These studies should be surveying these groups instead of individuals I think. Sometimes the "group" will just be one person, and that's okay. But I wouldn't want someone's opinion if they're making it on behalf of a group without the groups input, because that's not realistic