Co-production techniques
The room as the instrument
The co-production techniques, with how to run each one, what it produces, what it does about power and quiet voices, its cost in time, and how strong the evidence is.
Co-production techniques
The techniques that put the room itself to work. Each one is a way of getting material out of people that a questionnaire would miss, and each entry says what it is, how to run it, what it produces, what it does about power and access, what it costs in time, and how strong the evidence is.
26 techniques in 6 families. Every entry is drawn from the published record, and the evidence badge states what the research found rather than what would be flattering: 13 techniques carry moderate evidence, 12 thin, and 1 have none found.
Showing 26 of 26 techniques.
Building with things
Making a physical object carries the meaning, so the talk follows the model. 3 techniques.
LEGO Serious Play (brick builds) THIN EVIDENCEA facilitated process in which each participant builds a three-dimensional LEGO model in response to a facilitator's question; the models, not the talk, carry the meaning, and each person's answer comes first (the starting point is always an individual's answer, not the group's).
What it is. A facilitated process in which each participant builds a three-dimensional LEGO model in response to a facilitator's question; the models, not the talk, carry the meaning, and each person's answer comes first (the starting point is always an individual's answer, not the group's).
How to run it.
- Four-step Core Process: posing the question; construction (each participant builds a private model); sharing (each participant narrates their model to the group without interruption); reflection (clarification questions; the facilitator draws out surprises and connections).
- Practical shape reported in the one health-research application found: 4-6 participants per group, sessions of 2-3 hours, five sessions across twelve weeks, one facilitator plus a co-facilitator, all sessions audio-recorded and pseudonymised.
- The method's own guidance adds a warm-up build and treats a Core Process plus seven Application Techniques as the 'operating system' a facilitator must follow.
What the room ends up with. Named metaphors and models from every participant; themes extracted from the models and narratives (in the mental health study: 14 models, seven participants, six themes); a set of individual pictures that the group then connects into a shared picture.
Power, quiet voices and access. Strong claims, weak independent testing. The authors of one health study report that LSP 'enabled the removal of power imbalance' and gave voice to 'shy or quiet participants that might have been suppressed by dominant and self-appointed expert participants'; the method's own site claims the bricks are 'a common language that anyone can use, regardless of their education, position or culture'. A comparative study ran LSP alongside Community Philosophy and Design Thinking with people with a range of physical impairments. No study found tests LSP with people with communication disability, aphasia, intellectual disability or low literacy; the builds require hand function and the reflective step requires spoken language.
Cost and time. Reported: 2-3 hour sessions over a 12-week series in the one health study; facilitator skill is decisive ('if the facilitator doesn't know and follow this system the process will not deliver the benefits described'). Inferred: bricks/kits plus facilitator time are the main costs; the question series must be designed and piloted; the sharing step makes workshop size self-limiting.
Evidence. thin, base: health THIN EVIDENCE
Group model building / community-based system dynamics MODERATE EVIDENCEA scripted, facilitated workshop sequence in which stakeholders themselves build causal loop diagrams of the system under study, usually with a modelling team in the room.
What it is. A scripted, facilitated workshop sequence in which stakeholders themselves build causal loop diagrams of the system under study, usually with a modelling team in the room.
How to run it.
- Boundary charts, behaviour-over-time graphs and preliminary models built with group techniques; then model formulation, calibration, testing and simulation in learning labs.
- Applications run across multiple workshops: two half-day sessions with 92 participants plus 23 national stakeholders; a single day-long model-building workshop with 11 participants after separate focus groups with 42; 31 stakeholders across two communities.
What the room ends up with. Causal loop diagrams with named variables, causal links and feedback loops (one published model: 250 variables, 534 connections, 63 feedback loops); candidate leverage points; a formal causal graph usable in simulation or machine-learning development (16 variables, 42 links, 5 subsystems).
Power, quiet voices and access. Community members are routinely included alongside providers and officials. But participation is not automatic: people with personal experience of a stigmatised issue 'are essential model contributors, but are often underrepresented in modeling groups due to multiple barriers', and the remedy reported was extra short, brief, local group sessions.
Cost and time. Reported: 2-3 workshops to a workshop series; the published models required a modelling team plus analysis between workshops. Inferred: the highest-cost technique in this lane for skilled facilitation and analysis; transcribing and coding every session doubles the burden.
Evidence. moderate, base: health MODERATE EVIDENCE
Object and prop elicitation / tabletop mapping with objects THIN EVIDENCEUsing physical objects in the room as the elicitation device, instead of cards or questions alone.
