I would like to know if it is possible for participants to be automatically (alternately) allocated to one of two versions of a survey, please? Where each version is part of one LimeSurvey file. Both surveys are identical except for the order of answer options. I would like each survey to have a randomised order of presentation of question groups (using the randgroup descriptor). I was thinking perhaps I could do something like this:
1= welcome message, information & demographics AND {question groups 1 through 10, all randgroup1} AND end message etc OR 2= welcome message, information & demographics AND {question groups 11 through 20, all randgroup2} AND end message, with 1 and 2 alternating (automatic).
Please excuse the rudimentary attempt at (pseudo)syntax above, hope it gets my message across!
You cannot have "1 and 2 alternating (automatic)" but you can load a hidden equation question (let's call it random1) with a random number between 1 and 2 and then set relevance on the following groups based on the random number. This will not give an exact 50/50 split but it will be close if you have a large enough response pool. The exact 50/50 split is almost impossible to achieve as there will be drop-outs, concurrent responses. etc.