You can choose all answers on multiple choice questions, including choices that are mutually exclusive.

I’ll have to add that to my review checklist for next time. At least one of those was an error (ex. It doesn’t make sense to be in multiple levels of school currently). For others, it would be clearer to have a single selection with an option for “both about equally” (ex. Desktop vs. mobile usage).

a lack of definitions of terms. Is a town with 150 people “urban” because you have a street address and most people don’t work in another town/city, is it “suburban” because you need to go to a different town to buy groceries, or is it “rural” because that’s how most people who live there self-identify?

Many of those questions are intended to be self-identification, but we could have said that explicitly so that people aren’t uncertain. The reasons we didn’t have set definitions:

• People disagree on which definition/method is most appropriate, and we haven’t had the capacity to properly weigh the options / determine what value each definition might provide over the others.
• It seems that people are more curious about the self-identified groupings than the exact details. Both factor in to what the online experience is like, but the self-identification would play a larger role?
• Privacy. We want people to feel comfortable answering questions, without worrying that someone will figure out their real identity by aggregating the answers. It’s much harder to do that if it’s uncertain on why the user answered the way that they did.

Still, we are open to adding definitions to questions where it would make more sense to do so. For example, we added the fast.com and census/gov Canada links this time. Otherwise we can explicitly say that users should answer based on self-identification.

I appreciate the feedback! I’ve noted this down for next time