What is mixed methods research?
Also called: Mixed-methods research, Mixed methods design
Mixed methods research is a research approach that combines qualitative methods, such as interviews, with quantitative methods, such as surveys or product analytics, in one study, so the numbers show how much and the conversations show why.
Updated
What counts as mixed methods
Running a survey one quarter and some interviews the next isn't mixed methods on its own. The term means the two strands are planned together and their results are integrated: one shapes the other, or both are compared to reach a single conclusion. John Creswell and Vicki Plano Clark's work on mixed methods design is the usual reference for the three core designs below.
The three core designs
- Explanatory sequential: quant first, then qual. Product analytics show 40% of trial accounts never invite a teammate. Interviews with eight of those accounts explain why. This is the most common design in product teams, because the data already exists.
- Exploratory sequential: qual first, then quant. Interviews surface six reasons people switch tools. A survey to 300 customers then measures how common each reason is. Use it when you don't yet know what to measure.
- Convergent: both at once. Run a usability test and a satisfaction survey in the same release window, analyse them separately, then compare. Where they disagree, you've found something worth digging into.
How to plan a mixed methods study
- Start from one research question both strands answer, not two separate questions bolted together.
- Decide the order and what hands over. In a sequential design, the first strand should produce something concrete for the second: a list of segments to interview, or themes to turn into survey options.
- Recruit with the link in mind. If interviews will explain survey answers, recruit interviewees from survey respondents who agreed to follow up.
- Plan the integration step. Put both sets of results on one page, by research question, and say where they agree and where they don't.
When it's worth the extra effort
Mixed methods takes longer than a single method, so save it for decisions that need both scale and explanation: sizing an opportunity before building, understanding a drop in a key metric, or defining what goes into an MVP. For a narrow usability question, a single method is usually enough.
Go deeper
- Defining your MVP with mixed methods research
A worked use case: scoping an MVP with interviews and surveys.
- Quant research for qual researchers
Adding quantitative methods if your background is qualitative.
- UX research methods guide
Choosing which qualitative and quantitative methods to combine.
- Survey research guide
Designing the quantitative strand of a mixed methods study.
- Surveys in Great Question
Run the survey strand with your own customers.
Questions
What's the difference between mixed methods and triangulation?
Triangulation is a reason to use more than one source: checking a finding from different angles. Mixed methods is a study design that combines qualitative and quantitative strands. A convergent mixed methods study is one way to triangulate, but you can also triangulate with two qualitative methods.
Do you need a big sample for mixed methods research?
Each strand needs the sample its method calls for. The survey strand needs enough responses for the precision you want, and the interview strand needs enough sessions to reach saturation per segment. Mixed methods doesn't change those numbers, it just runs both.
Can one researcher run a mixed methods study?
Yes. Most product researchers run both strands themselves. If your background is mostly qualitative, start with an explanatory sequential design using product data your analytics team already tracks.
Run the study, not just the method
Great Question handles recruiting from your own customers, scheduling, incentives, interviews, tests and analysis in one place.
No credit card required.
Related terms
Full glossary- Triangulation
Triangulation is checking a research finding against more than one source, method or researcher, so you only act on it when different kinds of evidence point the same way.
- Generative research
Generative research is research done before there is a solution to test, to understand people's needs, behaviours and problems well enough to decide what to build. It produces opportunities and questions, not a verdict on a design.
- Evaluative research
Evaluative research tests an existing design, prototype or product against a goal, to find out whether people can use it and whether it does what the team intended. It checks a solution, where generative research looks for the problem.