Research
Evidence-informed — and clear about what that means
What “evidence-informed” means here
Evidence-informed means our design is shaped by relationship research. It does not mean RKINDLE has been trialled and shown to produce a particular outcome. Those are different claims, and we will not blur them.
RKINDLE’s seven areas and its own indicators are a working framework. They have not been psychometrically validated, and we describe them that way everywhere they appear.
If we ever run or commission validation work, we will publish what we did and what we found — including anything that did not work.
Design principles
Structure before interpretation
Scores, thresholds and eligibility are worked out by fixed rules, not by a language model. The same answers always produce the same result.
Interpretation stays tentative
Explanations are offered as possibilities for you to test — “this may be happening” rather than “this is what is wrong”. You can tell RKINDLE when a suggestion does not fit.
Measurement, not vibes
You answer the same check-in at the start and at the end of six weeks, so change is observed rather than assumed.
Behaviour is where change happens
Insight on its own rarely changes a relationship. The plan focuses on small, specific, repeatable actions.
Versioned and reproducible
Every question set and every scoring rule carries a version. A result can always be read against exactly what was asked at the time.
Honest about limits
RKINDLE’s own framework has not been through psychometric validation. We say so wherever it appears, and we keep our own measures separate from any established outcome measures we may add later.
References
Reference list in preparation
We would rather publish nothing than publish a reference we haven’t checked. Our source list will appear here as the assessment content is finalised.

