Integrating Numbers and Experience in Health Research

Mixed methods is more than collecting two data types

Mixed-methods research deliberately integrates quantitative and qualitative approaches to answer one scientific problem. Adding an open-ended survey item is not enough: each component needs a rationale, sampling plan, analysis and a prespecified point of integration.

Common designs

• Convergent: collect both forms of data in parallel, analyse separately and merge during interpretation.

• Explanatory sequential: quantitative results come first; qualitative work then explains selected findings.

• Exploratory sequential: qualitative findings develop concepts or instruments that are subsequently tested quantitatively.

Sampling and integration

Quantitative samples are planned around power and precision; qualitative samples around information richness and thematic adequacy. Explain how participants or data sources are connected. Integration can occur in design, sampling, analysis through joint displays, or interpretation.

Disagreement is informative

When components disagree, investigate differences in measurement, sample, timing and context rather than discarding one result. Divergence can reveal heterogeneity that an average effect conceals.

Quality criteria

Each component must meet its own methodological standards, and the study must demonstrate what integration adds beyond two parallel projects.

References

• NIH OBSSR Mixed Methods Resources.

• NIH Best Practices for Mixed Methods Research.

• GRAMMS reporting guidance.

Accessed 20 June 2026.