Research Design

From paradigms to the study matrix — a practical journey for design researchers

www.jaimeriveraphd.com

Agenda

~20'
The five research paradigms
Positivism → Post-positivism → Critical Theory → Constructivism → Participatory Research
~10'
My dissertation as a case study
Three stages, mixed methods, and the research questions cascade
~10'
Myths about design research
What they believe vs. what it is — debunking to liberate
~60'
Activity: Study Design Matrix
Each student builds their own matrix + 3-minute presentations

Why do paradigms matter?

A paradigm is a system of fundamental beliefs that guides how we understand reality, how we generate knowledge, and how we investigate it.

Each paradigm answers three different questions:

Ontology

What is the nature of reality?
Does an objective reality exist or is it socially constructed?

Epistemology

What is the relationship between the researcher and what is researched?
Can reality be known objectively?

Methodology

How should we investigate?
What methods are appropriate for generating knowledge?

Guba, E. G., & Lincoln, Y. S. (1994). Competing paradigms in qualitative research. In Handbook of Qualitative Research.
The five paradigms · 1 of 5

Five ways of seeing the world

Positivism

Ontology: Naive realism — reality exists and is apprehensible as it is

Epistemology: Dualist and objectivist — the researcher does not influence what is researched

Methodology: Experimental and manipulative — hypotheses verified as facts

Typical question: What is the measurable effect of X on Y?

Example: Measuring checkout flow efficiency with A/B testing
Post-positivism

...

Critical Theory

...

Constructivism

...

Participatory

...

Guba & Lincoln (1994). Popper, K. (1959). The Logic of Scientific Discovery.
The five paradigms · 2 of 5

Five ways of seeing the world

Positivism

Ontology: Naive realism

Epistemology: Dualist/objectivist

Methodology: Experimental

Empirical verification
Post-positivism

Ontology: Critical realism — reality exists but is imperfectly apprehensible

Epistemology: Modified objectivist — results are probably true

Methodology: Quasi-experimental, triangulation, falsification

Typical question: How likely is it that X causes Y?

Example: Evaluating usability with heuristics + testing + interviews
Critical Theory

...

Constructivism

...

Participatory

...

Popper, K. (1959). The Logic of Scientific Discovery. Guba & Lincoln (1994).
The five paradigms · 3 of 5

Five ways of seeing the world

Positivism

Ontology: Naive realism

Epistemology: Dualist/objectivist

Methodology: Experimental

Empirical verification
Post-positivism

Ontology: Critical realism

Epistemology: Modified objectivist

Methodology: Triangulation

Falsification, probability
Critical Theory

Ontology: Historical realism — reality is shaped by power, history, culture, economy, and gender

Epistemology: Transactional/subjectivist — the researcher's values influence the inquiry

Methodology: Dialectical — discourse analysis, ideological critique

Typical question: Who benefits? What is hidden?

Example: Analyzing how AI algorithms reinforce gender biases
Constructivism

...

Participatory

...

Marx, Horkheimer, Adorno, Marcuse. Guba & Lincoln (1994).
The five paradigms · 4 of 5

Five ways of seeing the world

Positivism

Ontology: Naive realism

Empirical verification
Post-positivism

Ontology: Critical realism

Falsification, probability
Critical Theory

Ontology: Historical realism

Dialectics, emancipation
Constructivism

Ontology: Relativism — multiple, socially constructed realities

Epistemology: Transactional/subjectivist — researcher and participant co-construct meaning

Methodology: Hermeneutic/dialectical — ethnography, grounded theory, phenomenology

Typical question: How do participants experience X?

Example: Understanding how designers construct meaning with AI tools
Participatory

...

Schwandt (1994). Guba & Lincoln (1994). Lincoln & Guba (1985). Naturalistic Inquiry.
The five paradigms · 5 of 5

Five ways of seeing the world

Positivism

Ontology: Naive realism

Empirical verification
Post-positivism

Ontology: Critical realism

Falsification, probability
Critical Theory

Ontology: Historical realism

Dialectics, emancipation
Constructivism

Ontology: Relativism

Interpretation, co-construction
Participatory

Ontology: Participatory reality — subjective-objective, co-created between mind and cosmos

Epistemology: Extended — experiential, presentational, propositional, and practical

Methodology: Cooperative — all are co-researchers and co-subjects

Typical question: How can we transform X together?

Example: Co-designing a community health system with communities
Heron, J. & Reason, P. (1997). A Participatory Inquiry Paradigm. Qualitative Inquiry, 3(3), 274-294.

Design as a paradigm integrator

Design research frequently operates across paradigms. A project may use constructivist ethnography to understand context, post-positivist experimentation to test a solution, and participatory design to co-create with communities.

