01 · Shared data
Context for reuse
What information helps a new reader use a dataset appropriately?
Read the briefProjects & questions
Small starting points.
Room for discovery.
Exploratory briefs that connect a real research challenge with a manageable first test.
Exploratory portfolio
Three questions worth working through
These are working briefs for learning and discussion. Each sets out a possible investigation, a small first test, and a decision it could inform. They do not report completed experiments or measured outcomes.
02 · Research translation
A finding into practice
What gets lost when a research finding becomes a practical instruction?
Read the brief03 · Evaluation
Learning from a pilot
Which observations would justify a larger test, a revision, or a pause?
Read the briefWorking brief 01 · Shared data
Make a dataset understandable beyond its first use.
A dataset may make sense to the people who assembled it while leaving a new reader unsure about units, exclusions, or missing values. This brief explores whether a short context sheet can reduce that ambiguity.
Question and first test
Take a small synthetic dataset and write a context sheet covering its purpose, field definitions, collection assumptions, and known limitations. Ask a reader unfamiliar with the sample to describe an appropriate use and one inappropriate use. Record the questions they still need answered.
Evidence and next decision
Compare the reader’s interpretation with the intended meaning of the fields. Look for recurring misunderstandings rather than a single overall score. Revise the context sheet before adding more data or building a repository.
Boundary: a clearer description does not establish data accuracy or permission to reuse real records. Those questions require separate assessment.
Use the handover checklist →Working brief 02 · Research translation
Keep the conditions attached to a finding.
Practical summaries need to be short, but compression can remove the conditions that make a conclusion meaningful. This brief examines the passage from a research explanation to a decision aid.
Question and first test
Write a short fictional research summary with a clearly stated setting, comparison, and limitation. Produce a one-page decision aid from it. Ask another reader which actions the aid supports and which claims it does not support.
Evidence and next decision
Trace every recommendation back to a statement in the original summary. Where an instruction depends on an unstated assumption, revise it or make that assumption explicit. Decide whether the aid is ready for a second reading before testing its use in practice.
Boundary: reader agreement is a check on communication. It does not validate the underlying research or demonstrate real-world effectiveness.
Read the evidence-to-action note →Working brief 03 · Evaluation
Design a pilot around a decision.
A small pilot can reveal practical obstacles before a larger commitment. The useful question is what it needs to teach, not simply whether an activity can be completed.
Question and first test
Choose a routine, low-risk process such as finding a document in a shared folder. Define the current process and a proposed alternative. Observe a few attempts, record the points of confusion, and keep the task and instructions consistent.
Evidence and next decision
Separate feasibility observations from claims about effectiveness. A process that works once may still depend on unusual support. Agree in advance which obstacles call for revision and what additional evidence would be needed for a wider trial.
Boundary: a small convenience sample cannot establish that the alternative will work for a wider population.
Use the pilot clinic agenda →Turn a brief into a focused discussion.
Use a workshop agenda to surface assumptions, compare methods, and leave with a testable question.