Q always asks this question when you import a new raw data file because the process of setting up multiple response questions can be very time consuming. By asking first it allows you to skip this step.
If you answer Yes here then Q will try to automatically identify multiple response questions in your raw data by looking for compatible variables that are adjacent and share a common label prefix. For example, the variables "Q2. Cool - Cola", "Q2. Cool - Fruit Juice", etc. would be combined to make the question "Q2. Cool."
This is useful because many raw data files do not designate multiple response questions, although Q will respect any it does find.