Analytical & Experimental Methods
The seven research strategies — and which question each fits.
Different questions need different strategies. Groat & Wang organise architectural research in three tiers — systems of inquiry, strategies and tactics — and catalogue seven strategies: interpretive-historical, qualitative, correlational, experimental and quasi-experimental, simulation, logical argumentation, and the case study. Learn each strategy's question and example, the difference between a true and a quasi-experiment, and why correlation is not causation. Try the research-strategy explorer.
Learning objectives
By the end of this lesson, you will be able to — mapped to the course outcomes for Research Methods & Field Studies:
Explain Groat & Wang's three tiers — systems of inquiry, strategies and tactics.
Describe the seven research strategies and the questions each fits.
Distinguish a true experiment from a quasi-experiment.
Explain why correlation does not imply causation.
Three tiers, one question first
A worldview frames the strategy; the strategy is the overall design; tactics are the techniques. Pick the strategy to fit the question — not the other way round.[1]
Inquiry, strategy, tactic
Groat & Wang separate three levels. A SYSTEM OF INQUIRY is the worldview/paradigm (postpositivist, constructivist…). A STRATEGY is the overall research DESIGN — the seven below. A TACTIC is a specific technique for collecting or analysing data — an interview, a survey, a simulation run, a statistical test. One strategy can use many tactics, and the same tactic (an interview) can serve different strategies. MISCONCEPTION→correct: 'a method is the whole approach' — pick the strategy to fit the QUESTION, then choose tactics to serve the strategy.[1]
The seven strategies
From interpretive-historical and qualitative through correlational, experimental, simulation and logical argumentation to the case study — each has its own question, data and rigour.[1, 2]
Meaning and context
QUALITATIVE research asks how and why people experience a setting — interviews, observation, thick description and thematic coding in real settings; its rigour comes from TRIANGULATION, member checking and audit trails (Lincoln & Guba's credibility/transferability/dependability/confirmability). INTERPRETIVE-HISTORICAL research asks what something meant in its context — archives, drawings and artefacts interpreted with internal and external source criticism. MISCONCEPTION→correct: 'qualitative research is just anecdote' — it has its own demanding standards of trustworthiness.[1]
Match a strategy to a question
Pick a strategy and read its core question, typical data, when to use it and a concrete architecture example — then ask which one your own question would call for.
Research strategies · pick one
Qualitative
Core question: How and why do people experience and use this setting?
Data & tactics: Interviews, observation, focus groups, document analysis; thick description and thematic coding in naturalistic settings.
When to use: Exploratory topics, understanding behaviour and meaning, building theory.
Architecture example: Observing and interviewing residents of a slum-rehab block to understand how the new layout reshapes community life and privacy.
The strategy follows the question — and strong research often combines more than one.
At a glance
| Aspect | Correlational | Experimental |
|---|---|---|
| Question | Correlational: are they related? | Experimental: does X cause Y? |
| Manipulation | Correlational: none | Experimental: manipulate the IV |
| Assignment | True experiment: random | Quasi-experiment: pre-existing groups |
| Establishes | Correlational: association | Experimental: causation (if controlled) |
| Realism | Lab experiment: high control | Quasi/field: high ecological realism |
Key terms
Worldview / overall research design / specific technique — Groat & Wang's three tiers.
Meaning-focused inquiry — interviews, observation, thick description; rigour via triangulation.
Measures the relationship between naturally-occurring variables — association, not cause.
Random assignment vs comparison of pre-existing groups.
Studying behaviour through validated physical or digital models.
In-depth study of a bounded instance; generalises to theory, not a population.
Studio task
Take one architectural topic — say, naturally-ventilated classrooms. Write a researchable question that would best suit EACH of three different strategies (qualitative, correlational, simulation), and name the data each would collect. Then explain, in two sentences, why you cannot answer “does the new ventilation design cause better concentration?” with a correlational study alone.
Self-assessment
1. The difference between a TRUE and a QUASI-experiment is —
2. Which strategy best answers 'how do residents experience this redeveloped block'?
3. A case study generalises mainly to —
Recap
References & further reading
- [1]Linda Groat & David Wang, Architectural Research Methods (2nd ed., Wiley 2013) — the seven strategies, three tiers, tactics.
- [2]Bent Flyvbjerg, 'Five Misunderstandings About Case-Study Research' (Qualitative Inquiry, 2006).
- [3]Robert K. Yin, Case Study Research and Applications (Sage) — multiple evidence, analytic generalisation.
- [4]John W. Creswell, Research Design — quantitative, qualitative and mixed strategies.
Further reading
- Linda Groat & David Wang — Architectural Research Methods.
- Robert K. Yin — Case Study Research and Applications.
- John W. Creswell — Research Design.
Sources gathered and fact-checked June 2026. Published values vary by source, sample and method — treat as indicative and confirm against the cited standard before structural use.