It’s time to reach out to your target group and ask them directly what you need to know for your project or you might also need to analyse existing data:

Checklist for identifying Methods

1. Review What You Already Know
   • Have you done a basic or detailed literature review?
   • Are there still open questions or gaps in knowledge?
2. Define What You Still Need to Learn
   • What exactly do you need to know to move your project forward?
   • Is the missing information about user needs, daily experiences, health issues, etc.? For example: What do users need from a new tech product? What everyday challenges do people with diabetes or endometriosis face?

3. Choose How to Get the Information

   Ask Your Target Group for

   • Interviews – for in-depth insights
   • Surveys – to reach more people and collect structured data
   • Focus groups – for group discussions and shared views
   • Observations – to see real-life use or behaviour
   • Samples/Medical data – to gather biological information (e.g. blood, cells, CT scans)
4. Use Existing Data
   • Public databases or statistics (e.g. on mobility patterns, energy consumption)
   • Past studies or reports
   • Internal company/user data (if available)

If you decide to ask your target group, you can use methods from the social sciences such as:

• (Expert) Interviews (see, for example for an overview here or here, and for some practical guidance here)
• Focus Groups (see, for example here or here)
• Surveys (see, for example)
• Observations (see, for example here or here)
• Experiments (see, for example here or here)

Please also take a look at the external resources at Support on Methods and Data Analysis and conduct your own research on your chosen method if you are looking for something more specific.

It is important do identify user needs in a professional way. Therefore, take care to have a Social Scientist in your team!

If you decide to use existing data, you will probably use:

• Quantitative Secondary Data Analysis
• Content Analysis

If you are looking for more co-creative, participatory methods please have a look at step 3.

Reflect on your choice of method: Make sure that your method fits your target group. For example, for marginalised groups, observations might be too sensitive. If you’re unsure whether your tool fits the target group, ask someone from the group or someone who knows them well. For example, if you want to interview people in a care home, the staff could help you understand what works best.

Consider the real-life circumstances of your target group. For example, think about how care responsibilities might affect scheduling. Please do some research on how this method can be applied in an inclusive way. If you use existing survey tools—such as question batteries from other questionnaires—make sure to briefly check in the literature whether the method carries any known bias(es).

Checklist for Designing Inclusive Surveys and Interviews

When you collect your data, you need to develop data collection tools/instruments such as interview guides, observation protocols, questionnaires etc. When creating these toolskeep these questions in mind:

• Is the language clear and simple? Avoid technical terms that might confuse people, especially older adults or people with disabilities.
• Do I ask about relevant categories? Always include your key characteristics (like gender or age) in your questions so you can analyse the results properly. For example, if you're studying gender differences, you need to ask people about their gender.
• Is the language inclusive? Use gender-sensitive wording and respect how people describe themselves. For example, offer open options for gender and never make sensitive questions mandatory. For example, you can find a guideline here by Morgan Klaus Scheuerman, here for survey design and gender by the gendered innovation website or here a more detailed report here by Suzanne Thornton et al. (2021).
• Am I being clear and specific enough? When formulating your questions, make sure they are easy to understand and not too general. If your wording is unclear, people might understand the question in different ways and the results might not be reliable.
• Are there accessibility needs? E.g. make sure tools work with screen readers or allow flexible times for interviews, especially for caregivers.
• Do the questions fit all users? Check if all parts of your target group can relate to your questions, even those with overlapping identities.
• Am I avoiding stereotypes? Be mindful of wording and visuals so you don’t reinforce social biases.

Different target groups may have different needs, so you might need to create more than one version of your tool – for example, in different languages.

At this stage, please also consider how you will analyse your data, as this may affect your design. For example, at what level of the scale do you want to analyse your data? This will influence how you design your questions. Similarly, if you would like to use a specific method of analysing interview data, your guidelines and interviewing style will be different. There are more flexible and more structured methods. For example, there is the narrative interview and Schuler's Multimodal Interview. These are rather contrasting examples.

If you’re using an intersectional approach (looking at how different factors combine, like gender and disability), check what that means for your methods. We offer some resources, but you may need to explore.

• Del Toro, J., & Yoshikawa, H. (2016). Invited Reflection: Intersectionality in Quantitative and Qualitative Research. Psychology of Women Quarterly, 40(3), 347-350.
• Zhang, B., Chang, B., & Du, W. (2021). Employing Intersectionality as a Data Generation Tool: Suggestions for Qualitative Researchers on Conducting Interviews of Intersectionality Study. International Journal of Qualitative Methods, 20.
• Christensen, A. D., & Jensen, S. Q. (2012). Doing Intersectional Analysis: Methodological Implications for Qualitative Research. NORA - Nordic Journal of Feminist and Gender Research, 20(2), 109–125.

