← Evaluation toolkit · Sparkworks Studios

Evaluation explained

The words funders and evaluators use, in plain English. Use it to read a brief, write one, or decide what kind of evaluation you need.

The three words

Monitoring
Routinely counting and recording what happens: who came, how often, where from. It tells you what is happening, not why or whether it mattered.
Evaluation
Stepping back to judge how well something worked, for whom and why, using the monitoring data and other evidence.
Learning
Using what you find to change what you do. Often written together as “monitoring, evaluation and learning”, or MEL.

Types of evaluation: what it is for

Formative evaluation
Happens while a programme is running, to improve it as it goes. Think of a cook tasting the soup.
Summative evaluation
Happens at or near the end, to judge what was achieved. The guests tasting the soup.
Process evaluation
Looks at how the programme was delivered: what happened, who it reached, what helped and what got in the way.
Impact (or outcome) evaluation
Looks at what changed for people, and how far the programme made the difference.
Economic (value-for-money) evaluation
Sets what the programme cost against what it achieved.
Developmental evaluation
The evaluator works alongside a new or fast-changing programme, feeding evidence in as it takes shape.

Types of evaluation: who does it

Independent (external) evaluation
Someone outside the delivery team designs the evaluation and reaches the conclusions.
Self-evaluation
Your own team collects and reviews the evidence. “Supported” self-evaluation adds an evaluator who sets it up, trains the team and checks the work.
Participatory evaluation
The people taking part help decide the questions, gather the evidence and make sense of it.
Critical friend
An experienced outsider who advises and challenges, without running the evaluation.

How conclusions are reached

Theory-based evaluation
Tests whether change happened in the way your theory of change said it would, step by step. Contribution analysis and realist evaluation are examples.
Experimental evaluation
Compares people who took part with a similar group who didn’t, chosen at random (a randomised controlled trial). Rare in arts and community work, and often not practical or fair.
Quasi-experimental evaluation
Compares participants with a similar group that wasn’t chosen at random, such as people on a waiting list.
Before-and-after
Compares the same people at the start and the end. Simple, but other things in their lives may explain the change.
Case-based evaluation
Studies a small number of people, groups or places in depth.
Mixed methods
Uses numbers and people’s own accounts together, so each checks the other.

Creative and participatory techniques

Evaluation café
An informal session over tea and cake. People talk in small groups around tables, each with a question, and move between them.
Learning walk
The evaluator walks through a space, session or neighbourhood with participants or staff, talking about what they notice and what has changed.
Audio diary
Participants take a simple voice recorder away and record their thoughts in their own time, across the life of a project.
Creative methods
Using drawing, song, photography, radio or games to gather evidence, especially where words or forms are a barrier.
Co-creation
The people the work is for share in designing it. In evaluation: they help choose the questions, gather the evidence and interpret it.
Sense-making session
A meeting where participants and partners look at the draft findings together and say what they think they mean, before the report is written.
Photo voice
Participants take photos that show what matters or what has changed, then talk about them.
River of life
People draw their time on a project as a river, with rapids and calm stretches, and talk it through.
Body mapping
People add to an outline of a body what they think, feel and can now do.
The evaluation song
A group writes a song about what has changed. The lyrics are the evidence. A Sparkworks original.
The radio show
Participants plan and record a short show reviewing their project. A Sparkworks original.
Findings quiz night
The findings are turned into a quiz for staff, volunteers or partners. Teams guess the answers before the real results are revealed.

The building blocks

Theory of change
A description of how your work is expected to lead to change: what you do, what changes along the way, why, and what has to be true.
Logic model
A simpler table of inputs, activities, outputs and outcomes. It shows the steps but not the reasons.
Evaluation framework
The plan: your questions, what evidence you’ll collect for each outcome, who collects it and when.
Inputs
What goes in: money, staff time, venues, materials.
Outputs
What you deliver, in numbers: sessions run, people attending.
Outcomes
The changes for people: what they feel, know, can do or do differently.
Impact
The longer-term, wider change your work contributes to.
Indicator
The sign you look for to tell whether an outcome is happening.
Baseline
The starting point you measure change against.

Words about evidence

Quantitative
Evidence in numbers: counts, scores, percentages.
Qualitative
Evidence in words, images and observation: what people say and do.
Attribution
Claiming the programme caused the change. Hard to prove outside an experiment.
Contribution
Showing the programme helped bring about the change, alongside other things. Usually the more honest claim.
Counterfactual
What would have happened without the programme.
Triangulation
Checking a finding against more than one source or method.
Validated measure
A set of questions that has been tested to show it measures what it claims to, such as a recognised wellbeing scale.
Sample
The people you collect evidence from, when you can’t ask everyone.
Informed consent
People agree to take part knowing what is collected, why, and how it will be used, and can change their mind.
Proportionate
The evaluation is in scale with the programme. A small project doesn’t need a large evaluation.

Techniques

The approaches and methods themselves, such as contribution analysis, Most Significant Change, interviews and wellbeing scales, are explained in the methodology menu, where you can also choose the ones you want.

Definitions here are written for clarity, in line with the Magenta Book (HM Treasury and the Evaluation Task Force, updated 2026) and BetterEvaluation’s glossary. The formative and summative distinction comes from Scriven, M. (1967) ‘The methodology of evaluation’; the soup comparison is credited to Robert Stake.

Have an evaluation brief? Let’s talk.

I’m Catherine Burgis at Sparkworks Studios. Send me your brief, or the outline of one, and I’ll tell you how I’d approach it and what it would take.

Email Catherine about your evaluation

hello@sparkworksstudios.co.uk

www.sparkworksstudios.co.uk