What it is. Using physical objects in the room as the elicitation device, instead of cards or questions alone.
How to run it.
- Place ordinary household objects and ask participants to talk through them (published example: the refrigerator as the object of an object-elicitation interview plus photograph).
- In River of Life, drawing can be replaced by locally sourced objects (leaves, beans, stones) laid along the river on a flipchart on the floor, and the guide explicitly allows this 'if drawing is challenging'.
What the room ends up with. Talk anchored in a concrete artefact rather than an abstraction; entry into subjects that self-report leaves out; a physical layout on the table or floor that the group can then annotate.
Power, quiet voices and access. The design logic is accessibility: objects substitute for writing and drawing. No study found measures outcomes for participants with low literacy or cognitive impairment, and objects can themselves require physical handling.
Cost and time. Inferred: cheapest technique in the lane (ordinary or locally sourced materials); cost sits in facilitator preparation and transport of materials.
Evidence. thin, base: health THIN EVIDENCE
Drawing, images and the body
The material comes out as a picture, an image or a map of the body. 6 techniques.
Photovoice MODERATE EVIDENCEParticipants photograph their own lives or community, then discuss and analyse the images together; the defining goals are to record and reflect community strengths and concerns, to promote critical dialogue in group discussion of the photographs, and to reach policymakers.
What it is. Participants photograph their own lives or community, then discuss and analyse the images together; the defining goals are to record and reflect community strengths and concerns, to promote critical dialogue in group discussion of the photographs, and to reach policymakers.
How to run it.
- Photo assignment; camera or phone training; photography period; group discussion and participatory analysis; exhibition or dissemination.
- In practice the components vary widely, with combinations of introductory, training, ethics, group, individual and exhibit sessions; a review of 22 studies found a mean of 18 participants (range 4-38).
- One health study ran a 2-week photo-taking period followed by a photo-interview, with interviews in participants' preferred language.
What the room ends up with. Participant-authored images with captions and themes; exhibits; priorities expressed 'through their own visual perspectives'; illness accounts the interview alone did not surface.
Power, quiet voices and access. The clearest inclusion case in this lane for limited health literacy: the Malaysian asthma study deliberately sampled adults with limited health literacy and ran interviews in their preferred language. Demands: a working camera or phone, two or more weeks of elapsed time, and enough mobility and confidence to photograph. Written captions add a literacy demand unless dictation is used. Reviews report that the researcher-participant relationship is unevenly described and that ethical reporting is patchy.
Cost and time. Reported: a two-week photography window before the photo-interview; a mean of 18 participants per study across a review. Inferred: equipment, ethical safeguards around photographing others, printing/exhibition costs, and two to three contact points per participant.
Evidence. moderate, base: health MODERATE EVIDENCE
Drawing and mark making (draw and write; draw, write and tell) THIN EVIDENCEParticipants draw in response to a prompt and then explain the drawing in their own words.
What it is. Participants draw in response to a prompt and then explain the drawing in their own words.
How to run it.
- Invitation to participate through open-ended questions, drawing, then writing or telling; administrable to a whole group or class at once; the published advice is to build trust first and to treat the child's own interpretation as the data.
What the room ends up with. Visual data plus the participant's own verbal account of it; children's perspectives that adult-led methods miss.
Power, quiet voices and access. Works with children from about age 4-6 upward, including multi-ethnic and deprived-area samples, without a reading requirement where scribing or dictation is allowed. Cautions: caregiver presence influences what is produced; the method needs an 'enabling climate' and is easily over-read by adult analysts.
Cost and time. Inferred: very low materials cost (paper, pens, a quiet space); 20-40 minutes per group, plus analysis of drawings and talk together.
Evidence. thin, base: health THIN EVIDENCE
Collage THIN EVIDENCEParticipants assemble images and materials into a composition as a data-elicitation task.
What it is. Participants assemble images and materials into a composition as a data-elicitation task.
How to run it.
- Interview first, then a collage task in which the participant represents their experience visually; both the talk and the artwork are coded and compared.
- It works alongside other methods in the same session: one mHealth co-design study ran collage and card sorting in its first design session.
What the room ends up with. Three distinct kinds of material: synthesis and confirmation of what was said, focus on salient aspects, and previously unarticulated experiences.
