Formative vs. summative evaluation
IEC 62366-1:2015 + AMD1:2020 · Formative and summative evaluation
The short answer
Formative evaluation is iterative and exploratory, takes place during development and has no fixed pass/fail — it improves the design and surfaces use problems early. Summative evaluation (validation) happens at the end of UI development with pre-defined acceptance criteria and provides objective evidence that the device can be used safely. A frequently cited practice (and FDA expectation) is around 15 participants per user group for summative evaluation — this is common practice, not a requirement of 62366-1 itself.
The difference at a glance
| Aspect | Formative | Summative |
|---|---|---|
| Timing | During development | At the end of UI development |
| Purpose | Improve design, find problems early | Objectively demonstrate safety (validation) |
| Character | Iterative, exploratory | Final, criteria-based |
| Acceptance criteria | None fixed | Pre-defined |
| Result | Design insights | Objective safety evidence |
On the number of participants
For summative evaluation, FDA practice and many guides cite around 15 participants per distinct user group. Important: this number is not a numeric requirement in 62366-1 itself — it is common practice and an FDA expectation. Justify your chosen sample size in relation to your user groups.
Why both are needed
Formative evaluations stop the summative evaluation from failing. Testing only at the end risks expensive redesigns after validation. The formative phase is your early-warning system; the summative phase is your evidence.
Frequently asked questions
Does 62366-1 require 15 participants?
No. 62366-1 names no fixed participant count. The "15 per user group" comes from FDA human factors practice and is a widely used but justifiable convention — not a standard requirement.
Can a formative evaluation replace the summative one?
No. Formative evaluations improve the design but do not replace the final, criteria-based safety evidence of summative evaluation.