Any academics: how do I tackle research re sex/gender

Which is worse? A small population of males may respond as female and vice versa to your data gathering about a non trans subject or every data gathering exercise has to drag every one through the assigned at birth and gender identity farce as if it's in any way necessary to anything at all the vast majority of the time. It's an enormous irrelevance to most subjects.

What is your research question?

Do you care about gender? If you are only interested in disaggregation by sex, then I would just ask sex at birth and explain why, and what sex means.

Presumably if you are doing a survey then you will also discuss potential weaknesses in the methodology, and one of those weaknesses will be respondents answering questions wrongly for whatever reason.

I also wouldn't ask about other protected characteristics unless relevant as you shouldn't be collecting personal data you don't need.

Also not an academic

Those upset you don't want to know about their inner trans feelings will not take part and that's such a tiny demographic it's no loss at all.

You would hope perhaps this would be so, but Unless you specify biological sex, and also ask about gender separately, a trans woman will check the “female” box by sex thus harming reliability of the data.

2 questions are In this age needed otherwise some respondents will answer “sex” with their gender identity.

When (as a physical scientist/mathematician) I used to sit on one of my University Research Ethics Committees, I once pointed out that saying sex/gender given the current issues was not good enough. If they meant sex, they should ask for that. course, depends (not that I can recall now), if it was medics/biologists or social scientists/psychologists, and what precisely they wanted to know.

" I would just like to ask what is your sex, but I don't think I will be able to do that."
Why not? If some wokey discipline, I would expect they already have a standard form of include all six billion genders and don't offend anyone. If you mean sex, then you need to say that, in whatever way makes it clear you mean sex. I would find adding sex assigned at birth to a biologist etc to be misleading and frankly wrong.

Have a sex question and a gender identity question if you really need to, just to make clear you are distinguishing between the objective and the subjective.

Unless you are researching the labrynthine inner feelings of the trans world just ask male or female as:

1. anything else is irrelevant to any other subject in the world
2. Those upset you don't want to know about their inner trans feelings will not take part and that's such a tiny demographic it's no loss at all.

I would ask every question, in order, as stipulated in the Equality Act. So add in for example whether participants have a disability etc etc
Then you'll be able to disaggregate sex data. Follow it to the letter.
Disclaimer: Im not an academic. But this is how I'd do it.

Also I would add an asterisk saying that the first question is sometimes referred to as either the sex assigned at birth or the sex observed at birth.

What is the sex on your original birth certificate (choose one)
Male:
Female:
Indeterminate / intersex / NA:
Prefer not to answer:

How do you identify your gender: _ (free response)

To avoid woke attacks It would be worth Noting that you are doing this specifically to help understand How the intersection of multiple Vulnerable identities impact whatever it is you are studying.

What are you asking? I would ask what is your sex and what is your gender as two separate questions if relevant to health.

You could ask for 'sex assigned at birth' and 'gender identity' if you're really concerned about treading carefully.

I'm not an academic and I agree with your original phrasing rather than having to bend to something that I think is utter nonsense, but my suggestion is including both questions, in that same order.