Warning message

The launch of the new ALA Connect, scheduled for April 25, 2018, has been delayed due to technical issues. During final testing some critical issues related to navigation, tagging, and the structure of parent-child relationships were discovered.

The vendor is working to resolve these issues. We apologize for the delay and thank you for your patience.

While we're waiting to get started, please check out the resources and training on the support site. We'll keep you updated, in particular, we will communicate when we have a firm launch date.

If you have questions, please feel free to contact Pam Akins, Community Engagement Specialist, at alaconnectmigration@ala.org.
Go to:
Online Doc
Meeting Request
Linda Crook (non-member)'s picture

Asking the "gender" question

I co-chair the STS Continuing Education Group, and this year we will be conducting our biennial survey.  One of the demographic questions asked in the past regards Gender.  In 2007 it was included as a multiple choice question with only the options male and female.  Is there a best practice for how (or whether) to ask survey takers about gender? Ideas I've come up with are:

We could include other options: e.g. other, transgendered

We could make it an open answer question

We could remove the question from the survey

We would be very grateful for your input! I really cringed at the phrasing of the question in 2007 and would love to have something more inclusive in the 2009 survey.

Previous surveys:

2007 Continuing Education Survey Results
2005 Continuing Education Survey Results
2004 Hot Topics Survey
2003 Continuing Education Survey Results
Michael Steffens's picture

How about "prefer not to answer"?

Johanna Riordan's picture

I like the idea of an open-ended box, but that leaves so much room for answers to vary that collecting statistics becomes pointless and impossible.

How about






-prefer not to answer


I don't like the idea of removing the question altogether.  The best way to challenge the idea of two, and only two, genders is to model a better option.

Meagan Brown's picture

I agree that "Prefer not to Answer" is probably the best response.  Some people who are still questioning may be uncomfortable using the open-ended answer box because either they will feel uncomfortable sharing their identity with people they don't know or because maybe they, themselves, are unsure of how they want to be identified.  I know that this doesn't necessarily help with statistics, but it is likely one of the better options that we can provide to individuals who feel shunned by society or confused about their identity.

Dana Longley's picture

Being trans myself, I like the idea of an open-ended text box - people can put whatever they like in the box (as many have their own definitions/labels of their own gender). I cringe at the "other" designation as it feels to me like a "non normal" category and the "prefer not to answer" smacks of having asecret. But then again, I suppose a text box makes it hard to get data out of it.

I guess the main question to ask is: is the gender data used in some meaningful way? Is useful data/insight gained from including that question? If not, leave it off altogether.


My 3 cents,



Michael Steffens's picture

Dana's got a good point that cuts right to the chase. Does the gender marker provide statistical information that we use? If not, why collect it? I've often wondered why someone would need to know my gender for any number of things. The problem with providing labels, as I see it, is that there are so many possible labels. I have known people who identified as tranny, transman/woman, TG, TS, gender warrior, zie, etc. -- and male, female, other. I do still like "prefer not to answer" because I use that one in a lot of areas of surveys, such as income (which gets a little complicated for a grad student), whether I'm employed or not (do I count internships?) and some other areas, although gender has never been offered as an option. It's something they like to pin down in advertising.

David Vess's picture

Linda, first of all, thank you so much for thinking to touch base with our RT.  ;-)  Secondly, on a personal note,  I'm thrilled to see you say, "I really cringed at the phrasing of the question in 2007."   It makes me happy to know that people like you are out there who _get_ how important a question like this can be.

To stir up some more conversation, I posted a note about your question on the glbtrt list directing people to come over here to Connect to post replies but many have chosen to reply to my email on the list.  I will collect all these replies and post them here later today or tomorrow as the conversation dies down.  Since people decided to email the listserv, I'll remove names and addresses from the messages posted here and just label them email 1, email 2, etc.  I can also just send these emails on to you with emails and names but feel odd about placing those emails out of their original context.

More soon, d

Camden Tadhg (non-member)'s picture

I'm trans and my favorite way I've seen the gender question asked is to have the options Man, Woman (if we're talking gender Male & Female aren't the right words, imo), Trans, and a Prefer Not To Answer choice, with each box having a space directly after it for writing in and the option to "check all that apply".  Don't know if the formatting will come out here, but it looks like this:

__ Man, ____________

__ Trans, _____________

__ Woman, ____________

__ Prefer not to answer, ___________


That way people can get as specific about their gender as they chose and things can be broken down a little more in the statistics by coding the responses, but you can also get some straight-forward numbers out of it.  It's similar to how many surveys are doing the ethnicity question these days, recognizing that a person can have multiple identities and backgrounds or may have identities that don't fit neatly in a category, but that it's still important to gather these numbers.

