Throughout my grad school experience, from conference registration forms and university-wide surveys to actual grad school applications themselves, I have often run into the following question:
Always making me think: Why is this still a thing?
Okay, certainly there are legitimate reasons to collect demographic information, including information regarding gender. Whether to explore the diversity of the field (i.e. graduate admissions or university surveys) or whether it is needed for grant purposes (i.e. conferences) there are times when inquiring about an individual’s gender identity is reasonable. However, why must this information be collected in such biased fashion?
The fact of the matter is that for many individuals gender is a complicated social construct, with which they have a nuanced relationship. As phrased above, the question “Gender?” takes all of these complexities and nuances and smashes them; smashes them into two extremely narrow little circles. It takes individuals and smashes their agency, forcing them to squeeze themselves into a narrow space predefined by someone else.
Moreover, to some who are “gender-expansive”, this question is yet another instance of someone — a colleague, a peer, a fellow mathematician — neither recognizing nor accepting part of who they are. It is yet another instance in which they are in part invisible:
“invisibility may seem like a small price to pay … But invisibility is a dangerous and painful condition. When those who have the power to name and to socially construct reality choose not to see you or hear you … when someone with the authority of a teacher, say, describes the world and you are not in it, there is a moment of psychic disequilibrium, as if you looked into a mirror and saw nothing. Yet you know you exist and others like you, that this is all a game with mirrors.”
~Adrienne Rich, “Invisibility in Academe” 
In short, this question invalidates the feelings and experiences of those who do not find themselves comfortably within the confines of the two narrow little bubbles provided to them.
If I agree there are legitimate reasons to collect gender demographics, and if I find the above format highly problematic, what do I think the answer is? Well I think a better approach is to give agency to the respondents. As the Human Rights Campaign notes the least restrictive and most preferable option is to allow individuals to self-identify. For example, by re-phrasing the question in the introduction as follows:
Doing this empowers the respondent to define their gender identity as they see fit. I should also point out – not that it should matter – that phrasing the question in this way changes little how the question writer interacts with the responses. (Note other parts of academia have already begun moving to eliminate such biased and discriminatory questions.)
Some of you might be wonder why I have bothered spending ~500 words on a topic so “insignificant” as to how to ask someone their gender. The fact of the matter is that these things are not “insignificant” – they matter, and they have substantial effects on the lives and careers of many people. (I am reminded, in part, of the following tweet.) Sure, this might not have an effect on your life, but that does not mean the opening question does not make others feel uncomfortable, outcast, and discriminated against. And to say this is “insignificant” is to say these people are, well, insignificant …
Which is unacceptable. No person is insignificant. No one should ever be made to feel insignificant. No one should ever be … invisible.
For us to create a more inclusive, supportive, and welcoming mathematical community, these are the sorts of things we need to begin to think and care about. We need to stop seeing things like this as “insignificant”. We need to begin to see the world and individuals as they are: complexly. Most importantly, we need to step up and change the ways we act. Changing how we phrase this question is a small first step.
 Roberts, T. et al. The Broadview Anthology of Expository Prose 2nd Ed. Broadview Press. 2011.