Twelve statements. Tick the ones that are true of your program today, then read your band in the table below. The score says whether to run a summative OSCE next term, run a formative one first, or go back to the blueprint. Every check maps to a section of this guide.
The test takes ten minutes and is worth running twice: once by the exam office, once by the clinical lead. Where the two scores differ, you have found the thing to fix.
The twelve checks
What your score means
Score | What it means | Read this first |
|---|---|---|
10 to 12 | You can run a summative OSCE. Fix the gaps before results carry weight. | |
6 to 9 | Run a formative sitting first. The exam will work. The defense will not. | |
0 to 5 | Start at the blueprint. Stations written without one are hard to defend. |
A score is a prompt, not a verdict. Two checks carry more weight than the rest. A school with no blueprint and no standard-setting method has an exam that produces marks but not decisions. Those two are the ones to fix first, whatever the total says.
Where the checks come from
The list is the operational spine of AMEE Guide 81 Part II, cut to the items a school can verify in an afternoon. The ordering follows the seven-step framework in Nyangeni 2024, which runs from policy and standard operating procedures through the blueprint, the station bank and peer review, the marking sheet and the standard-setting choice, examiner and simulated-patient training, the venue, and finally piloting. That paper describes an exam of 10 to 20 stations of 5 to 10 minutes across five campuses, so the steps are tested at scale.
Two checks are worth reading closely. Station count determines reliability, and the 2020 Ottawa consensus holds that fewer than 12 stations or less than 150 minutes of testing time will not yield reliable results. The standard-setting check matters because the method has to be in the regulations before the exam runs, not chosen once the marks are in.
Frequently asked questions
What score is safe for a high-stakes exam?
Ten or more, with the blueprint and standard-setting checks among them. Below that, the exam can still run. What it cannot do is carry a progression decision that a student might challenge. The gap is rarely the stations. It is the paperwork that shows why the pass mark is what it is.
We fail on examiner training. Can we still run the exam?
Yes, for a formative sitting. Not for a summative one. Examiner stringency moves pass and fail decisions more than most schools expect, and untrained examiners widen that spread. Run the sitting, collect the scores, and read the examiner and site effects page before the marks count.
How long does it take to move from six to twelve?
One academic year is realistic for a school starting from written exams. Station writing and examiner training take the longest, because both need clinicians. The venue, the run sheet and the results process are weeks of work, not months. Piloting one circuit shortens all of it.
Does buying software change the score?
It changes four checks, not twelve. Software holds the bank and its versions, prints the pack, collects the marks and computes the statistics. It cannot write your blueprint, train your examiners or ratify a result. Use the software requirements checklist once the other eight are in hand.