I'm new to limesurvey and trying to understand what you can and can't do with anonymous surveys that use tokens.

Can you personalize emails sent to recipients of an anonymous survey (for example by beginning the email with "Dear FirstName")?

Can messages be personalized to groups of recipients of the survey based on on their answers to previous survey questions? (For example, if in a survey they told us they took the course Math 100, could we later send them an email that says, "We would like to ask about your experience in the course Math 100 last semester?").

Is there any way to add attributes to the list of survey participants that are connected to their anonymous answers? (For example, when uploading names/emails, identifying which ones belong to the control vs experimental group, and then having that information connected to the survey answers without connecting names/emails.) I think I understand the answer to this to be no, but I just wanted to check...

As the survey administrator, what exactly is visible to me if I conduct an anonymous survey with tokens? I expect that I would be able to see names and emails, and separately responses based on token ID numbers, but no connection between the two - have I understood that correctly?

Thanks for taking the time to read my questions.

If a survey is anonymous, there is no connection between their personal data from the token table (name, email, etc.) and the results. Thus you can't later make a connection to that in another survey.

I am not sure, but I think the tokens won't be stored in the answer tables in case of anonymous surveys. Otherwise you could make connections again, which would undermine the purpose of the survey being "anonymous".

So I think what you are trying to do, will only be possible if your survey is not anonymous.

If you need the survey to be completely anonymous, you will have to live with the disadvantages of this (not being able to recontact them, not being able to connect the survey results with the token table).

I am also not sure, but I don't think you can pass on the information about "control vs. experimental group" from the token table. Because if this would be possible, you could undermine anonymity as well.

Just create a question asking them whether they belong to control or experimental group at the start/end of the survey?