I'm sure there must be setting that I have wrong, but I noticed if a quota is set at say 5 and if one of the initial survey takers (1-4) lingers in the survey and doesn't respond, goes away, comes back later after the quota has passed, that lingering survey taker can finish the survey even though the quota has already been met and the survey has been closed.

I'm pretty sure there is no way to defeat this except to accept that any lingering participants may have their choices put into the survey even after the survey has closed.

At least that's what I'm seeing. .lss attached.

Hi, this is more or less the normal behaviour of any survey tool.

Example:
You set a quote: 50 male, 50 female.
After you collected 48 female answers, four females start the survey at nearly the same time.
All will pass the quota check, because the program doesn't know if they will complete the survey. Only after completion the quota is updated.
And participants who passed the quota check once will not be checked again if the resume later.
So, if all complete the survey you will have 52 females. So, what?


You may set the quota check at the end of the survey. Some of the four will answer faster than the others.
But this - of course - will annoy the participants who answer the entire questionnaire to see "Sorry, quota full"

Joffm

Yes I figured. If rewards are tied to completed surveys it just makes things a tad bit more complicated if there are a finite group of rewards yet extra participants partake. But I can deal with that. Thanks for the assessment what you says make perfect sense.

Joffm wrote: Hi, this is more or less the normal behaviour of any survey tool.

The behavior is no bug when we take the quota features of LimeSurvey into account.

You might would expect, that resuming a survey can be done with or without checking quota criteria again.
If you resume and the quota is filled, the survey session can be ended.
Depends on how much you care about respondent health

If LimeSurvey would offer Cross Logic Quotas you might could be more easily move respondents to quota left open. That is now mostly done by hand and with the loss of a few respondents.

Quotas in LS are still a little bit "rough", but as Joffm said, I don't think there is much that can be done in those cases described. You can try to reduce the session time to avoid "lingering", but this might cause problems on other ends. So you need to see what is more important to you. And: I think the number of additional participants should be fairly low in most cases. Yes, you might have to pay them additional incentives.

currently a solution is to create quota on Hidden question.

See bugs.limesurvey.org/view.php?id=11597#c40585

Need to create the Feature request AND the pull request …