ipollert May I first ask how you are planning to distribute the survey? If you are using a site intercept, you could create two different surveys, with the intercept randomly assigning respondents either to the survey with the relevant question at the beginning of the survey or to the survey with the question at the end.
Also, if you are working with a mailing list, ,you could use the two survey approach while using an anonymous link, creating two anonymous links and randomly assigning half the population to each one.
You would then combine the responses after closing the survey so you could analyze them together. That may be the most effective approach--any other ideas from the Community?

Thank you or the quick response AdamK12 . The plan is to use a mailing list, but I really like the idea presented.
I have not performed something like this before though, are there any pitfalls I should be aware of when doing something like this? It seems straightforward enough to split and then recombine in my head, but better safe than sorry.

My pleasure, ipollert -- I would suggest considering and planning how to recombine the data. If you are going to have two independent population samples, which are asking the same question but at different positions in the questionnaire, you could combine them in an Excel spreadsheet or in another tool like Tableau, and also assign a dummy variable so that you can easily determine which responses fall in which buckets. You may be able to do this within Qualtrics by combining responses from two projects, or externally to Qualtrics. I hope that helps!