Hi all,

I just finished making a survey with which I want to validate 150 images. Each page shows 1 with 5 questions, the questions they have to answer about each image are always the same. The 150 question groups are randomized, so the ratings aren’t influenced by the order in which the images are presented. For this I wrote MYIMAGES in the “randomization group” field of all 150 groups. This seems to work perfectly. As every time I open the questionnaire, the order in which the questions groups (and thus the images) are presented is different. However, now that I’ve finished the Limesurvey questionnaire I wanted to use Prolific to get my sample. When I preview the questionnaire in Prolific, the questionnaire seems fine. The only think that isn’t working is the randomization: the question groups are presented in the order I’ve inputted them in Limesurvey instead of randomized. This is strange to me as Prolific just refers to the Limesurvey link.

Anyone who has worked with Limesurvey, Prolific, and randomization, who knows what’s going wrong here?

Kind regards,
D

doomen wrote: HThe only think that isn’t working is the randomization: the question groups are presented in the order I’ve inputted them in Limesurvey instead of randomized. This is strange to me as Prolific just refers to the Limesurvey link.

Without having a link or a complete LSS-Export a debuggin is difficult.

From you description you might now enter the survey different for testing and are not using a new session for every testrun. Try to use the anonymous mode of your browser to check the survey and open a new anonymous window with every testrun.

Hi Jelo,

Thank you for your quick reply.
Here is the link to the survey: tpsurvey.ugent.be/limesurvey315/index.php/446526?lang=en
As you can see, I have a few demographic questions (4) and a few questionnaire questions (16) before the actual image validation study starts. Using this link you will start with an different image each time you start the questionnaire.
I can't share a LSS-Export right now, but can do so tomorrow.

I may be incorrect, or not understand you fully. But I don't think that not entering a new session for every testrun is the problem. Going incognito/ changing computer etc. gives the same result. When purely using the limesurvey link or limesurvey preview the image groups are randomized. But using Prolific preview (which supposedly just directs to the Limesurvey link), I will always get the first image group first, then the second, then the third etc. regardless of computer/ incognito/ etc.

Kind regards,
D

doomen wrote: When purely using the limesurvey link or limesurvey preview the image groups are randomized. But using Prolific preview (which supposedly just directs to the Limesurvey link), I will always get the first image group first, then the second, then the third etc. regardless of computer/ incognito/ etc.

You might elaborate what "Prolific preview" you conduct. A simple redirection via HTTP cannot change the behavior of LimeSurvey to that extend. So just inviting people via a panel cannot result into a fixed question order which is randomized when accessing the survey directly.

I expect you just doing a simple preview check.
E.g. researcher-help.prolific.co/hc/en-gb/art...reviewing-your-study

Is the survey activated inside LimeSurvey while testing via Prolific?

See attached for the Prolific preview I use.

It seems to me that it redirects you to the limesurvey like it will redirect future participants to the limesurvey. So it's very strange to me, because indeed, I thought that a simple redirection would not change the behavior of LimeSurvey at all.
The Survey is activated right now inside Limesurvey, but not inside Prolific.
See attached for the LSS-Export you requested yesterday.