Saturday, February 22, 2014

Troubleshooting Social Science Surveys, Part 2: So You Wanna Make A Gender-Inclusive Survey, Huh?

This is the second half of my bloggings about gender-inclusiveness on surveys, mostly directed toward grad student researchers but including really anybody who ever needs to have people insert a gender in a form (dating site and social network profiles, for instance).  In part one I talked about really considering why you feel the need to separate male and female trans folk out of your results for male and female people anyway.  There are some things I didn't really detail, though, including both what to do if your situation does require disclosure and how to express non-binary genders.  That's what this is about.

First, let's revisit the survey questions that inspired me to write this:

This was a grad student's research survey, and it's clearly written by somebody who is at the very least recognizing that trans folk and non-binary folk exist and will likely be filling out the survey.  The problem is that it's pretty inundated with issues.  "Transwoman" and "transman" are rendered as one word rather than two ("trans woman" or "trans man").  A trans man's identified gender will be "man," not "trans man," and a trans woman's identified gender will be "woman."  "Intersex*" is not typically assigned as a sex at birth.  And so forth.

I don't want to pick on this one survey as if it's the cause of all ills in the world.  The problem is that these surveys are regularly cissexist or otherwise problematic.  What follows are some common mistakes people make and how to correct them.

1. Using "transgender" as a stand-alone category.

The above survey didn't do this.  As problematic as it was, I actually applaud it for that because it's one of the most common ways people try to be inclusive of transgender people.  They come up with a survey question that goes something like this:
  • What is your gender?
    • Male
    • Female
    • Transgender
This is actually really, deeply insulting to me and it's one of the shittiest ways you can try being inclusive.  Here's why:
  1. It strips trans men of maleness and trans women of femaleness.
  2. It lumps all trans people into the same gender category despite us being just as different as cis men are from cis women.
The first point is the most obviously insulting part.  "Transgender" isn't my gender.  "Male" is.  And yet when I see this on a survey, I know that the author expects for me to lump myself in with the other trans people and be happy to be included.  I would legitimately rather you not attempt to be inclusive than try making do with this M/F/T way of doing things.

The second point is especially problematic.  Trans women and trans men are in many ways as different as cis women and cis men.  Historical trans male invisibility has made it that when somebody identifies as "transgender," a lot of cis people will assume "trans woman," and I really worry about how many of these studies assume that trans respondents are female.  It's also important to recognize that trans men and trans women receive very different treatment and have very different histories.  In many cases it'll just make more sense to categorize trans men with other men and trans women with other women.

Oh, while I'm on this, I'd like to mention that this is particularly frustrating on dating websites.  I don't even bother using sites that use this format, because practically nobody browses the "transgender" section of a dating site looking for a trans guy.  The worst offender I saw, though, had these categories:
  • Man
  • Woman
  • Couple (M/F)
  • Couple (M/M)
  • Couple (F/F)
Not only are all trans folk--including crossdressers--lumped into one category, there are no trans options in the "couple" categories.  Because presumably we'll all die alone?

2. Using the concept of "gender identity" (or "gender you identify as/with") improperly.

One of my favorite "stare at computer screen, stunned" moments was when somebody posted a survey in a Facebook group that specifically called for "female-identified participants."  After it had been posted for over a week, a trans woman--not rare in this group--asked if the survey was open to trans women or if she was just looking for cis women.  As it turned out, this person was only looking for cis respondents.

This should never have happened, because the use of the term "identified" is specifically meant to be an inclusive term.  For trans people who are aware of the lingo, "female-identified" or "male-identified" are big, neon signs telling us that we're supposed to be included.  So if you're not willing to include trans folk, don't use "identity" to describe it.

The same issue is less prevalent in my original survey example.  The problem this survey author had was that they created an entire field dedicated to the gender a person identifies as and still separated "trans woman" from "woman" (and "trans man" from "man").    "Trans woman" is  not a gender identity.  Trans women identify as women.

There are two overlying options, here.  Option one is to ask two questions, one about gender identity and the other about assigned sex, like this one does.  When you do this you will have no reason to specify "transgender" at all.  For instance:
  • What sex were you assigned at birth?
    • Female
    • Male
    • Other (Specify)
  • What gender do you identify as?
    • Female
    • Male
    • Other (Specify)
This is a good option if you're only dealing with men and women, and it solves quite a few of the problems that I'll go over below, as well.  It does not specifically mention any non-binary gender identities.  I'll talk about that, later.

Option two is to only ask one question and include both gender identity and assigned sex in one answer.  Which actually brings me to this next problem people have...

3. Requiring trans people to mark themselves without requiring the same of cis people.

One of the most common mistakes people make when creating surveys--and especially when creating social network/dating profile questions--is to separate people into four categories:
  • Male
  • Female
  • Trans Male/FTM
  • Trans Female/MTF
This is less frustrating than lumping all trans folk together, but there are still problems.  In addition to ignoring non-binary people, it marks trans folk while allowing cis folk to be unmarked.  This separates "trans men" from "men" and "trans women" from "women" just as having a standalone "transgender" category does.  It just separates that category slightly better.

