Research & Evaluation
Our social research and evaluation teams combine rigour with depth. Take a look at the methodologies we use - usually woven together to suit context.
Theory of Change
Theory of Change makes explicit the assumptions about how interventions create impact, then tests whether they hold. We map causal pathways, identify necessary intermediate outcomes, and surface contextual conditions required for mechanisms to work.
We treat Theory of Change as reflective practice, revisiting assumptions as complexity reveals itself. The power lies in conversations about what we think will work, why, and for whom: not in producing perfect diagrams. We've developed Theories of Change for national Just Transition frameworks and complex social interventions in EU industrial settings, always treating them as living documents that evolve with learning.
Measuring and evaluating success in the Scottish Just Transition
Drabble, D. and Simeone, L. (eds.) (2021). Strategic Thinking, Design and the Theory of Change. London: Routledge
Scenario building
We use scenario building to help organisations prepare for multiple possible futures rather than optimising for one imagined outcome. We ask: what could happen, and how would we respond? Scenario building creates coherent narratives of how different contextual factors might interact to shape distinct futures over the medium to long term. We identify key drivers of change, understanding their uncertainties and interdependencies, then develop scenarios that stretch strategic thinking beyond familiar patterns.
Scenarios aren't predictions or preferences: they're structured explorations that reveal what might otherwise remain invisible, helping leaders test strategies against divergent futures. We've used scenario approaches to support strategic planning in rapidly changing policy environments, including Europe's labour market trends aligned with the European Commission's Strategic Foresight process.
Realist evaluation
Realist evaluation uncovers the underlying causal mechanisms that explain why interventions succeed or fail in contexts, referencing the often hidden reasoning and responses of participants. We ask: what works, for whom, in what circumstances, and how? We build and test context-mechanism-outcome configurations iteratively, using mixed methods to understand generative causation in real-world complexity.
A parenting programme might work brilliantly in one community and poorly in another, not because the intervention changed, but because different social norms, resources or relationships triggered different mechanisms. Realist evaluation reveals not just whether something works, but how and why it works differently across varied settings. We're currently applying theory-based realist evaluation to understand AI tool implementation across European labour markets.
AI implementation research across European labour markets
HM Treasury (2020). Magenta Book Supplementary Guide: Realist Evaluation
Mixed methods research
Numbers tell part of the story, and the views and experiences of individuals tell another. We integrate quantitative methods (surveys, statistical analysis, validated measurement tools) with in-depth qualitative work (interviews, focus groups) to understand what's happening and why it matters.
Survey data might reveal patterns across hundreds of participants whilst interviews and focus groups surface the dynamics behind those patterns. This integration allows us to move between breadth and depth, between measuring outcomes and understanding the human processes that generate them. We've combined large-scale surveys across global tech companies with focus groups and systematic literature reviews and have the experience and capacity to launch projects rapidly when clients need.
Developmental evaluation
We evaluate in real-time as interventions unfold, because complex initiatives don't wait for neat endpoints to adapt and learn. We facilitate learning and reflection to support adaptive decision-making through iterative inquiry. Rigorous evidence-gathering takes place, enabling innovations to evolve intelligently as contexts shift and insights surface.
The evaluation is part of the innovation process, not a judgement delivered afterwards. Our developmental evaluation with international social change organisations facilitated learning spaces as each adapted wellbeing interventions to their unique cultural contexts.
Participative research
Lived experience shapes our work from inception. Through sustained partnerships with people who may be affected by research processes and outcomes, we explore how power, anxiety and unconscious dynamics affect collaboration. We co-design questions, methods and interpretation that bring collective knowledge to life.
Conflicts, competing perspectives and power asymmetries are surfaced and worked with, not smoothed over. We don't ignore messy realities. Our learning partnerships are built on sustained relational practice that challenges who holds expertise and how change happens. We've co-produced evaluations with young people with lived experience of mental health issues, women affected by domestic violence, industry partners struggling to weigh up change.
Contribution analysis
When your intervention is emerging amidst interacting systems, responding to constantly evolving contexts, traditional evaluation won't tell you what works. We use contribution analysis to build a credible story of what's shifting, working with the uncertainties. Through iterative inquiry with stakeholders and those affected, we trace how your activities contribute to change across complex, shifting environments where simple cause and effect doesn't apply.
This approach reveals plausible causal contributions rather than definitive proof, acknowledging the messy reality of transformation processes. We've used this to evaluate large-scale transformation change processes across multiple sectors.
Junge, K., Cullen, J., & Iacopini, G. (2020). Using contribution analysis to evaluate large-scale, transformation change processes. Evaluation, 26(2), 227-245
What is our approach to evaluation?
Action learning
We work alongside your team as co-investigators, not distant observers. Action learning at TIHR is a blend of inquiry and intervention, grounded in organisational learning and practical application. Through iterative cycles of planning, acting, observing and reflecting, we surface and address the hidden dynamics that often hinder change: the unspoken tensions, defensive routines and anxieties about uncertainty.
We explore the relationship between the individual, the group, and the wider organisational system. Combining systems thinking with depth psychology, we’ve supported change in mining industries, the EU’s clothing sector, health leadership teams and global IT companies. Our teams understand the unconscious processes influencing behaviour and decision-making and build your capacity to think clearly under pressure and create lasting transformation.
What is Action Research?
Menzies, I. E. P. (1960). A Case-Study in the Functioning of Social Systems as a Defence against Anxiety
Socio-Technical Systems
Socio-technical systems (STS) are work systems — like supermarket self-checkouts, hospital wards, and sales organisations — in which the social and technical elements must be jointly designed and optimised if the system is to perform well.
The principle originates in the Institute's 1950s coalfields research, where Eric Trist and colleagues demonstrated that new technologies fail when they fracture the social structures that make work meaningful. Subsequent studies across Vauxhall Motors, Courtaulds and the process industries extended this into high-automation environments.
The founding principle — that neither subsystem can be improved at the expense of the other — remains the working grammar of socio-technical practice today.
Recent scholarship has extended this into balanced optimisation, which treats demands from the wider ecosystem as equally pressing; and adaptive autonomy, recognising that in human-AI systems, autonomy is not fixed but distributed and renegotiated as tasks and contexts shift.
We bring an STS lens wherever new tools, agents or platforms are reshaping how people work. We are developing a multi-level STS evaluation framework for complex human-AI systems to promote effective labour market policy in Europe. We are also pioneering experiential and arts-based approaches to STS learning, including participatory performances such as Faust Shop and Memory Machine.
Lawlor, D., Sher, M. with C. Child (2023), ‘Socio-ecological: Working with Large Complex Collaborative Partnerships (Whole Systems)’, Systems Psychodynamics: Innovative Approaches to Change, Whole Systems and Complexity , D. Lawlor, M. Sher, New York, London.
Trist, E.L. and Bamforth, K.W. (1951). Some social and psychological consequences of the longwall method of coal-getting. Human Relations, 4(1), pp.3–38.
Emery, F.E., Marek, J., Miller, E.J. and Murray, H. (1960). Analysis of automation: a preliminary statement. Tavistock Institute of Human Relations.
Designing effective labour market policy in Europe
Creative Digital Futures Lab