Power, quiet voices and access. No reading or writing required; the evidence found does not report adaptations for motor impairment or cognitive disability. Cutting and gluing imposes a physical demand that drawing-free variants would not.
Cost and time. Inferred: low (magazines, glue, scissors, card); one session of 30-60 minutes; analysis must handle image and text together.
Evidence. thin, base: health THIN EVIDENCE
Body mapping MODERATE EVIDENCEParticipants create and annotate a body outline, using the body as the organising frame for experience.
What it is. Participants create and annotate a body outline, using the body as the organising frame for experience.
How to run it.
- Workshops in which participants make full-body, mini- or legend maps, then narrate them; the published projects run multi-day workshops (3 days with 24 young people), and the body maps are analysed alongside written narratives and the transcripts of the workshops.
What the room ends up with. Body map visuals plus written narratives; themes about how social context is experienced in and through the body; material for priorities (in the review: body mapping used in 17 studies with 3-48 participants).
Power, quiet voices and access. Reported benefits include reaching 'difficult-to-access emotions and experiences', a strengths and resilience focus, therapeutic effect, and the engagement of children and young people; some studies recruited only women or only children and young people. Requires a private, safe space (the body is the topic) and physical capacity to draw or trace at scale.
Cost and time. Reported: multi-day workshops in the Adolescent X project (3 days, n=24); participant numbers 3-48 across a review. Inferred: high facilitation and room cost per participant; session length must cover both making and narrating.
Evidence. moderate, base: health MODERATE EVIDENCE
River of Life THIN EVIDENCEEach participant draws their experience as a river: calm water, fish and flowers for good periods, rocks, rapids and crocodiles for difficulties, tributaries for people and events.
What it is. Each participant draws their experience as a river: calm water, fish and flowers for good periods, rocks, rapids and crocodiles for difficulties, tributaries for people and events.
How to run it.
- Part A: explain the metaphor, set the time period, give each person flipchart paper and markers (and natural materials if drawing is hard), provide the prompt question, ask for symbols.
- Part B: each participant presents their river without interruption; the group then identifies common influences, enablers and challenges; a further step ranks these and agrees actions.
- Group size should be no more than eight to allow equal participation; the TGHN guide states up to eight for 'equal participation and a welcoming and safe environment'.
What the room ends up with. Individual visual timelines; a group list of shared enablers, barriers and turning points; ranked issues if the optional ranking step is used. One documented use fed a PRA matrix ranking exercise to identify the most pressing issues.
Power, quiet voices and access. The clearest literacy statement in the lane: 'The method is accessible for all literacy levels'. Adaptations named: allow a road or journey metaphor instead of a river; draw on the ground outdoors; use leaves or beans instead of drawing; separate groups by gender, ethnicity or religion in some contexts. Requires safeguarding and signposting because personal and potentially distressing material is shared. No peer-reviewed health-services evaluation was found.
Cost and time. Inferred from the guides: drawing time plus several minutes of presentation per person; the max group size of eight means low throughput per facilitator. Inferred: low materials cost; the real cost is facilitation time and emotional support.
Evidence. thin, base: grey THIN EVIDENCE
Photo and film elicitation (image elicitation, catalyst films) THIN EVIDENCEImages or filmed narratives used as the prompt for discussion rather than as the data themselves.
What it is. Images or filmed narratives used as the prompt for discussion rather than as the data themselves.
How to run it.
- Participants bring or take photographs and the interview is conducted around them; in experience-based co-design, interviews with patients and staff are edited into a film which is then shown at the joint event to anchor the discussion of priorities.
What the room ends up with. Empathy and a shared reference point; in EBCD, a common account of the journey that participants then map onto.
Power, quiet voices and access. The image carries part of the meaning, but the discussion is still spoken, so language and hearing demands remain. Producing the film requires additional participants willing to be filmed and a production step.
Cost and time. Inferred: film production is the expensive variant (filming, consent, editing); photo elicitation is cheap by comparison, needing only participants' own images.
Evidence. thin, base: health THIN EVIDENCE
Cards, sorting and ranking
Participants move items around, so the judgement is visible and fast. 5 techniques.
Card sort MODERATE EVIDENCEParticipants group content written on cards: either into categories they create themselves (open sort) or into categories the researcher has predefined (closed sort), and explain their reasoning.
What it is. Participants group content written on cards: either into categories they create themselves (open sort) or into categories the researcher has predefined (closed sort), and explain their reasoning.
How to run it.