The key is not choosing one paradigm and staying there — it's being aware of when and why you shift from one to another.

Positivism
Post-posit.
Critical Th.
Constructiv.
Participatory
DESIGN
Creswell, J. (2014). Research Design. Owen, C. (1998). Design Research: Building the Knowledge Base.
Real case study · My doctoral dissertation

Research Through Provocation

A Structured Process to Design New Information Technologies · IIT Institute of Design · 2023

STAGE 1

QUAL + qual

3 case studies with provocative prototyping (provotypes).

Projects: SkyWords · Asthma tool · Visual workflows
Paradigm: Constructivism + Participatory research
STAGE 2

QUAL

Provotyping tool applied in industry (Harman/Samsung).

Projects: Smart home · Autonomous car 2030
Paradigm: Post-positivism + Constructivism
STAGE 3

QUAN + QUAL

Controlled experiment. Repeated Measures ANOVA + interviews.

Framework: Self Determination Theory
Paradigm: Post-positivism + Constructivism
Schoonenboom, J. & Johnson, R. B. (2017). How to Construct a Mixed Methods Research Design.
?
Case study · Questions cascade

Questions don't emerge alone — they converge

Each field contributed a foundational question. The central question emerges from the intersection of all three.

METHODOLOGICAL INSIGHT

The best research questions emerge at the intersections between fields.

DESIGN

How can early prototyping be structured in the design process?

BEHAVIORAL

How can behavior change models guide the design project?

TECHNOLOGY

What are the key interaction elements for new technologies?

↓ converge into ↓

How can key interaction elements be manipulated to structure a design process based on provocations?

Paradigmatic awareness as a superpower

My dissertation operates across three different paradigms over three stages. This is not a mistake — it's a conscious methodological decision.

Rigor is not about staying in one paradigm — it's about knowing when and why you shift.

Paradigmatic progression

1
Explore → Constructivism

Provotypes to understand interactions in context

2
Apply → Pragmatism

Tool validated in real industry

3
Measure → Post-positivism

Controlled experiment with ANOVA

Mixed methods ≠ "doing both"
Mixed methods = articulating paradigms with intention

Main Research Areas
Research Questions Venn
Structured Provocation Model
Six Case Studies
Three Stage Methodology
Provotyping Tool
Smart Home Provotyping
Academic Settings
Controlled Experiment
ANOVA Results
Heuristics of Provotyping

What they believe vs. what is

Before building your study matrix, we need to deactivate some beliefs that block design researchers.

01
01

"Research through Design is not structured research"

RtD is structured research. The design artifact is not the final result — it's the vehicle through which new understanding is generated.

Structure comes from rigorous documentation of the process, design criteria, and systematic reflection.

Frayling, C. (1993). Research in Art and Design. Zimmerman, Stolterman & Forlizzi (2010).
02
02

"Design research = a design project"

A project solves a specific problem for a client. Research generates transferable knowledge to other contexts.

🎨 Design project

Solves a specific problem.
Result: artifact, service, or system.

🔬 Design research

Generates transferable knowledge.
Result: new, documented understanding.

03
03

"Design is unique and so is its methodology"

Design does have unique characteristics (wicked problems, abduction, reflexive praxis), but it shares methodological foundations with the social sciences.

We don't need to reinvent the wheel — we need to adapt it with awareness of our particularities.

Owen, C. (1998). Design Research: Building the Knowledge Base. Buchanan (1992). Cross (2001).
04
04

"Qualitative research is not rigorous"

Qualitative rigor exists — it's called trustworthiness: credibility, transferability, dependability, confirmability.

These are the equivalents of internal validity, external validity, reliability, and objectivity.

Tip for your thesis: Use member checking, audit trail, and triangulation to demonstrate rigor.

Lincoln & Guba (1985). Naturalistic Inquiry.
05
05

"Mixed methods = doing both"

Mixed methods requires a paradigmatic justification for integration. Having quantitative and qualitative data isn't enough — you must explain how and why they integrate.

Tip for your thesis: Define the sequence (concurrent, sequential), the weight (QUAL→quan), and the integration point.

Creswell & Plano Clark (2017). Schoonenboom & Johnson (2017).

The Study Design Matrix

A tool to align all components of your research into a coherent system.

Question
+ Unit of analysis
🎯
Objectives
General + specific
🌐
Paradigm
Ontology + epistemology
🔧
Method
Instruments + sources
📊
Analysis
Form of analysis

The question everything must answer:

Are all components aligned with each other?

Your turn

Open the interactive Study Design Matrix and build yours.

01
Define your research question and unit of analysis
02
Formulate objectives and select a paradigm
03
Choose method, instruments, and sources
04
Verify the alignment of all components
05
Present your matrix in 3 minutes

The template has presentation mode — use it when it's your turn to present.