I would like to filter out some survey-takers after a few initial questions, and have the rest take the entire survey. Fake example: I'm surveying students' eating habits. First 2 questions are yes/no: "Do you have a kitchen?" followed by "Do you cook?" I want people who respond "no" to "Do you cook" to then exit the survey (with an appropriate message that would be different from the end-message for those taking the whole survey), while those who respond "yes" to "Do you cook" would then go on to the subsequent questions. (I don't want to simply have an initial statement saying that the survey is only for people who cook, because I want data on how many people have kitchens, and if so, whether they cook or not.)

This is not simply a conditional, unless I make every subsequent question conditional on "yes" to "Do you cook". This is awkward and risky. Surely there's an established way to end the survey early for some survey-takers, contingent on responses to particular questions? And, to provide them with an end-message (e.g., "Thanks, your further participation is not necessary")?

Thanks for any help!

Look for Quotas.

docs.limesurvey.org/Quotas

There you can use special Endmessages and EndURLS.

Thanks, that seems to make sense, but...In using quotas to screen out everyone (as opposed to some specific number of people, as illustrated in the user manual) who answers "no" and include everyone who answers "yes", do I set the quota to "0" for the answer "no"? I'm trying this, but when I test-run the survey (i.e., I enter responses, but my responses are not saved), there seems to be no quota affect--that is, the survey continues whether I respond "yes" or "no".

Activate the survey to test the quota.

How should responsed be saved, when you are in the demo mode?

Yes.

Jelo is correct, the survey must be activated to test the quota.

Thanks. I figured that activation was required, but I'll add that it's not obvious, at least to a non-programmer, that you can't test quotas but CAN test conditionals, as both involve a response to one question triggering a subsequent question.

A suggestion to the writers of the LimeSurvey user manual:
Specifically state that a quota can be set to 0. I'd guess that it's more common for a survey designer to want to screen out an entire group of ineligible survey-takers than to want to set some non-zero limit on a group. So, calling the function "Quota" (rather than, say, "Filter" or "Screen") and not providing a screen-all-ineligible-survey-takers example of using it, suggest to the (still-learning) user that using "Quota" is not the way to go. It also makes it hard to find when Google-searching for a way to screen survey-takers, as "quota" is an unlikely term to use. Finally, it would be helpful to explicitly say that Quota use is not testable unless activated.

For the record: The person for whom I'm designing a survey has made a financial contribution. It's the only fair thing to do! Thanks for making LimeSurvey publicly available.

sD960 wrote:
Since the manual is an wiki you can edit the text to help get newbies started.

Screenouts and Limits are both used in studies.
In Limesurvey there is a lot of programmer thinking.
To use a quota of 0 as a screenout is not that strange.
Would be interesting to know how many people are using quotas at all.

Other packages offer a Questiontype "End of Survey / Redirection" and then you
need to set your conditions / filters to screenout.
That would be helpful since not all questiontypes can be used in quotas.

Perhaps LS3

Well, I thought I was a newbie myself, so I'm less than confident about editing the manual--but maybe the perspective of someone who thinks like a researcher (my background) rather than like a programmer (not my background) would be helpful to other new users.