- Open sort: give the cards, ask participants to group them into whatever categories make sense to them, then label the categories and explain the grouping.
- Closed sort: give the cards and the predefined categories, then ask why each card went where it did.
- Practical shape in health studies: 36 argument cards from which each participant selects fewer than ten and ranks them by importance, with the selected cards then placed anonymously on the table; suitable for groups of two to three as well as individuals.
- Reported time: 1-2 hours to set up, 15-30 minutes per participant;
- A closed sort has also been used to validate a framework with 12 researchers, though that source is not part of this ledger and is not cited as evidence.
What the room ends up with. Grouped and labelled content; a quantified sort matrix that can be analysed (inverted factor analysis, cluster analysis, multidimensional scaling); a written explanation of the categories that can be quoted.
Power, quiet voices and access. A low-literacy-compatible variant exists in the literature but was not retrieved in this run, so it is not cited (see gaps). Reading is nevertheless required for word-based cards unless pictorial cards are used; a Tanzanian study deliberately excluded women with fewer than seven years of primary education from the sorting task and interviewed them instead, which is direct evidence that card sorts carry a literacy entry requirement as usually run.
Cost and time. Reported: 1-2 hours preparation plus 15-30 minutes per participant for a card sort session. Inferred: cheap to produce and to run; the cost is duplicated per participant if sorts are individual, and analysis time if the sort is to be treated as data.
Evidence. moderate, base: health MODERATE EVIDENCE
Q sort (Q methodology) MODERATE EVIDENCEParticipants rank-order a set of statements drawn from the 'concourse' of opinions on a topic; the resulting Q sorts are correlated and factor-analysed by person, not by variable, to identify shared viewpoints.
What it is. Participants rank-order a set of statements drawn from the 'concourse' of opinions on a topic; the resulting Q sorts are correlated and factor-analysed by person, not by variable, to identify shared viewpoints.
How to run it.
- Build the concourse, derive a representative statement set, sample participants (commonly 30-40), have each participant sort statements onto a distribution grid (typically forced normal), then run by-person factor analysis and interpret factors against the highest- and lowest-ranked statements.
- Health-services use in prioritisation has combined a Q sort with a focus group.
What the room ends up with. A small number of named viewpoints or 'types', each with a distinguishing statement ranking; a defensible representation of the range of views in a population rather than the majority view.
Power, quiet voices and access. The task is physical sorting of printed statements, so reading ability is required and so is the patience for a long, forced-choice task. Q methodology has been applied in disability research, where four factors reproduced the social, medical and disability-rights models plus a family perspective. No adaptation for low literacy or communication disability was found.
Cost and time. Inferred: statement development and piloting is the largest cost; the sort itself is typically 30-60 minutes per participant and is often done individually, so it scales linearly in researcher time; factor analysis needs software and expertise.
Evidence. moderate, base: health MODERATE EVIDENCE
Ranking (importance and difficulty/feasibility) MODERATE EVIDENCEParticipants order items on a named dimension, most often importance, but also difficulty, feasibility or priority.
What it is. Participants order items on a named dimension, most often importance, but also difficulty, feasibility or priority.
How to run it.
- Items are generated first (from elicitation), then each participant or group places them in order; scoring can be aggregated to an importance score and used to prioritise.
- Ranking is frequently the final step inside another technique: after NGT idea generation, after a World Cafe conversation, A virtual World Cafe study in this corpus ranked implementation strategies by priority against feasibility and used the two axes to sort them into act-first, act-if-possible, consider-later and discard, which is the clearest documented example of the two-dimension ranking ([no source retained in this ledger]; see gaps).
What the room ends up with. An ordered list with scores that can be reported numerically (for example a 0-1 importance score per outcome); a defensible prioritisation that a systems-map workshop can adopt as its focus.
Power, quiet voices and access. Ranking is cognitively lighter than writing but still requires participants to hold many items in mind; with long lists, pile sorting into groups first is a documented alternative. Numeric scales and grids presuppose numeracy unless the ranking is done physically.
Cost and time. Reported: ranking took place inside a single focus-group session in an NGT study. Inferred: quick and cheap once items exist; the cost is in generating and cleaning the item set.
Evidence. moderate, base: health MODERATE EVIDENCE
Pile sorting / free listing (cultural domain analysis) MODERATE EVIDENCEFree listing produces the items in participants' own vocabulary; pile sorting asks participants to group those items into piles that go together, which are then analysed for dimensions.
