Good morning,
I would like to create an anonymous survey in which a user can save his/her own responses so that he/she can resume it later.

At this point I show you 2 scenarios identified by me:

1. After accessing the survey via an invitation link, user saves responses via the appropriate button, he/she closes the survey window and then he/she accesses the survey again via the same invitation link. Responses are saved but when user inserts new responses or modify the previous ones and he/she clicks button to save responses again, the responses are deleted, ie they all come back on "no answer"
2. After accessing the survey via an invitation link, user saves responses, he/she closes the browser or turns off the PC and then he/she accesses the survey again via the same invitation link, but the survey starts from the beginning, ie you have to enter again the captcha and then start a new survey.

What I would like is that in both scenarios the user can access the partially completed survey and that when saving new responses or modofy the previous ones all the answers are not deleted (as in scenario 1)

The settings entered are:
- Anonymized responses: ON
- Enable token-based response persistence: ON
- Participant may save and resume later: ON

The version in use is: Version 3.15.0 + 181008

I hope I was clear, thank you!

Scenario 1: user saves responses via the appropriate button
Scenario 2: user just close browser window

Just to be sure, what is the "appropriate button" in scenario 1?

When "Anonymized responses: ON" the token is not saved in the responses.
So reentering the survey via the invitation link with token, should always start with a fresh survey.

So when "Anonymized responses: ON", the option "Enable token-based response persistence:" is no longer relevant.

The "Participant may save and resume later" option could be still valid, when entered with a code/cookie, which is owned by the respondent /browser and in to connection with the token.
Reentering the survey via the URL with token would be OK, as the token would not be saved in the response data set.

The function "Participant may save and resume later" may need an improvement. The cleanest approach would be to not allow this option, when "Anonymized responses: ON".

Just to be sure, what is the "appropriate button" in scenario 1?

It is the "Save insert data" visible at the top of the survey.

The "Participant may save and resume later" option could be still valid, when entered with a code/cookie, which is owned by the respondent /browser and in to connection with the token.
Reentering the survey via the URL with token would be OK, as the token would not be saved in the response data set.

This is valid when browser is not closed, at the reboot of browser, when reentering the survey via the URL with token it starts with a fresh survey.

The function "Participant may save and resume later" may need an improvement. The cleanest approach would be to not allow this option, when "Anonymized responses: ON".

This means that it is not possible save responses and complete survey at second time when "Anonymized responses: ON"?

jelo wrote: …
The function "Participant may save and resume later" may need an improvement. The cleanest approach would be to not allow this option, when "Anonymized responses: ON".

No ,

I think it's OK : anonymous ON or answer persistance OFF : show a save forum where user can put anyname and a good password.
This user/pass reload the survey saved.

Else :

After accessing the survey via an invitation link, user saves responses via the appropriate button, he/she closes the survey window and then he/she accesses the survey again via the same invitation link. Responses are saved but when user inserts new responses or modify the previous ones and he/she clicks button to save responses again, the responses are deleted, ie they all come back on "no answer"

This part must be corrected : the saved response must not update current response.
I try to reproduce and put clear step on bugs.limesurvey.org/view.php?id=13861

Because the cirrent bug report is useless since we don't know how it's the survey about anonymous …

jumpyjumpy wrote: [It is the "Save insert data" visible at the top of the survey.

Save insert data? Never seen in LimeSurvey. Screenshot possible?

This means that it is not possible save responses and complete survey at second time when "Anonymized responses: ON"?

When you enter the token again (via URL or directly), LimeSurvey cannot find the old response, because the TOKEN is not saved with the response dataset. If you would be able to access the response again, it wouldn't be an anonymized response.

I think it's OK : anonymous ON or answer persistance OFF : show a save forum where user can put anyname and a good password.
This user/pass reload the survey saved

Ok, then this is the only way to partially save survey and resume it at a second time?

Save insert data? Never seen in LimeSurvey. Screenshot possible?

It's in Italian but the meaning should not change.

DenisChenu wrote: I think it's OK : anonymous ON or answer persistance OFF : show a save forum where user can put anyname and a good password.

I would prefer more options here. Username/password is fine. But a unique accesscode, which is created by LS would be nice too. Accesscode, URL and QR-Code on a screen or the option to create your own username/password.

jumpyjumpy wrote: It's in Italian but the meaning should not change.

Thanks, we are talking about the same button

jelo wrote:

DenisChenu wrote: I think it's OK : anonymous ON or answer persistance OFF : show a save forum where user can put anyname and a good password.

I would prefer more options here. Username/password is fine. But a unique accesscode, which is created by LS would be nice too. Accesscode, URL and QR-Code on a screen or the option to create your own username/password.

You mean :
Save create an unique access code to reload THIS response in this survey and show it to user ?

Seems a good idea : gitlab.com/SondagesPro/coreAndTools/reloadAnyResponse/issues/2 :)