I do think these numbers are important.  As long as gender and sex based privilege exist, it's important to collect statistic.  I agree there's lots of places the question is required  that are unnecessary...ahem, myspace, ahem,...but we don't live in a "gender-blind" society and our demographic measurements need to reflect in order to catch the ways that gender and society's perceptions about gender influence our lives and our professional opportunities.  There's an option for not answering if you don't want to share, or you're not comfortable with the question or you're not sure of the answer or you don't believe that gender has any bearing on the profession.

Michael Steffens's picture

I'm not even going to go into the whole man/male/masculine-woman/female/feminine thing. Too many writers far more knowledgeable on the nuances than I will ever be have weighed in on that. I think people understand male and female to be basic designators and would generally not quibble over them, unless they're really into semantics and gender theory, as well. More important, to my mind, is whether a gender marker is useful in classifying us professionally at all, and whether that marker is inclusive of those who do not live within a binary world.

While I agree statistics can be important to collect, by basically allowing "prefer not to answer" and "all of the above" and allowing write-ins, the statistics become basically meaningless, IMO. I guess I'm more interested in being inclusive than exact and more interested in feeding valid statistics than engaging in battles of theory. However, I also recognize this would not be the first time I stand outside whatever might be accepted by others -- one reason I do so little within the "community" any more.

Chris Darling (non-member)'s picture

I like your suggestion.  Many people can comfortable place themselves in either Man or Woman categories.  The rest really are on some sort of continuum of transitioning or floating back and forth with different things going on.  Then, with the prefer not to answer.. you can effectively cover all of the ones that feel it's really none of your darn business now is it?! 

Statisticians should be able to work with that a little more clearly I think.   I'm so glad my life is one of holistic happiness and I don't have to spend a bunch of time categorizing little parts of people into numbers that can then be manipulated in the way that would get the desired results.

I fix Windows based computers.  Delaware computer repair pc technician services provided for folks in Hockessin, Wilmington, Bear, and Newark.

David Vess's picture

Hi All

There's been a discussion on the GLBTRT Listserv about this issue.  I've taken the liberty of pasting those emails in this post.  If other messages come into the list, I’ll place then here in this forum so that all users can see the discussion.

I've removed any names and email information in each message for the sake of privacy.  I've placed these messages here in the spirit of open communication/discussion.  If you see your message and don’t want it here just let me know.

The GLBTRT Listserv is a forum for open communication among GLBTRT members and others. In keeping with the ALA open meeting policy, anyone wishing to follow discussions on the list may subscribe and participate.

  • To subscribe, send mail to sympa@ala.org with the subject: subscribe glbtrt-l
  • To unsubscribe, send mail to sympa@ala.org with the subject: unsubscribe glbtrt-l

david s. vess



from    David Vess
to    GLBT-RT List
date    Thu, Aug 20, 2009 at 9:27 AM
subject    Gender Question on our ALA Connect discussion board

Hi All,

The co-chair of the STS Continuing Education Group started a discussion about gender in survey questionnaires in our ALA Connect space.  The person wants to hear from us about what we believe should appear on a questionnaire in addition to "female" and "male."

If you have a moment, go weigh in on this topic at :  <>< />

I'm thinking that having and alphabetical list of the below three would be good but is it good enough?
* Female;
* Male;
* Transgender

Of course what the data will be used for matters the most.  I think gender data is often collected with no real purpose in mind and just is - perhaps because gender is just trained in us by our culture?  If the survey is about gender then I'd vote for the more granular FTM, MTF and others like genderqueer, queer, etc.

Attached is a pdf on logging into ALA Connect.  On the next line is a link to the same information on Connect:
<>< />

David S. Vess
Visiting IMLS Portal Librarian &
Assistant Professor of Library Administration
Grainger Engineering Library Information Center
University of Illinois at Urbana-Champaign
1301 West Springfield Avenue, 158
Urbana, Illinois 61801




Hello all
Just my $.02 worth!  As someone who uses surveys for statistical purposes -- gender is sometimes important if the statistician wants to do a demographic comparison.  For example, females prefer one option more than transgender people or males.  If we want ALA to be more reflective of all peoples, then it is important that all have the opportunity to be heard.  So I would encourage people to, if so moved, to participate in this discussion.



