[Oh, forgetful me ...]
So, for the umpteenth time, I got ticked off at a web site
that required new registrants to pick a gender of
"male" or "female" before leaving a comment, and for the
several-th (but less than umpteenth) time, I fired off a note
to the contact address for the site, asking why that was a
mandatory field, and if it was mandatory then why didn't it
include the right answer.
Uh, unlike most other web sites I've prodded about this,
instead of getting completely ignored, or getting a patently
insincere "to serve you better" with no farther explanation,
or a buzzword-based brush-off, I got a polite note from a
manager, explaining (acceptably vaguely) why the question is
there[1], promising to take it up with the relevant
director when he gets back from someplace-else[2], ...
... And asking for suggestions for labels to add
to the list for that field, other than "other", which I
had already pointed out was, by definition, 'othering'
when it's the only option shown beyond M and F.
I want to make sure I'm giving good advice.
I remember reading some useful discussion of exactly
this question sometime in -- uh, the last eighteen
or twenty months? -- but can't remember where.
It was the kind of discussion where folks supported their
opinions and tried to take into account data-analysis
messiness as well as the feelings of us folks with not
quite "standard" gender (and, IIRC, at least part of the
conversation looked specifically at "understand our
readership better" and "tell advertisers what our
demographic is" reasons for collecting the data in the
first place, as opposed to medical situations or dating
sites, for example.) Do any of y'all happen to remember where
that/those discussions took place, or maybe even have
them bookmarked? One of the trans-issues sites?
Folks trying to get LJ to improve the gender field
in user profiles? DW trying to decide how to set up
the same field?
I'm mostly looking for a pointer to the discussion I
remembered seeing before, rather than wanting to hash it
all out from scratch in comments here, but I'll take
folks' thoughts here too.
Off the top of my head, I'm inclined to suggest
"male", "female", "both", "neither", "other", as a
reasonable (though imperfect) starting point, assuming
that they want to keep a pull-down list, don't want to
try to list every gender-identity label currently in
use, and think too many people would pick "decline to
answer" if that were offered[3]. But I've got this
nagging feeling that there were some problems with
that scheme that came out in the last discussion,
that I really ought to remember.
[1] "This helps us better understand and define our
audience which will in turn define and shape the future of our
business. This is primarily a research question." This
makes me wonder just what they think the gender info is really
telling them -- are they working from a stale (or exaggerated)
list of expected behaviours/tastes based on gender, or keeping
careful track of how what correlations there are change over
time, or just tweaking their content by trial and error to
try to keep the male-identified:female-identified ratio in a
range that makes their advertisers happy? But ultimately,
not my problem. I just want the gender field to be made
more inclusive, optional, or both. Whether they're being
smart about what they do with the data, I'll probably never
know.
[2] But no promise that anything would actually
get done, because he doesn't know whether they have the
ability to alter that part of the form -- which I'm guessing
means that the registration/comments section of the site
is a package they bought somewhere else or a setup hosted
by someone else, rather than a system developed in-house.
[3] LiveJournal, which has "male", "female",
and "unspecified", appears to have about 28% of users
listed as "unspecified" (though the
stats page
makes the male & female numbers add up to 100% and
ignores the unspecified precentagewise). InsaneJournal,
with the same list of options (and the same way of counting
percentages),
shows about
two-thirds of users picking "unspecified". Dreamwidth,
with "male", "female", "other", and "rather not say",
shows 39%
under "rather not say" and 1.6% under "other".
I'm guessing that the
folks I'm talking to aren't going to like the idea of
having one to two thirds of the answers to the gender
question fall into a "myob" category if they do change
the registration form.