Sorry, Helen,
but there is nothing in it.
The relevance equation only selects the start point.

But where do you count the wrong answers?

You see the provided sample survey with the "counter" and the relevance equation ...strpos(counter,"00000") (for five consecutive wrongs) works without problem.

Joffm

Hi Joffm,

Thank you for your patience. This has been a bit of a learning curve for a technophobe like me.

I went over your dummy survey several times to see where I had gone wrong.
Because it might help others who are struggling like me, I list the mistakes I made:-

I needed the relevance equation in both the vocabulary questions AND in the vocabulary equation questions. Maybe obvious to most, but wasn't to me.
I noticed you created the 'counter' in the 'A bit about you' group, whereas I had put it in the vocabulary section, so I moved it. I don't know whether this would have made a difference.
I FINALLY noticed that the relevance equation you created referred to AgeYear whereas in my survey, the question code was AgeYears, so no wonder it didn't work.
I also noticed that there were differences in the syntax in the relevance equation in the dummy survey, compared to your message (differences in brackets, " versus ' , AND versus and ). I don't know enough about coding to know whether this made a difference or not.

But it finally works. Hurrah! I'm off to have a glass of wine and a lie-down.
Many thanks, again, from an exhausted mature PhD student.

I hope my errors help somebody else, some day.