[X person] has an excellent point.  To extend it, if I may, it really depends on the survey design and what the survey is attempting to gather.  Gender as in identity may well be much more complex than even male, female or transgender.  Use of categories on a survey would also require the survey writer to think very carefully about how that information is going to be used, especially if it's being subjected to statistical analyses.  On that point I would defer to someone with a much stronger social sciences research background than I have at this time.

I think the GLBT-RT can make a statement about the nature of gender identity, inclusiveness, etc., but I'm not sure anyone beyond the specific survey writer can really decide what to use on a given survey.

Thank you!



How about a blank?



>The co-chair of the STS Continuing Education Group started a discussion
>about gender in survey questionnaires in our ALA Connect space.  The person
>wants to hear from us about what we
>believe should appear on a questionnaire in addition to "female" and "male." ...
Lines: 18
X-Antivirus: avast! (VPS 090819-0, 08/19/2009), Outbound message
X-Antivirus-Status: Clean


Thanks for posting this url. We should weigh in.  But I suspect there will not be total agreement among us.

In my view, "male" and "female" is too limited.  Although they are few, there are intersexed people among us.  It seems to me that most transgendered among us would be comfortable giving the gender with which they currently identify, as opposed to their birth gender.

I would suggest simply adding "other" as a category, rather than  a laundry list of "intersex", "MTF transgender", "FTM transgender", as well as stages of transition.  Some, might consider being formerly of another gender personal information not for general disclosure.



Genderqueer?  Queer?  I must be out of touch because I've never heard of the former term, and the latter has always struck me as so vague (and not necessarily about gender) that it's not a helpful description.

If we do offer a lengthy list of terms we recommend, could we establish that they are in common use, whether that be in a particular community or even (here's the librarian in me) a controlled vocabulary?

Ultimately, I like [X person]'s idea of a blank.  Maybe we don't need to recommend a comprehensive list but instead suggest that the question allow the respondent to write a label or description that the respondent feels fits best.



I am thinking that we might also include the category "intergender," also sometimes called "gender queer."  So perhaps the list might look like this:


On the other hand, as another poster has pointed out, there are still more categories that could be added, and a simple "other" may do.



Hi all,

I think in cases like this, we will never be able to come up with a list of inclusive terms. People have varying degrees of openness about their gender, and I think the best solution would be to simply ask "Gender:_____________" and let the categories create themselves.

We could try and come up with them all in a list, but the point of doing that would be to make sure that we are not limiting the choices to prescribed gender categories, no matter how exhaustive a list might seem. So I think the best way to accomplish that would be to let people tell us how they identify.

I'm against putting male, female, and other, because the other category serves to kind of "un-gender" a person. It seems exclusive to me.




I also like the idea of choices plus a blank to allow for self-identification.

Often, a trans person's identity may shift over time (example: from "genderqueer" to "Transgender" to "FTM" to "Male"). A blank would allow for this and other categories.




I agree with [X person], this is a good and simple way to avoid the use of "Other" (which may feel offensive), and people can just define themselves without anyone coming up with a "correct" list.




So often, I think, there is no need for this question.



Genderqueer is absolutely in common use, largely among younger people.



I do like this idea, although perhaps it should say "Gender Identity______." There are many people who don't categorize themselves in binary gender terms, who might answer "genderqueer," "fluid," or in other terms.

I do, however, wonder why we are asking as a general question. I realize our bylaws call for co-chairs of differing gender identities, and that we try for gender variation/balance in the make-up of our committees, and the Steering Committee. However, I think using a binary gender system to do so no longer serves us, given the changing realm of gender identity. I think it makes more sense to ask people their gender identity when we are trying to put together committees, and we can then try to find some range that represents our membership.



I agree that "other" can be offensive, and it should not be used.  Just a blank box to fill in one's preferred term would be better.

If there is indeed a real reason to gather this info in a survey, it should be included, and enough categories and a blank box should be provided.  If we are challenging the gender binary we need to model good solutions instead of just deleting the question.



Keep it coming gang.

I didn't think we would come up with a definitive answer or list for them.  It will be up to Linda's group to decide what to make of all this.

I personally am leaning towards [X person], [X person], [X person] and other's solution.  The most neutral so far seems to be, "Gender OR Gender Identity: ________"   I realize using this on a questionnaire requires some work on the data analysis side to normalize responses but this method is ideal for a lot of reasons.  Also, I agree that "Other________" in a list of "F, M, Other" just feels pejorative to me.