A more equitable option would be something like this:
  • Cis Male
  • Cis Female
  • Trans Male/FTM
  • Trans Female/MTF
(Note: The reason I include "FTM" and "MTF" here as well as "trans male" and "trans female" is because it makes it slightly less likely that a newer or inexperienced trans person will pick the wrong one.  Writing the whole thing out--"female to male" or "male to female"--might help even more.  You'd be surprised how many trans women think they are supposed to call themselves "transgender males" and vice versa.)

 "But people won't understand cis!" you may be saying.  Listen, people are only going to be confused as long as people refuse to actually use the word regularly.  Be the movement.  There is, after all, a marvelous invention called the "footnote."  Or you can put it in parentheses (parentheses exist now!).

But if you're certain this is going to taint your results, you could always do this slightly-less-appealing but more-readily-understood version:
  • Male (Non-Transgender)
  • Female (Non-Transgender)
  • Male (Transgender, FTM)
  • Female (Transgender, MTF)
4a. Treating "male/female," "transgender," and  "intersex" as mutually exclusive.

I don't want to speak too much on intersex folk here because as I am not intersex I don't have the authority of lived experience to go by.  However, it's important to recognize that being trans or intersex does not disqualify one from being a man or a woman... nor does being trans disqualify one from being intersex, nor vice versa.

Example?  A friend of mine is an intersex trans man.  He has an intersex condition, but was assigned and raised female.  When he was older he transitioned female-to-male.  His gender identity is male.  He was assigned female.  He is also intersex.  So just as "transgender" should never be used as a standalone gender category, neither should "intersex."

A lot of this has to do with peoples' confusion about what "assigned sex" means.  Hence...

4b. Treating "transgender" and especially "intersex" as assigned sexes.

The above example treats "intersex" like it's an assigned sex... but it really isn't.

There are very, very few cases in which somebody will be assigned anything other than "female" or "male" at birth.  A tiny minority of parents are attempting to raise their child gender-free, but this is independent of any intersex conditions those children may or may not have.  Intersex people are typically assigned male or female depending on convenience, preference of the parents/doctor, or closest outward appearance... even if no medical intervention is sought.  People often think of "biological sex" as being a rigid, easily-classifiable and quantifiable set of characteristics, but really it's as much a social construct as gender is (See: "Sex Is Not Between Your Legs").

It's rare, but occasionally I see this same thing applied to "transgender."  Have you ever really seen somebody "assigned as transgender" at birth?  Cultures where parents might raise a child in a gender we wouldn't associate with their appearance exist, but they typically don't use the word "transgender" to describe this.  A child who is raised outside of male and female is being raised as non-binary or gender-neutral... not "transgender."

That said, if you wanted all this information you could set it up in the same way you would set up later questions that don't relate to gender.  For instance:
  • What was your assigned sex at birth?
    • Male/AMAB (Assigned Male At Birth)
    • Female/AFAB (Assigned Female At Birth)
    • Other (Explain)
  • Do you identify as transgender?
    • Yes  (Explain)
    • No
  • Do you identify as intersex?
    • Yes (Explain)
    • No
The "explain" field is there because people have different reasons for identifying with all these terms.  Not every AFAB person who identifies as transgender is a trans man.  An intersex person might want to mention a specific intersex condition.  And so forth.

5. Using gender as a write-in with no standard options to try including non-binary folk.

If you're getting confused and freaked out, it might be tempting to just put gender as a write-in and do the categorizing later at your own leisure.  This might be fine if you have a small number of respondents and time to do this all manually (...or assistants), but there are some problems you might come across if you do.

First, realize that people are prone to use fields like this somewhat sarcastically.  Write-ins at a forum I used to frequent--not one for trans folk--included a lot of cis people identifying themselves as "dude," "guy," "penis-owner," "chick," and other things that they assume people will understand but which are baggage-heavy or cissexist ("penis-owner" doesn't say anything about your gender, after all).  If you are trying to get people to identify themselves as trans if they are, this isn't going to do it:  I know that I, personally, almost never put "trans man" in a write-in asking for gender.  I put "male."  The point here is that this method is likely going to be more trouble than it's worth.

However, write-ins are probably the most useful way of getting accurate identifications from non-binary-gendered folks... if you use them in combination with other options.

What I mean is the fabled "other" category.  It's great... if you choose not to use it just as an afterthought.  What I mean is that there should be an indication through your other options that you're putting it there in all seriousness and not just to avoid hearing people complain afterwards.  To use an example:
  • Male
  • Female
  • Genderqueer
  • Gender Neutral/Neutrois
  • Unsure
  • Other __________
This particular one would be for people who either don't require disclosure of trans status or have a supplementary question asking about it, but again a lot of trans folk who are accustomed to people insisting we disclose at every turn might stick "trans woman" or "trans man" in the "other" category (which, of course, you would then categorize with "female" and "male" respectively, right?).