• Descriptive Data Analysis: Check the distribution of demographic variables (e.g., gender, ethnicity) in your data.
  • Are certain groups underrepresented?
• Fairness Metrics: Use statistical measures to evaluate your data:
  • Demographic Parity: Positive predictions should be equally distributed across groups.
  • Equal Opportunity: True positive rates should be equal.
  • Equalized Odds: Both true and false positive rates should be balanced.
  • Disparate Impact: The ratio of positive decisions across groups should be close to 1.
• Test Subgroup Performance: Evaluate your data separately for key subgroups (e.g., women, minorities).
  • Does the model perform worse for specific groups?
• Run Counterfactual Fairness Tests: Change only a sensitive attribute (e.g., gender) and check if the prediction changes.
  • Predictions should remain stable if all other features are the same.
• Consider External Audits: Have third parties (e.g., NGOs, researchers) review your model for potential bias.
  • Independent audits can reveal hidden issues.
• Conduct Manual Error Analysis: Review misclassifications and edge cases.
  • Look for systematic errors or fairness-related anomalies.

If you have quantitative data, start by describing your sample: How many women, men, or gender-diverse people took part? What are their ages or other key traits? How many female, male, gender diverse skin samples? Cross tables can help show this clearly.

Next, look at your main questions. For example, if you're studying public transport use by gender, begin with charts or cross tables. You can compare groups (e.g. women, men, gender-diverse), but also look at differences within each group. Gender differences for example might change depending on age, ethnicity, social class, or other factors. In long-term studies, you can observe how gender-related experiences and patterns develop or shift over time. If you are using existing data, think about the social or cultural context it came from – some bias may be built in.

If you're doing statistical testing:

• If gender or diversity is part of your hypothesis, focus your analysis on it.
• If not, use these dimensions as control variables and include them in your interpretation.

If you work with intersectionality, you can find some support here:

• The CIHR provides an article on quantitative intersectional analysis.
• Else-Quest, N. M., & Hyde, J. S. (2016). Intersectionality in Quantitative Psychological Research: II. Methods and Techniques. Psychology of Women Quarterly, 40(3), 319-336.
• Dubrow, J.K., Ilinca, C. (2019). Quantitative Approaches to Intersectionality: New Methodological Directions and Implications for Policy Analysis. In: Hankivsky, O., Jordan-Zachery, J.S. (eds) The Palgrave Handbook of Intersectionality in Public Policy. The Politics of Intersectionality. Palgrave Macmillan, Cham.
• Greta R. Bauer, Siobhan M. Churchill, Mayuri Mahendran, Chantel Walwyn, Daniel Lizotte, Alma Angelica Villa-Rueda (2021): Intersectionality in quantitative research: A systematic review of its emergence and applications of theory and methods, SSM - Population Health, Volume 14, 100798, ISSN 2352-8273,

As with quantitative data, start by looking at your sample and its key characteristics. Then:

• Use intersectional coding: Use codes that capture intersecting social categories (e.g. gender, race, class, disability) rather than treating gender as a stand-alone variable.
  • This avoids oversimplified or stereotypical interpretations.
• Let the participants' language guide the interpretation: Respect how respondents describe their identities, experiences and realities.
  • Avoid imposing normative categories or assumptions.
• Identify power structures in the narratives: Look for how power relations (e.g. gender hierarchies, discrimination, marginalisation) shape participants' experiences and statements.
  • This brings structural inequalities into focus.
• Include minority and marginalised voices: Ensure that underrepresented perspectives are not diluted or sidelined in thematic analysis or reporting.
  • Actively look for patterns in these voices, rather than treating them as outliers.

We advise you to do some research into the method you are using to analyse your data, such as grounded theory, qualitative content analysis, thematic analysis or narrative analysis, and how inclusion comes into play there.

If you're doing interviews with an intersectional focus, you can find support resources here:

• Cuádraz, G. H., & Uttal, L. (1999). Intersectionality and In-depth Interviews: Methodological Strategies for Analyzing Race, Class, and Gender. Race, Gender & Class, 6(3), 156–186.
• Christensen, A. D., & Jensen, S. Q. (2012). Doing Intersectional Analysis: Methodological Implications for Qualitative Research. NORA - Nordic Journal of Feminist and Gender Research, 20(2), 109–125.