What it is. Free listing produces the items in participants' own vocabulary; pile sorting asks participants to group those items into piles that go together, which are then analysed for dimensions.
How to run it.
- Free list first (participants name everything they can think of on the topic), then pile sort (each participant arranges the items into groups of their own making), then multidimensional scaling and cluster analysis of the pooled sorts, then participant review and interpretation of the resulting dimensions.
- Application in a veteran mental-health study: free lists produced barriers, pile sorting categorised them, scaling and clustering produced five dimensions, and participants then reviewed and named the dimensions.
What the room ends up with. A culturally grounded item list and a validated set of dimensions with participant-given names, which is a stronger basis for a map than a researcher-imposed category system.
Power, quiet voices and access. Free listing requires only speech; pile sorting requires physical handling of cards or slips and enough items to be manageable. The analysis is quantitative and needs software.
Cost and time. Inferred: two contact points (list, then sort); analysis is more technical than a simple card sort, and the participant interpretation step adds a third contact.
Evidence. moderate, base: health MODERATE EVIDENCE
Card sort as data versus card sort as decoration THIN EVIDENCEThe same physical activity can be run either as a measurement (sorts recorded, pooled and analysed) or as an ice-breaker that generates talk which is then read impressionistically.
What it is. The same physical activity can be run either as a measurement (sorts recorded, pooled and analysed) or as an ice-breaker that generates talk which is then read impressionistically.
How to run it.
- As data: fix the card set in advance; record each participant's sort; analyse (inverted factor analysis; multidimensional scaling and cluster; importance scoring after ranking); report the analysis.
- As decoration: sort cards to prompt discussion, then report only the discussion.
What the room ends up with. As data: a sort matrix and factor/cluster structure that can be reanalysed by others. As decoration: themes from the conversation, with the sort itself leaving no analysable trace.
Power, quiet voices and access. The distinction matters for inclusion, because the 'data' version raises the literacy and numeracy bar (recorded sorts, grids, scoring) while the 'decoration' version is easier but produces weaker evidence.
Cost and time. Inferred: the data version costs analysis time and a fixed instrument; the decoration version costs facilitation only.
Evidence. thin, base: design THIN EVIDENCE
Structured talk
The turn-taking is designed, which stops the loudest voice setting the agenda. 5 techniques.
Nominal group technique (NGT) MODERATE EVIDENCEA structured small-group consensus method with four stages: silent generation of ideas, round-robin sharing, clarification, and voting by ranking.
What it is. A structured small-group consensus method with four stages: silent generation of ideas, round-robin sharing, clarification, and voting by ranking.
How to run it.
- Classic four stages; variations reported across 57 studies, with between 1 and 41 groups and 2-30 participants per group; 30 identifiable design decision points.
- In use: 14 groups of 123 living kidney donors identified and ranked 35 outcomes; three meetings over four months with a five-GP panel produced agreement on the content of electronic records for child health examinations [no source retained; see gaps].
What the room ends up with. A ranked list with aggregate importance scores; a documented priority set with the reasons members gave.
Power, quiet voices and access. The silent generation stage means every participant produces input before discussion, which is the method's stated defence against dominance, and the private ranking produces a group score rather than a floor vote. A virtual variant has been documented for participants in different locations and time zones, reported as producing 'more effective engagement and focus with fewer distractions compared to in-person meetings', though the same paper is a reflection on practice rather than a comparison. Written idea generation can be replaced by spoken round-robin, which lowers the writing demand.
Cost and time. Reported: the scoping review's range of 1-41 nominal groups and 2-30 participants per group is the only size evidence carried here; meeting lengths of 2.5-5 hours are reported in NGT studies but the specific paper is not retained in this ledger. Inferred: moderate cost; the staged process needs a trained facilitator and someone to aggregate rankings.
Evidence. moderate, base: health MODERATE EVIDENCE
Delphi (including e-Delphi and modified Delphi) MODERATE EVIDENCEAn iterative multistage questionnaire process with controlled feedback between rounds, designed to transform opinion into group consensus without the participants meeting.
What it is. An iterative multistage questionnaire process with controlled feedback between rounds, designed to transform opinion into group consensus without the participants meeting.
How to run it.
- Questionnaire rounds with individual feedback of group responses between rounds; 'modified Delphi' variants add a meeting or a first round derived from literature.
- Remote variants run entirely online: three rounds of online engagement with 12 experts; other studies combine e-Delphi consultations with real-time online workshops and run twin panels of researchers and knowledge users, but those sources are not retained here.
