PART 2. From Evidence to Judgment
Different Evidence Answers Different Questions
A common research mistake is to treat all evidence as interchangeable. Ten interview quotes do not become quantitative evidence because there are ten of them. A survey percentage does not explain why respondents behaved as they did. A trend report may describe a market but not the experience of a specific community. A scientific paper may establish a material property without telling a designer whether users will accept the resulting object.
Research literacy therefore requires a simple question: what kind of claim am I trying to make? If the claim is about prevalence, the evidence should support prevalence. If it is about meaning, the method should allow people to describe experience. If it is about behaviour, direct observation or usage data may be stronger. If it is about technical feasibility, expert knowledge and testing become essential.
Strong design research often triangulates. A team might combine public data about heat with street observations, interviews with residents and temperature measurements in a prototype. None of these sources replaces the others. Together they create a more reliable picture because the weaknesses of one form of evidence can be challenged by another.
Triangulation also makes disagreement productive. If people say they want one thing while behaviour suggests another, the aim is not to decide which source is ‘correct’ immediately. The difference becomes a research question. Perhaps aspiration and action are separated by cost, time, habit, stigma or infrastructure. That gap may be where design can intervene.
Secondary Research Is Context, Not Decoration
Secondary research is often reduced to a page of statistics placed at the beginning of a presentation. Used well, it has a much more demanding role. It can reveal history, policy, precedent, technical constraints, demographic change and earlier attempts to solve the same problem. It helps students avoid rediscovering knowledge that already exists and prepares them to ask better questions in the field.
IDEO.org’s guidance on secondary research describes it as a way to build a foundation of context before and during fieldwork. For students, this means looking beyond the first search result. Who produced the information? When? For what purpose? Does the source describe the same population, geography and conditions as the project? Is the number original, or repeated from another publication without context?
The aim is not to make every design project academically exhaustive. It is to create a chain of evidence that can be traced. A student should be able to distinguish what was learned from a source, what was observed directly and what remains an interpretation or hypothesis.
Source:
https://www.designkit.org/methods/secondary-research.html
Case Study: Todd Bracher and Science-Based Design
Todd Bracher’s 2026 monograph Observations, Research, and Design presents more than twenty-five years of work through a science-based approach to design. Phaidon describes his practice as translating scientific discoveries into solutions across furniture, product, interior design and consulting, rather than treating research as an appendix to form-making.
The example is useful because it expands what design research can mean. A designer may begin with biomechanics, light, acoustics, material behaviour or another body of specialist knowledge, then translate that evidence into an object people can use. Research is not limited to asking users what they want; it can connect creative practice to disciplines that explain how bodies, environments and systems behave.
For students, the lesson is to search for the knowledge a project actually requires. A seating project may need ergonomics. A lighting project may need visual perception. A circular product may need failure data and disassembly analysis. The designer does not need to become the expert in every discipline, but must become capable of finding, questioning and working with expertise.
Source:
https://www.phaidon.com/en-us/products/observations-research-and-design
From Notes to Insight
Collecting information is only half the work. Research becomes useful when the designer can make sense of it without forcing it into a preferred concept. This is where synthesis matters: grouping observations, comparing contradictions, mapping journeys, identifying recurring barriers, tracing cause and effect and deciding which findings are strong enough to shape a brief.
An insight is not simply an interesting quote. It connects evidence to an implication. ‘Three participants said they dislike the app’ is a finding. ‘People avoid the app when the task feels high-risk because they want reassurance from a person before submitting information’ is a possible insight — but it still needs to be tested against more evidence. The second statement is useful because it suggests what the design needs to support.
Students should resist producing too many insights. A project becomes stronger when a small number of findings clearly change priorities. Each important insight can be linked to a design criterion: what must the solution enable, avoid, preserve or make possible?
This traceability is a powerful defence against aesthetic drift. When a concept changes, the team can ask whether the new direction still responds to the evidence or whether it has become attractive for reasons unrelated to the problem.
Who Is Missing from the Research?