A more comprehensive example:
  • What was your assigned sex at birth?
    • Female
    • Male
    • Other ___________
  • What is your current gender identity?
    • Woman
    • Man
    • Bigender
    • Genderqueer
    • Gender Neutral/Neutrois
    • Unsure
    • Other ____________
  • Do you identify as intersex?
    • Yes (Explain)
    • No
A quick note:  The non-binary gender options I have put here (genderqueer, gender neutral/neutrois, and bigender) are there because in my experience they are the most common gender identities aside from "male" and "female."  If you are aware that there are other common options among the people you are surveying, you should use those instead (for instance, there are gender categories that are tied to a particular country, culture, subculture, or religion... like hijra, travesti, sworn virgin, two-spirit, and so forth).

6. Assuming everybody is sure of their gender.

I already included this in point #5, but it's worth its own point.  Just as sexual orientation is often described as "questioning" or "unknown," there are plenty of people out there who are not certain about their gender identity yet.

7. Using sarcastiquotes--or "scare quotes"--to describe peoples' identities, names, and parts.

This one is not really about surveys themselves, but about the things you write after the surveys.  It is, however, directly related to a survey I took once, and it's also relevant on the day I write this as there's shit going down about a person's putting the word "genderqueer" in quotes.

The very first survey I ever filled out that was specifically about gender identity and expression was one in which somebody was trying to figure out the most sensitive ways to refer to trans people.  One question in particular was about what is and isn't appropriate to put in quotation marks when referring to a trans person (it was a long time ago and I am paraphrasing):
  • Which of the following sentences do you feel is most correct?
    • "Mary," a trans woman, was pleased with her new vagina.
    • Mary, a trans "woman," was pleased with her new vagina.
    • Mary, a trans woman, was pleased with "her" new vagina. 
    • Mary, a trans woman, was pleased with her new "vagina."
The correct answer here is "none of them, and what the hell is wrong with you that you even felt the need to ask this question?"

There are, of course, times when putting an identity in quotation marks will be acceptable.  In fact, I put the word "genderqueer" in quotes earlier on in this section.  Basically all of these instances are those in which you are talking about the word and not the identity, in other words you are specifying or defining.  A good rule of thumb is if you can put "the word" in front of it (or some variation on that, like "called" or "known as"), then quotes are acceptable.  For example:
  • A person who was assigned male at birth but who has a female gender identity is known as a "trans woman."
  • After he came out as a trans man, he took the name "George" after his late father.
This is a distinction that will require practice before it becomes intuitive.  If you're confused, it's best to try writing in such a way that quotes won't be needed.  It's better to use no quotation marks somewhere it would be appropriate to use them than to put quotation marks on something that doesn't need it.  But before I finish this section, I'd like to explain a bit why using quotation marks to describe trans identities and body parts is so insensitive.  It has to do with a phenomenon called "scare quotes," although I prefer "sarcastiquotes."  This means that when you use quotation marks around a person's identity, name, body, pronoun, or so forth it comes off as saying "This person calls themself/their body that, but we all know the truth, right?  Wink wink nudge nudge..."

So for instance, when you're talking about a trans person's gender confirmation surgery and decide to put quotation marks around "penis" or "vagina," you're denying the validity and realness of that person's body as they experience it.  When you put quotes around a name, a gender identity, or a pronoun, you're denying the realness of their identity.  Keep in mind that this is something that cis people rarely need to worry about.  A cis celebrity can change their name on a whim and still not have the media constantly put their new names in quotation marks.  A cis man who loses his penis stepping on a landmine and has it reconstructed in the exact same way a trans man would have gender confirmation surgery isn't usually going to have people putting "penis" in quotation marks.  So just don't do it.

This is extra-specially important when you're talking about non-binary identities.  Unless you are in the process of defining that identity (for instance: "According to Wikipedia, "genderqueer" is..."), avoid them.  Even other trans people and allies are relatively likely to use sarcastiquotes around non-binary identities as a way of calling their validity into question.  This is inappropriate and something to avoid engaging in.

The conclusion, ending notes, additions and corrections...

Creating reasonably inclusive surveys can feel daunting, and unfortunately due to ever-evolving attitudes and regional differences it'll probably never actually be easy.  It is, however, worthwhile to make a concerted effort in doing so when you are doing research that affects gender-diverse communities (which, by the way, is all communities).  Maybe you'll be able to use one of my suggestions as-is... there's always a possibility it won't work for you, in which case you can of course merely use this list as inspiration or guidelines.

A final note:  When I started writing this, I tried asking people to give input, especially regarding the issue of non-binary gender... and I got very little response.  I'll be editing this document in such case that I've made a mistake or somebody brings up something I didn't think of.  If you want to print this document and distribute it--and you are welcome to do so--please keep all timestamps and the web address intact so that people who read it have the opportunity to view corrections and updates.  Thank you!

** Original document finished February 22, 2014 **    
** Edited March 16, 2014 to add section on quotation marks ** 

* - I should mention here that I have heard tell that "intersex" is beginning to fall out of favor.  It is still the predominating usage as far as I am aware, which is why I use it in this piece.