What is Mixed Methods Research?
Mixed methods research collects and analyses both quantitative and qualitative data in one study, and integrates them, so that the combination answers the question better than either alone. Numbers show how much or how often; words show how and why. The integration is what distinguishes mixed methods from a study that simply contains a survey and some interviews.
Why Use a Mixed Methods Design?
- Explanation: interviews explain surprising survey results.
- Development: focus groups inform a questionnaire that is then tested at scale.
- Corroboration: two kinds of data checking each other, a form of triangulation.
- Completeness: complex problems, such as evaluating a health programme, need both outcomes and experiences.
The Three Core Designs
John Creswell and Vicki Plano Clark describe three core designs:
- Convergent: quantitative and qualitative data collected at the same time, analysed separately, then compared and merged.
- Explanatory sequential: quantitative first, then qualitative to explain the results (written QUAN → qual).
- Exploratory sequential: qualitative first to explore, then quantitative to test or measure what was found (QUAL → quan).
More complex designs embed these inside experiments, evaluations or participatory projects. Capitals in the notation show which strand has priority.
Example
A study of a university's new feedback system surveys 600 students on satisfaction and grades (quantitative), finds that satisfaction rose but grades did not, then interviews 20 students to learn why (qualitative). A table that sets the survey results beside the interview themes, called a joint display, shows how they fit together.
How to Conduct Mixed Methods Research
- Write a research question that needs both kinds of data.
- Choose a design and decide where integration will happen: in sampling, analysis or interpretation.
- Collect and analyse each strand with its own standards of quality.
- Integrate: compare, connect or build one strand on the other.
- Report what the combination showed that neither strand could alone.
Limitations
- Needs skills in both traditions, or a team that has them.
- Takes longer and costs more.
- Integration is hard to do well and easy to skip.
Mixed methods often draws on pragmatism: choosing methods by what answers the question. See research design.
Use case: AI for research paper writing