What the room ends up with. A consensus statement, definition, competency set or guideline with reported agreement levels, and a record of items that did not reach consensus.
Power, quiet voices and access. Because it is written and asynchronous it can include participants who cannot travel, but it demands reading and writing, and often several rounds of it. Expert panels are the standard unit, which raises the literacy and status bar relative to a community workshop.
Cost and time. Reported: two to five rounds, with the median study running several weeks. Inferred: cost is dominated by questionnaire development and chasing non-response; the absence of a room is the saving.
Evidence. moderate, base: health MODERATE EVIDENCE
World Cafe MODERATE EVIDENCEA cafe-format conversation method: small tables of four or five, progressive rounds of about 20 minutes, one table host who stays while everyone else moves, and a final 'harvest' of collective insights.
What it is. A cafe-format conversation method: small tables of four or five, progressive rounds of about 20 minutes, one table host who stays while everyone else moves, and a final 'harvest' of collective insights.
How to run it.
- Seven design principles: set the context; create hospitable space; explore questions that matter; encourage everyone's contribution; connect diverse perspectives; listen together for patterns; share collective discoveries.
- Practical format: four to five per table, at least three rounds of about 20 minutes, table host stays and travellers move, participants write and draw on the tablecloths, then a whole-group harvest.
- Health-services applications have adapted it for research prioritisation with marginalised communities, for a culturally and linguistically diverse community, and across rural, urban-multicultural and indigenous communities.
What the room ends up with. Annotated tablecloths, clustered ideas and a harvest of themes; in prioritisation work, a ranked set of research priorities.
Power, quiet voices and access. The format lowers the entry bar relative to a plenary discussion because contributions can be spoken or written directly onto the paper, and the source guidance explicitly allows people to participate by listening. The CALD application shows adaptation for a language-minority community. Small tables of four to five are argued to avoid the parallel-conversation and silent-participant problem of larger groups. Practical limits: the method is 'complex and demanding when collecting high-quality data' and the hosts' behaviour substantially determines data quality; hosts need a structured instruction, not the abstract principles alone.
Cost and time. Reported: three rounds of about 20 minutes plus a 20-30 minute harvest. Inferred: room set-up (staggered cafe tables, tablecloths, pens, refreshments) plus at least one host per table; large group size is handled by adding tables rather than lengthening rounds.
Evidence. moderate, base: health MODERATE EVIDENCE
Marketplace and carousel stations THIN EVIDENCEA set of stations, each carrying its own question or task, through which sub-groups rotate, adding to what earlier groups left behind.
What it is. A set of stations, each carrying its own question or task, through which sub-groups rotate, adding to what earlier groups left behind.
How to run it.
- Set up one station per aspect of the task; divide the group into teams, each starting at a different station; teams generate ideas onto the flipchart or sticky notes; after a set time they rotate, read what the previous team wrote and build on it; continue until all teams have visited all stations; reconvene to consolidate, identify overlaps, then prioritise and finalise.
- A shorter variant uses 5-10 charts, 2-3 minutes per station, and a closing 5-10 minute review.
- The same source gives the virtual form: breakout rooms for the rotating topics and a shared whiteboard per room.
What the room ends up with. A dense, layered set of ideas per topic, already grouped by topic station, that can be consolidated into themes and priorities.
Power, quiet voices and access. Movement between stations is the mechanism, so the method works poorly for participants with limited mobility unless the papers are brought to them or the rotating content is digital. Station prompts can be pictorial. No health-services evaluation found.
Cost and time. Inferred: low materials (flipcharts, markers, sticky notes); costs scale with the number of stations and the facilitation needed to keep rotation on time.
Evidence. thin, base: grey THIN EVIDENCE
Fishbowl THIN EVIDENCEAn inner circle discusses while an outer circle observes; in the open form, observers can take an empty seat in the inner circle.
What it is. An inner circle discusses while an outer circle observes; in the open form, observers can take an empty seat in the inner circle.
How to run it.
- Two concentric circles; the inner group discusses and the outer group observes and takes notes; observers then ask clarifying questions or swap in.
- In the medical-education use reported, students interview a simulated patient in the inner circle while peers and a tutor observe, and the study produced guidelines for running the tutorial because 'there is little evidence to guide teachers on how best to structure a fishbowl tutorial'.
- Variants keep one or two seats empty so observers can join briefly and then return.
