If they word it like they did in the article I think asking their research question (asked to kids) was itself reinforcing a harmful understanding of gender and the kind of characteristics that can be applied to them. Individuals lie. Entire genders do not. Basically nothing applies across the entire category. In high school (the age of a lot of these respondents) they teach you to watch out for tricky multiple choice questions that sneak in an over inclusive option.