Gender identification in survey

[rootwork]

Saw the survey on the Diversity and Inclusion Drupal Slack channel, and since I wasn't on during that workgroup meeting, I thought I'd post some thoughts on the gender identification question.

I would consider making these checkboxes rather than radio buttons. In a workshop I co-led on inclusive data collection in 2015, audience feedback included folks who identified as more than one gender identity, and noted that even in otherwise inclusive surveys they still couldn't fully identify themselves.

I don't remember how that person identified specifically, but one could for instance identify as both male and FTM -- and having people choose one or the other could be read as implying FTM folks are not "really" male.

The National Transgender Discrimination Survey from 2012 splits the issue into four questions (see pg. 189 of the PDF):

1. Do you consider yourself to be transgender/gender non-conforming (yes, no)
2. What sex were you assigned at birth? (male, female)
3. What is your primary gender identity today? (male/man, female/woman, part time as one gender and part time as another, a gender not listed here)
4. For each term listed, select to what degree it applies to you (transgender, transexual, FTM, MTF, intersex, gender non-conforming or gender variant, genderqueer, androgynous, feminine male, masculine female or butch, a.g. or aggressive, third gender, cross dresser, drag performer, two-spirit, other)

That may be more detail than you want to try to capture in this survey. But I do like the mechanism of asking how someone was assigned at birth and how they identify now (questions 2 and 3). It allows for identifying people who are on the trans* spectrum, but doesn't imply that their current identity is somehow different or other than cisgender identities.

[drnikki]

@rootwork thanks for this thoughtful contribution! I'd concede that including all four of those questions would give us more detail than we're planning on collecting. We're also a bit limited by the options available on the survey platform itself. I'll do some fiddling with this over the weekend and see if I can find a way to include more inclusion without adding too many more questions.

[nstielau]

Thanks @rootwork, good suggestions. That survey is gold; great data and one of the few I've found that seems to get it right.

I've made a change in #74 . Basically taking questions 1 and 3 from your suggestions, which I think allows for inclusive responses and good data, without adding extra questions or complicated responses.

What do you think?