and potential injuries. The research team wants to understand why people choose that crossing point. Mixed methods enters the picture as a way to combine measurement (how often, where, what patterns) with meaning (why people do it, what they believe, whether they’re in a rush, and where they’re going).
One common structure is a concurrent or convergent design. Here, qualitative and quantitative data are collected at the same time and often independently—for instance, administering a survey while also conducting interviews or observations. The key requirement is that the two streams must later be brought together in a coherent integration step during analysis (the transcript points to separate material for how that integration works). Without that planned linkage, the study risks becoming two parallel studies rather than a single mixed methods design.
A second structure is a sequential design, where one component follows the other. The crucial nuance is that the second phase should build on the first, not merely occur after it. In an exploratory sequential design, qualitative work comes first to uncover participants’ perspectives and generate ideas that can be turned into survey items. This is especially useful when the phenomenon is under-researched or when researchers don’t yet know what questions should exist for the quantitative instrument. After interviews reveal themes—such as motivations, timing pressures, or destination patterns—researchers use those insights to develop the quantitative survey (e.g., Likert-scale items like agree/strongly agree/disagree).
The reverse pattern is an explanatory sequential design. Quantitative results come first, and then qualitative follow-up is used to interpret or clarify puzzling findings. If many respondents strongly agree with a statement, or if an open-ended response introduces confusion, interviews can probe the reasons behind the numbers. In this setup, survey outcomes shape the interview guide.
The central warning ties these structures together: mixed methods is not just about using both qualitative and quantitative methods. It is about the relationship between them—an overlap where one component informs the other. External pressure can lead researchers to collect both types of data without a clear rationale for why both are needed together. Triangulation alone may be acceptable, but it does not automatically satisfy the defining logic of mixed methods design.
Cornell Notes
Mixed methods research counts only when qualitative and quantitative components are intentionally linked. Collecting interviews and surveys at the same time (concurrent/convergent) is not enough unless the data are integrated in analysis. Sequential designs make the linkage explicit: exploratory sequential starts with qualitative work to generate survey items, while explanatory sequential starts with survey results and then uses interviews to explain or clarify them. The key error to avoid is running methods as independent add-ons—without a clear overlap showing how one phase informs the other.
What makes a study “mixed methods” rather than just “qualitative plus quantitative”?
How does a concurrent (convergent) mixed methods design work in practice?
Why is exploratory sequential design especially useful when researchers don’t know what to ask on a survey?
What is the logic of explanatory sequential design?
What mistake undermines mixed methods designs most often?
Review Questions
- In what ways do concurrent and sequential mixed methods differ in how they connect qualitative and quantitative components?
- Give an example of how qualitative findings could be converted into quantitative survey items in an exploratory sequential design.
- What kinds of quantitative results would typically trigger qualitative follow-up in an explanatory sequential design?
Key Points
- 1
Mixed methods requires a planned relationship between qualitative and quantitative components, not just the presence of both.
- 2
Concurrent/convergent designs collect both types of data at the same time, but integration during analysis is what makes them truly mixed.
- 3
Sequential designs must be more than “one after the other”—the later phase should build on the earlier phase.
- 4
Exploratory sequential design uses qualitative findings first to develop quantitative measures when researchers don’t yet know what to ask.
- 5
Explanatory sequential design uses quantitative results first, then qualitative follow-up to explain patterns or clarify confusing responses.
- 6
A common failure mode is collecting both methods due to external pressure without a clear overlap or rationale for combining them.
- 7
Triangulation can be useful, but it does not automatically satisfy the logic of mixed methods design.