Research can look participatory while repeatedly hearing from the easiest people to reach. Students recruit classmates, confident digital users, people who speak the same language or participants already enthusiastic about the topic. The resulting design may appear user-centred while representing only a narrow part of the user population.
The UK Government Service Manual explicitly asks teams to include users with different levels of digital access and confidence, disabled users and people who may need support. It also recommends researching people in the environments where they actually use services when that context matters. The broader principle applies to all design: inclusion begins in the research sample, not at the accessibility checklist after the concept is finished.
Students can ask who would experience the highest cost if the design failed. Who is least visible in conventional recruitment? Who is affected without being a direct ‘user’? A public-space project may need maintenance staff as well as visitors. A food system may involve delivery workers, shopkeepers and waste handlers. Expanding the research frame often changes what counts as a successful solution.
Source:
https://www.gov.uk/service-manual/user-research/understanding-users-who-dont-use-digital-services
Research Ethics: People Are Not Raw Material
Creative research can become extractive when people are treated as sources of stories, photographs and emotional detail rather than participants with rights and agency. A powerful quote does not automatically belong in a portfolio. A photograph taken during fieldwork is not automatically available for public presentation. Sensitive information may be useful to understand a problem but inappropriate to display.
Students should know what participants have agreed to, explain how material will be used, avoid collecting unnecessary personal data and create space for people to refuse documentation. When research involves vulnerable groups, cultural difference or unequal power, the method itself should be designed with care.
Ethics also improves the quality of research. People are more likely to speak honestly when they understand the purpose of the project and feel that boundaries are respected. Research literacy therefore includes not only methods for learning, but responsibility for the relationship that makes learning possible.
AI Can Accelerate Research — and Flatten It
Generative AI can be extremely useful at the beginning of research. It can suggest vocabulary, identify areas to investigate, compare explanations, help structure interview questions or surface possible precedents. It can also help students enter an unfamiliar field quickly enough to know what they need to investigate next.
The risk is that fluency can be mistaken for evidence. A concise answer may merge strong and weak sources, remove disagreement or present a disputed interpretation as settled. It can also encourage students to remain at the level of what is already easily describable rather than encounter a place, person, object or system directly.
A research-literate workflow therefore treats AI output as a map of questions, not a final evidence base. Important claims should be traced to original or authoritative sources. Uncertainty should be preserved rather than smoothed away. If a source cannot be found, the claim should remain provisional. If the project concerns lived experience, generated summaries should not replace contact with the people involved.
AI becomes more valuable when students already have a research structure. They can ask it to generate counter-hypotheses, reveal missing stakeholder groups or suggest what evidence would disprove an assumption. The human role is not simply to approve the answer; it is to decide what deserves to be believed and what still needs to be learned.
Case Study: UNESCO and Human-Centred AI in Research
UNESCO’s guidance on generative AI in education and research is built around a human-centred approach. It emphasises human agency, ethical validation and the need for institutions to examine risks, bias, privacy and pedagogical appropriateness rather than treating the availability of a tool as sufficient reason to use it.
UNESCO’s AI Competency Framework for Students develops the same principle through critical judgement: students should be able to understand, apply and create with AI while examining its ethical, social and technical implications. For design education, this aligns closely with research literacy. The student must remain responsible for the question, the evidence, the interpretation and the consequences of using the information produced.
This changes how AI fits into the studio. It can expand a research process, but it should not erase provenance. Students can show where AI helped generate directions, which claims were verified independently and where human observation, expert knowledge or primary sources changed the initial machine-generated picture.
Source:
https://www.unesco.org/en/articles/guidance-generative-ai-education-and-research
https://www.unesco.org/en/articles/ai-competency-framework-students
Knowing When to Stop Researching
Research can also become avoidance. Students sometimes continue gathering references because making a decision feels risky. The goal is not perfect knowledge. Design always moves forward with uncertainty, constraints and incomplete evidence.
A useful stopping rule is decision-based: do we understand enough to choose the next experiment? Have the highest-risk assumptions been tested? Are new interviews repeating what we already know, or revealing new patterns? Is an unanswered question likely to change the concept, or is it merely interesting?