What the room ends up with. A visible discussion that a larger group can watch and then interrogate; observation notes that can be harvested as data; in the simulated-patient form, observed consultation practice.
Power, quiet voices and access. The observer role is a legitimate, low-demand way to take part, and observation notes are a route into the data for people who do not want to speak. The structure also lets a small number of people command the discussion, so the outer circle needs a real route to contribute (clarifying questions, swap-in seats, a facilitated debrief).
Cost and time. Inferred: cheap (chairs and a room); the cost is in briefing observers and running the debrief, and a simulated patient adds a paid role.
Evidence. thin, base: health THIN EVIDENCE
Enactment, walking and shadowing
People act the service out, or walk it, rather than describing it. 4 techniques.
Forum theatre (Theatre of the Oppressed) MODERATE EVIDENCEA participatory, arts-based method rooted in Boal's Theatre of the Oppressed, in which a scene of oppression is performed and then re-run with audience members replacing the protagonist to test alternative actions.
What it is. A participatory, arts-based method rooted in Boal's Theatre of the Oppressed, in which a scene of oppression is performed and then re-run with audience members replacing the protagonist to test alternative actions.
How to run it.
- Participatory workshops co-develop a script from lived experience; public performances follow; the audience chooses the scene it identifies with most and the play is 'forumed', with spectators invited to replace the struggling person and demonstrate how they would handle the situation.
- One pilot co-produced the work with people with lived experience of psychosis, carers, performers and Theatre of the Oppressed experts, and evaluated feasibility and acceptability.
- A public-engagement use gathered story workshops first, then performed ten plays to 1,175 community members, with focus groups afterwards.
What the room ends up with. A co-created script and performance; audience-level knowledge change and preference data; in the staff-learning use, changed individual and collective understanding of abuse.
Power, quiet voices and access. It does not require literacy: scenes are performed and interventions are enacted. It is strongly language-dependent (script and dialogue) and demands a room, performers and rehearsal time, and it can expose participants to distressing material. Reported limitations include participants' ambivalence before taking part and difficulty moving from insight to practice change.
Cost and time. Reported: ten performances plus five preparatory workshops in one project; a three-day workshop culminating in five short plays in another. Inferred: the highest production cost of any technique in this lane (rehearsal, venue, performers), partly offset by the size of the audience reached.
Evidence. moderate, base: health MODERATE EVIDENCE
Role play, re-enactment and simulated patient journeys THIN EVIDENCEParticipants act out an interaction or walk through a service journey in role, often with a simulated patient or a stand-in service user.
What it is. Participants act out an interaction or walk through a service journey in role, often with a simulated patient or a stand-in service user.
How to run it.
- Used inside forum theatre (audience members step into the scene) and inside a fishbowl tutorial, where students interview a simulated patient while observed; journeys are re-enacted by mapping stages and touchpoints (including emotional touchpoints) rather than by acting.
What the room ends up with. Observable interaction practice and observer notes; a staged journey with emotional high and low points that a service can redesign against.
Power, quiet voices and access. Enactment can bypass literacy, but it requires physical presence, mobility and a willingness to perform, and can be uncomfortable for people who do not want to be watched. Journey mapping in the EBCD lineage requires spoken interviews to build the journey in the first place.
Cost and time. Inferred: moderate; a simulated patient is a paid role; journey mapping needs interviews plus a workshop, so it duplicates contact time.
Evidence. thin, base: health THIN EVIDENCE
Walking interviews and go-alongs MODERATE EVIDENCEInterviewing participants while moving with them through the environment being discussed, combining interview with participant observation.
What it is. Interviewing participants while moving with them through the environment being discussed, combining interview with participant observation.
How to run it.
- Accompany the participant on a route that matters to them and talk as you go; audio-record and supplement with field notes and photographs.
- Reported uses include refugee maternal health and the built environment of dementia care homes.
What the room ends up with. In-situ accounts and observations: not only what participants say, but where and how they move; identification of barriers that are hard to articulate sitting down; interaction between participant and surroundings.
Power, quiet voices and access. The strongest EAL/access case among the embodied methods: it can 'assist in identifying the needs and challenges of people with limited language skills or low educational levels', and has been used with people with visible and invisible disabilities. Exclusion is built into some designs: the dementia walking-interview study required 'the ability to express themselves verbally in German', which excludes non-German speakers by design. Physical mobility, weather, and being on unfamiliar ground all matter; recording quality outdoors is poor, and talking while walking is difficult for people who use walking aids.
