Skip to Content

Method selection

Selecting methods for your projectKey considerations in method selectionMixed methodsWhat reviewers look forRelated resources

This article is part of the Iris Ethics Knowledge Base; you can find a directory of all articles here.

Selecting methods for your project

Method selection is an expression of merit: research/evaluation with merit is designed or developed using methods appropriate for achieving the aims of the proposal (NS 1.1(b)). The chosen methods must be capable of producing the intended findings. Method selection also has direct ethical implications: different methods carry different risk profiles and impose different burdens on participants.

Key considerations in method selection

  • Fitness for purpose: the chosen method must be capable of answering the research/evaluation questions. A purely quantitative survey is unlikely to adequately answer questions about lived experience; an unstructured interview is unlikely to reliably estimate population-level prevalence.
  • Risk profile: different methods carry different ethical risk profiles. Face-to-face interviews introduce re-identification risks and potential for distress not present in anonymous surveys, but anonymous surveys that ask questions likely to cause discomfort or distress still require a plan for identifying and managing risk to a respondent. Observational methods, which have limited control over who participates and in what ways, raise distinct consent and privacy considerations. Method selection should account for these differences, identify relevant risks, and demonstrate how they are mitigated (NS 2.1.1).
  • Proportionality: the intrusiveness of data collection should be proportionate to the value of the research/evaluation. If a sensitive topic can be adequately explored using less intrusive methods and questioning, those methods should be preferred.
  • Validity: where validated instruments exist for measuring the constructs of interest, their use supports the merit of the research/evaluation (NS 3.1.2). Validated instruments may also reduce ethical risk by building in established safeguards to minimise harm and distress where possible. This does not mean other instruments may not be used, but research and evaluation using non-validated instruments should explicitly demonstrate how they support the project's objectives and minimise risks to participants.
  • Population appropriateness: methods must be appropriate for the characteristics of the participant population, including age, literacy level, language, disability, and cultural background.

Mixed methods

Mixed-methods designs that combine quantitative and qualitative approaches are common in evaluation and social research. Each component of a mixed-methods study should first be assessed independently for its ethical requirements and risk profile. The combination of methods should then also be assessed, since combining data sources can raise additional considerations around data linkage and re-identification. For example, a person's responses in a semi-structured interview may provide information that could identify them in an administrative dataset. Practitioners need to identify the risks that could result from this and put safeguards in place to prevent it, such as de-identifying interview information before analysis, or working only with aggregated and/or suppressed administrative datasets.

What reviewers look for

Reviewers assess whether the methods described can actually answer the questions posed, not just whether the method is common in the field. In evaluation and social research applications, reviewers typically look for: an explicit link between each method and the research/evaluation question it addresses (a matrix is a useful way to show this; see Demonstrating research merit); a description of the risk profile of the method, not just its logistics; a justification where a non-validated or bespoke instrument is used; and, in mixed-methods designs, an account of how findings from different methods will be combined and what re-identification risk that combination might create.

Related resources

This page was last updated on 16 September 2026.

Was this helpful? We'd appreciate your feedback!

This article makes reference to the National Statement on Ethical Conduct in Human Research 2025. It is intended as practical guidance and does not constitute legal advice.

AI Disclosure: Resources in this library were prepared using Large Language Models with input from Iris Ethics staff who guided the scope and design. Subsequent revisions and final versions were developed and approved by Iris Ethics staff.