Client is asking about "DualMaxdiff" meaning that we will offer respondent to pick the "best" item twice using 2 separate layers of comparison (i.e. most motivating and most believable). I think it is expected that 2 scales will be shown on the same screen.
If somebody can share script for putting second row of radio buttons and labels, it will be greatly appreciated.
I can do 2 separate Maxdiffs (and I would prefer it), but resulting scores may correlate less than expected, and finding the absolute winners could be more challenging. Anyway, I would like to hear your thoughts. Maybe use a split sample?