Research and making should alternate. A prototype creates new questions; fieldwork changes the prototype; technical testing exposes another constraint. The best projects do not have a research phase followed by a design phase. They create a learning loop in which evidence and making continuously correct each other.
How Design Students Can Practise Research Literacy
Research literacy can be built through small habits. Before searching for references, students can write three facts they know, three assumptions they are making and three questions that would change the project if answered differently. This simple separation prevents assumptions from hardening into facts.
They can keep an observation notebook in which concrete description is written separately from interpretation. They can interview one person in context rather than five people in a classroom. They can choose one secondary claim and trace it back to its original source. They can deliberately look for one participant or stakeholder whose experience is likely to contradict the dominant story.
During synthesis, students can attach evidence to each major design criterion. If a criterion has no evidence, it can be labelled as a hypothesis and tested through a prototype. If evidence conflicts, the conflict can remain visible instead of being edited out of the presentation.
Most importantly, students can practise changing their minds. A research process has succeeded when it alters the brief, the priorities or the concept. If research only confirms what the designer wanted to do at the beginning, it may be worth asking whether the questions were genuinely open.
Research in the Portfolio
A research-led portfolio should not become a wall of sticky notes. Employers and educators need to see the relationship between evidence and decisions. A concise research story can show the initial assumption, the question, the method, the critical finding and the design change that followed.
Students can include a field observation that overturned an assumption, a map that revealed a hidden stakeholder, a failed prototype that produced new evidence or a source that changed a material decision. The point is not to prove that a large amount of research occurred. It is to demonstrate judgment about what mattered.
This makes research a visible creative capability. The portfolio shows that the designer can enter an unfamiliar problem, build enough understanding to act, recognise the limits of that understanding and revise decisions when reality disagrees with the concept.
What Design Schools Can Teach
Design schools can strengthen research literacy by grading the quality of questions as well as the quality of outcomes. Briefs can require students to identify assumptions explicitly, justify why a method fits a question and show how evidence changed the project.
Critique can ask: What do you know? How do you know it? What would make you change your mind? Who did you not speak to? Which claim comes from direct observation, which from a source and which is still a hypothesis? These questions make research part of studio culture rather than a separate theory module.
Schools can also protect time for fieldwork, source evaluation, synthesis and ethical reflection. In an environment where visual output can be produced faster than ever, education has an opportunity to reward the slower capability underneath it: learning how to notice, verify, interpret and decide.
Key Takeaways for Creative Students
- Begin with what you need to learn, not with the research method you want to use.
- Separate facts, observations, interpretations, assumptions and hypotheses.
- Choose evidence that matches the kind of claim you are making.
- Use interviews to understand experience, observation to understand behaviour, and secondary research to understand context and scale.
- Triangulate important findings instead of asking one source or method to do everything.
- Treat contradictions as useful research questions rather than inconvenient data.
- Include people who are hardest to reach and those who bear the cost if a design fails.
- Use AI to generate questions and directions, but verify important claims and preserve provenance.
- Translate research into design criteria and show how evidence changed decisions.
- Remember that research is complete enough when it supports the next responsible experiment, not when uncertainty has disappeared.
Looking Ahead
Creative practice is often described as the ability to produce original answers. In a world of abundant images, tools and generated possibilities, the quality of the question may become just as important. A designer who can ask what others have overlooked gains access to problems and possibilities that cannot be reached by faster execution alone.
Research literacy gives students a way to move through information without becoming overwhelmed by it. It helps them recognise the difference between a source and a summary, between a quote and an insight, between a pattern and an explanation. It also makes room for uncertainty, because an unanswered question can be named rather than disguised by a confident presentation.
For the next generation of artists, designers and architects, research is not a retreat from creativity. It is one of the ways creativity becomes more specific, more surprising and more accountable to the world it enters.
The strongest project may therefore begin not with the most original idea, but with a better act of attention: a question precise enough to reveal something that the brief, the reference library and the algorithm did not already know.