> "What is your gender?" The multiple-choice options were "Male", "Female", and "Transgender". I was very disappointed at this 101 mistake, and sadly opened an issue referencing the question. The body of my issue read:
> "'Transgender' is not a gender. Transgender people may be male, female, gender queer, non-binary... If you want to know if a survey respondent is transgender, you need to explicitly ask that question."
> The next day I got an urgent request for a call with my manager. She told me that the data scientist who had written the survey questions was very upset and had gone to her manager to complain about me.
I can't help but shake my head reading thru the whole piece. It's like a soap opera. Everyone's so caught up in the social issues it's fucking impossible to not offed anyone.
Maybe she is right, but I've never in my life seen a form that explicitly asks "Are you transgender". So I don't agree that this is a 101 mistake. Most forms ask "Male" or "Female", and that's it.
The fact that Github asked "Transgender" in addition to this shouldn't be seen as offensive, as I am sure they were doing this simply to make sure transgendered people felt included too. Sadly a good deed is not always appreciated.
UK perspective: it may depend on why the data is being collected.
UK government guidelines [1] for general forms to do with benefits/services &c are fairly thin
Organisations that have to maintain equality metrics in some form [2] have a more complex approach. This second approach seems like the nearest to the Github situation. The example I have cited (third or fourth in goog search results) maps directly to 'protected characteristics' in diversity legislation in UK.
Meta comment: Survey questions are tricky to formulate. Kicking ideas round in a group on a whiteboard with reference to external advice/exemplars seems to help get wording sorted and clarify the taxonomy. Genuine question (I don't work in the software world) is it normal for a lone web developer person to be left just making up the questions as they are coding them?
The best (worst) part is that this thread is now repeating that dynamic writ small... I admit I may be somewhat complicit in that, but it's important to be able to take a step back.
> "'Transgender' is not a gender. Transgender people may be male, female, gender queer, non-binary... If you want to know if a survey respondent is transgender, you need to explicitly ask that question."
> The next day I got an urgent request for a call with my manager. She told me that the data scientist who had written the survey questions was very upset and had gone to her manager to complain about me.
I can't help but shake my head reading thru the whole piece. It's like a soap opera. Everyone's so caught up in the social issues it's fucking impossible to not offed anyone.