Limesurvey version: 6.4.10
Limesurvey hosting
Theme: Bootstrap Vanilla ( vanilla )

Hello, 

I want to carry out an experiment in which 8 different question lists must be pseudo-randomly distributed to participants. From what I know, pseudo-randomization using limesurvey is no simple task and maybe even impossible without complex coding procedures. Instead, I want each participant to complete a random list out of those 8 lists. What I have in mind is to create a "selection" questionnaire and provide a single link for my participants, which then randomly generates one of the links for 8 different questionnaires. 

To accomplish this task, I understand that I need to create a question and generate a random number between 1-8.  

And then, I need to change my endurl to this by using the text elements option:https://example.com/index.php?r=survey/index&sid={if(random == 1, 111111, if(random == 2, 222222, if(random ==3, 333333, if (random==4, 444444, if(random==5, 555555, if(random==6, 666666, if(random==7, 777777, 888888))))))}I have the following questions in this regard: 

1) What kind of question do I need to create? What code should I use to generate a number between 1-8 and where should I write this code? (Equation part or something?)

2) Should I just enter the link above into the endurl part directly? (Just like we do in providing a single url at the end?)

Are these "questionnaires" so long and so different that you can't use just one and display the respective groups or parts to the participant by conditions according to your random number.

If I read "8 different question lists" it sounds that it shouldn't be too long.

And, please, answer the questions that appeared at the beginning!
Not only, that you ignored them, you removed them.
 

Joffm

Thank you for your reply, I added the required information.

Both question orders and little details in the questions are different across 8 different questionnaires. My intention is to investigate the effect of these little details. Each participant should complete only one of the questionnaires, which takes around 35 minutes.

Still you did not answer the questions about version, hosting

So, if there are "only little details" then I'd think it better to use just one "questionnaire", meaning one survey and use "micro-tayloring" according toi the random number.

Best you provide the lss export of one or two of your "questionnaires"

Joffm
 

Still you did not answer the questions about version, hosting

Seems like they have been added to the original post.