Please help us help you and fill where relevant:
LimeSurvey version:  Version 4.3.4+200713
Own server or LimeSurvey Cloud: Own server
Survey theme/template: Sea Green, Default LimeSurvey admin interface template.
==================
(Write here your question/remark)

Hi,

I'd like to randomize the order of answer options both for radio button like and check box like questions. I therefore selected the Yes option for the Random order option in the Display section but to no avail. The answer options are always displayed in the same order. What is happening here and how could I solve the issue.

Thanks in advance for your help,

Nicolas

Hi,
we do not see what you really did.
So send the lss export of the survey (only the relevant questions)

Joffm

LimeSurvey version: Version 4.3.4+200713

So you are running a over 4 year old installation, that hasn't received any updates, bug fixes and security patches since 13/07/2020?

Also, the Version 4.x has soon be replaced by LS 5.x and is EOL for a long time now. You really should consider updating to LS6 (LS5 will be EOL soon - actually I see that it is EOL since june 1st already).

Dear Joffm,

Please find enclosed one lss export. All sociodemographic questions are similar to this one.

Best,

Nicolas 

Hi,

Thanks for your answer. I'll ask our IT specialist to install a newer version. I guess it is best to switch version only after data collection is over, not while there's a questionnaire active?

Best,

Nicolas

Hi,
You sent a lsq export.
We always ask for lss.

Reason: lsq and lsg exports are language sensitive.
You can't import them into a survey with a different base language.
So we have to guess the base language of your export, have to create a new survey with this language to be able to import.

Furthermore all surveywide settings are missing which may be important.

Please ease the work of us volunteers.

Joffm 

I am so sorry. I did not pay attention to the format. As a matter of fact, lsq is the only format I can use to export a single question. Enclosed please find the lss format of the survey. The questions for which I'd like to use the random order feature for answer options are questions 24 to 27. They are all in group 4.

Nicolas

Hi,
1. I do not see any issue.
 
But if you have a look at the release notes
[url] github.com/LimeSurvey/LimeSurvey/blob/ma...cs/release_notes.txt [/url]
you see

hanges from 4.3.16 (build 200915) to 4.3.17 (build 200921) September 21, 2020
...
-Fixed issue #16229: Random answer order currently broken (GabrielJenik)
...

And in the bug report
[url] bugs.limesurvey.org/view.php?id=16229 [/url]
and you see the history.
So this is a bug in your version.

You may remember, LS 4.x. had a very clumsy start, so that there was a new launch with version 4.4.

But I have no idea why you want to confuse your respondents by this randomizations. Without randomization everything is clearly ordered that the respondents finds his answer quickly.
And the "no answer" option really shouldn't appear in the middle of the options.

Two other things I want to mention.
1. Why didn't you create a multilingual survey
2. In my opinion it is really annoying to have to answer so many arrays with dropdowns (without need). For each item the respndent has to click two times.
 


Joffm

Thank you so much for the follow-up. This is so helpful and very much appreciated.

I strongly considered creating a multilingual survey but many years ago I experienced major issues with that. I guess I'm still under the negative impression it has generated in me.

We want to randomize the ordre of presentation of the answer options for demographic questions to go against heterocentrism. As is, the respondent always sees the Cisgender man and Cisgender woman options first thereby reproducing social hierarchies, this is at least a recommendation that was given to us by experts in gender studies we consulted.

From your troubleshooting I deduce that the solution to the issue is to install a new version of Limesurvey.

Thanks again for your prompt and helpful replies.

All the best,

Nicolas

always sees the Cisgender man and Cisgender woman options first thereby reproducing social hierarchies, this is at least a recommendation that was given to us by experts in gender studies we consulted.

I do not think so.
It is too confusing.
To be honest: If I see this "disordered muddle", I'd think "This is badly programmed" and do not trust the whole survey.

At least you should keep together
cis  (men, women)
trans (men, women)
the two "person" items

And btw: Do normal people know that they have to select "cis"?

Joffm