Typically national and government surveys - even if they are quite large - only have a few gay and bisexual men in them, they’re good at comparing gay and bisexual men to other men, but there are too few to do comparisons within the group.