Coming from a research background with human subjects and questionnaire instrument design I feel comfy saying that a, "choose one item from this list," method is always very tricky.  These kinds of lists should be formed out of data collected in a preliminary research phrase where investigators learn a great deal about the people being studied - what matters to them, etc.  That kind of preliminary work can still easily miss things, such as people who identify outside of a binary gender construct.

I'll send all these emails on to Linda in a day or so as conversation dies down with the proviso that she keep them to herself and not send them on to others.  I'll also post all emails about this to the Connect space Linda created with your names and email addresses removed.  They will be labeled as 'email 1,' 'email 2,' etc. in chronological order.  I'll do this so that others, not on our listserv but on Connect, might stumble upon this discussion and jump in.

To [X person], about your statement, "If we are challenging the gender binary we need to model good solutions instead of just deleting the question."  SNAP  ;-)  Awesome answer!  You've made me rethink ever throwing out the question.  Though I still feel that people ask this question out of habit and don't really intend to use it in their analyses.  I think any groups in ALA need to be asking about and using gender as well as race, ethnicity, etc in any data collection projects as well as assessing those variables in results.  Assessment should not only be about who _is_ at the table with us but should also be about who _isn't_ at the table with us and asking, "why is that?"  [X person] said it perfectly, "If we want ALA to be more reflective of all peoples, then it is important that all have the opportunity to be heard."

David S. Vess



I personally like either a blank box or gender identity:



We are doing an lgbtiqa needs assessment survey at my university, and they've split out sex and gender as two different categories--the gender categories being genderqueer, man, transgender, woman, and a blank box--which works pretty well. Another method I personally dig is offering genderqueer, male, transgender, and female, but allowing people to check multiple boxes and still having a fill-in-the-blank spot.
[X person], who worries that he thinks about these things too much



>I agree that "other" can be offensive, and it should not be used.  Just
>a blank box to fill in one's preferred term would be better.

Having "male" and "female" to tick, plus a blank box to fill in, would be fine.  But just a blank for all to fill in would not work for automated analysis.  I suspect that for must automated surveys, those using the box would be treated as "other", with no further analysis, so there would be no effect on the survey, just avoding people calleing themselves "other".

Jess Briggs (non-member)'s picture

I've found this thread after wrestling with the question myself for work. I know it's very very old but thought it might be worth adding something as I quite like the idea. My social geography background is pulling me one way and the marketing job the other. I like the BLANK rather than Other and I'll be using that myself I think. But also, perhaps the question itself can be altered to help reflect the sense of a continuum. What if we said "Which gender identity do you most identify with?"... It's a question I bet would be found in sexuality research that would assume that the individual were on a continuum of sorts, but why can't we assume that in 'general' research? In time, perhaps there would even be a Likert scale?  But then, I suppose there are people whose gender identity is utterly seperate from a masculine-feminine dichotomy anyway so the scale may not work.

Anyway, just throwing that in there. Basically, we can use question construction to make assumptions of gender (as opposed to sex, a whole other ballgame of course), less absolute and discrete.

Anyway, not really expecting a reply; more thinking out loud ;)

Rachel Morton's picture

I have done a lot of reading on gender identity, and there are two categories that I see mentioned in my reading that I don't think anybody has brought up: "agender" and "gender nonconforming". If I was creating such a question on a survey I would include those.

Also, this comic might give you some ideas of how to create a more inclusive survey:


Elizabeth Bruch's picture

Loved the comic. Agree on blank only or including agender and gender nonconforming and many other points in earlier posts. I didn't see anyone mention including cisgender. So if there were a list of options where you would choose as many as you want, you could pick cisgender and female or transgender and female or whatever works for you. I think this would also help it not seem like cisgender is a default. 

Amy Brunvand's picture

As was mentioned in other comments, to format a survey question you need to decide why you are asking in the first place.  

Unless the survey is actually about gender identity it seems like asking people for detailed information about gender identity is simply not that useful.  Still, it would be nice to have a standard pattern that could be used on any survey form. 

Usually the  purpose of asking about gender  is for some kind of statistical correlation.  For that purpose, all of the non-binary answers are probably going to get lumped together for the final data analysis regardless of what is on the survey form.    

In order to both respect the gender identity of survey fillers and  make the life of assessment librarians a little easier, How about asking the question like this: 

Gender Identity: