Research Idea / Blueprint
Defines the question, rationale, possible method, risks, and what would need to be learned. It is planning evidence, not outcome evidence.
Methodology / Instrument Draft
Specifies how data might be collected or analysed and how an instrument might work. A well-written method does not prove that data were collected, ethics approval occurred, analysis happened, or validity was established.
Pilot
Tests feasibility, comprehension, workflow, safety, data quality, or early signals in a bounded context. A pilot should produce learning and corrections. It should not be presented as broad effectiveness or national evidence.
Case study
Documents one case with consent, source, starting state, intervention/action, evidence, result, limitations, and corrections. It can show delivery or learning in that case. It cannot prove universal causation or effectiveness.
Validation work
Validation must name exactly what is being validated, against which criterion or evidence, for which population/context/language/version, by what method, and with which limitations. 'Validated' without scope is too broad.
Impact evidence
Impact claims require outcome definitions, baseline/comparison logic where relevant, time period, attribution limits, missing-data handling, harms/adverse outcomes, and transparent analysis. Appreciation or originator belief is not impact evidence by itself.
Current meUus research position
Until verified results are recovered, the safe current language is: developing research foundation, research direction, blueprint, methodology draft, instrument/question work, pilot preparation, under review, not validated yet. This protects the real work without inflating it.
Reflection
Questions to sit with
- What exactly would success mean, and how would it be measured?
- What result would make us reduce or reject the original claim?
- For whom and under what conditions would any validation apply?