Hello. I am trying to run a survey wherein I ask respondents a question about 10 randomly-selected pairs of states. Since there are 50*50 = 2500 possible state pairs, it would be very unwieldy to create a block in the survey for every single possibility and then have Qualtrics randomly-select 10 blocks.
One solution I thought of was the following: Create State1 = Alabama, State2 = Alaska, ..., State50 = Wyoming as 50 embedded data objects. Then, generate StateCodei_1 through StateCodei_10 and StateCodej_1 through StateCodej_10 by taking a random draw from 1 to 50. (This may lead to i and j being the same state on occasion, but let's put that aside for a moment.) Finally, use "piped text within piped text" along the following lines: ${e://Field/State${e://Field/StateCodei_1}} and ${e://Field/State${e://Field/StateCodej_1}} to call the relevant state names that have been randomly selected for the first of my 10 state-pair questions.
However, using piped text within piped text in this way doesn't seem to work. Am I doing something incorrectly, and is there a way of getting this to work? If not, do you have any suggestions about how I could implement this in a different way?
Thank you!