Cost and time. Inferred: one interview-length contact per participant plus travel; transcription is harder (noise, movement) and analysis involves field notes and photographs as well as talk.
Evidence. moderate, base: health MODERATE EVIDENCE
Shadowing THIN EVIDENCEA trained observer follows a patient (or a member of staff) through a care episode and records what actually happens.
What it is. A trained observer follows a patient (or a member of staff) through a care episode and records what actually happens.
How to run it.
- Follow the person through the pathway with a structured record; interrogate what you observed; feed findings back into a service improvement cycle.
- Published work derives ten principles for shadowing as a quality improvement method from interviews with 20 participants of a national improvement programme.
What the room ends up with. A first-hand observed account of the pathway, including waits and transitions that participants do not report; an evidence base for improvement priorities.
Power, quiet voices and access. Shadowing requires consent at a vulnerable moment and does not give the patient a speaking role, so it is a method about people rather than with them unless combined with interview. It is physically demanding for the observer and ethically demanding in wards and homes.
Cost and time. Inferred: high observer time per person observed (a full episode each); requires staff cooperation, governance approval and a structured record.
Evidence. thin, base: health THIN EVIDENCE
Online and distributed
The room is a screen, or the contribution happens over days. 3 techniques.
Online whiteboards and virtual mapping THIN EVIDENCEA shared digital canvas on which participants place, draw and connect content during a remote or hybrid workshop.
What it is. A shared digital canvas on which participants place, draw and connect content during a remote or hybrid workshop.
How to run it.
- The carousel guidance gives the minimal version: breakout rooms for each rotating topic, with a shared whiteboard per room for notes.
- The richer version is map-based planning: an online geographic platform on which participants locate and negotiate features together, replacing physical presence with 'virtual mobility'.
What the room ends up with. A digital artefact (board, map or diagram) that persists and can be screenshotted into a report; in virtual mapping, a shared spatial plan produced under lockdown conditions.
Power, quiet voices and access. Can include people who cannot travel or who live far from the venue. Excludes people without reliable connectivity, a suitable device, or the digital confidence to manipulate a canvas; also excludes participants whose contributions come from physical movement and face-to-face handover, which have to be abandoned in the virtual form. A pictorial, nonverbal variant has been documented for children with developmental disabilities, using a shared presentation and the annotate function.
Cost and time. Inferred: low direct cost (licensed software or free tiers) but high preparation cost (boards built in advance), plus a second facilitator to work the technology.
Evidence. thin, base: design THIN EVIDENCE
Live polling NO EVIDENCE FOUNDInstantaneous voting or rating by participants, displayed back to the room.
What it is. Instantaneous voting or rating by participants, displayed back to the room.
How to run it.
- No source found in this search describes a live-polling procedure in a health-services co-production workshop, nor any evaluation of it.
- Nothing is reported here rather than an estimate.
What the room ends up with. Not reported in the sources found.
Power, quiet voices and access. Not determinable from the sources found; the plausible exclusions (devices, connectivity, literacy for typed answers, numeracy for scales) are inference, not evidence.
Cost and time. Not reported in the sources found.
Evidence. none_found NO EVIDENCE FOUND
Asynchronous contribution (online forums and threaded discussion) MODERATE EVIDENCEParticipants contribute over days rather than in real time, usually through a moderated threaded forum.
What it is. Participants contribute over days rather than in real time, usually through a moderated threaded forum.
How to run it.
- Moderators post prompts on a schedule and participants respond when they can: eight groups of transgender and gender-diverse young adults, each running four consecutive days with two prompts per day; seven groups and 100 participants across the ALS Talk project, with weekly questions on a secure threaded board; four asynchronous groups on sensitive sexual-health topics with 29 participants.
What the room ends up with. Threaded textual data with a traceable record of who said what and when; branching interaction between participants rather than turn-taking; high reported acceptability (98% rated the experience excellent, very good or good; 95% would sign up again).
Power, quiet voices and access. Reduces travel and disability-related barriers and offers anonymity for stigmatised topics. The price is a written medium: it demands reading and typing, and sustained participation across several days, and it loses the nonverbal and relational texture of a room.
Cost and time. Reported: four consecutive days to several weeks of participation, with a moderator posting prompts. Inferred: platform licensing or setup, moderator time spread over the whole window, and slower recruitment-to-analysis cycle than a single workshop.
Evidence. moderate, base: health MODERATE EVIDENCE