The blog for the annual survey of humans worldwide whose genders or lack thereof are not fully described by the gender binary.
Gender Census homepage
Mastodon: @[email protected]
There's a survey on Australian LGBTIQA+ youth experiences going around. Figured I'd share it here! (Please share with anyone you know 14-21 and lives in Aus).
"For more than 20 years, the In Our Words survey has shed light on LGBTIQA+ young people’s health and wellbeing in Australia. Last time, over 6000 LGBTIQA+ young people shared their stories, leading to real service and policy changes for the community.
In Our Words is now open and we want to hear from you!
If you’re:
- 14-21 years old
- Lesbian, gay, bi+, trans, intersex, queer, ace/aro, or anywhere on the rainbow
Take part in this anonymous survey to help shape the future of LGBTIQA+ young people’s health and wellbeing in Australia.
👉 For more info visit: https://latrobearcshs.qualtrics.com/jfe/form/SV_8kN2NovmmlzaHhc Your voice, real change.
Ethics Refs: La Trobe - HEC25389, AIATSIS - REC-0603, WAAHEC - HREC1476"
Hello, finally got around to reading this year's report (thanks as always!) and wanted to send a brief correction to some of the information about Spivak pronouns.
Specifically, Michael Spivak's book "The Joy of Tex" was first published in 1982, though it was reprinted in 1990, and Michael did not coin the e/em/eir pronoun set, though his book did help popularize it. You can see some sources for this at the Spivak pronoun Wiktionary entry.
Additionally, I'm curious where you heard that the set was generally capitalized as E/Em/Eir? Michael capitalized them in the Joy of Tex, but as far as I can tell this pattern hasn't stuck? Most hits I'm seeing for the set on Twitter (sorry) with the following search
"e" "eir" lang:en -grok
are lower case, such as this one with a photo of a published book.
Please take care!
Ahh, thank you for pointing this out! I've now fixed this in the report, and pointed the link to your more reliable source page.
For capitalisation, I think this might require a dedicated mini-survey so that I know how to capitalise it in future checkbox lists. Perhaps I will just do a few social media polls first? Because if it's 100% capitalised or 100% uncapitalised on social media then I can maybe safely assume it'd be way more than 50% in the general population.
I don't always get to it right away due to busy life things, but I always appreciate people sending in corrections, because I want my reports to be correct and reliable resources.
Who? Everyone whose gender doesn’t tidily fit into the female/male binary. (Oceania participants isolated)
What? An online survey asking participants how they describe themselves and how they would like other people to refer to them.
When? July to August 2025.
Oceania Participants: 2,419
Worldwide Summary: https://www.gendercensus.com/results/2025-worldwide-summary/
Full Oceania Report: [GC2025] Analysis of Oceania’s results from the Gender Census
Q1. Identity words
The Top 5 were:
nonbinary: 58.8% (up 1.2%pt.)
queer: 53.5% (up 1.9%pt.)
trans: 44.5% (up 1.7%pt.)
transgender: 40.1% (up 3.2%pt.)
a person / human / [my name] / "I'm just me": 36.3% (down 1.0%pt.)
Q2. Titles
Titles (with name)
Here’s the top 5:
No title at all: 38.2% (up 1.2%pt.)
Mx: 19.7% (down 3.1%pt.)
Mr: 10.8% (up 0.3%pt.)
Non-gendered prof/acad.: 6.4% (up 0.5%pt.)
Ms: 6.0% (down 0.7%pt.)
Titles (without name)
The top 5:
No title at all: 70.0% (up 3.4%pt.)
Friend: 31.3% (up 1.8%pt.)
Sir: 28.9% (up 0.7%pt.)
Comrade: 20.3% (down 0.4%pt.)
Mx: 18.3% (down 2.0%pt.)
Q3. Pronouns
The top five pronouns (or lack thereof) are:
They - they/them/their/theirs/themself: 76.9% (up 1.1%pt.)
He - he/him/his/his/himself: 39.6% (down 0.9%pt.)
She - she/her/her/hers/herself: 32.9% (down 4.3%pt.)
It - it/it/its/its/itself: 21.1% (down 0.3%pt.)
Avoid pronouns / use name as pronoun: 12.1% (down 0.9%pt.)
SmartSurvey have changed their plans! Next year it's going to be EIGHT TIMES as expensive.
Last month it was £70 per month for two months before VAT and non-profit discount (£151.20 per annual survey), and that gave me unlimited responses.
Now, that plan has a 30,000 response limit. The next plan up is Business Teams, which can only be billed annually, and it's £1,080 per year before VAT (£1,296 after VAT), and it has a 75,000 response limit.
Remember when a bunch of people were unable to visit gendercensus.com or get to the survey via survey.gendercensus.com a few weeks back because their malware detectors were blocking the domain?
What happens when you click the above gendercensus.com link now?
It works now; during the survey I was malware-blocked
It works now; during the survey it also worked and I wasn't blocked
I get malware-blocked now; during the survey I was also malware-blocked
I get malware-blocked now; during the survey it worked and I wasn't blocked
It works now; I don't know/remember if it worked before
I get malware-blocked now; I don't know/remember if it worked before
Remember when a bunch of people were unable to visit gendercensus.com or get to the survey via survey.gendercensus.com a few weeks back because their malware detectors were blocking the domain?
What happens when you click the above gendercensus.com link now?
It works now; during the survey I was malware-blocked
It works now; during the survey it also worked and I wasn't blocked
I get malware-blocked now; during the survey I was also malware-blocked
I get malware-blocked now; during the survey it worked and I wasn't blocked
It works now; I don't know/remember if it worked before
I get malware-blocked now; I don't know/remember if it worked before
Remember when a bunch of people were unable to visit gendercensus.com or get to the survey via survey.gendercensus.com a few weeks back because their malware detectors were blocking the domain?
What happens when you click the above gendercensus.com link now?
It works now; during the survey I was malware-blocked
It works now; during the survey it also worked and I wasn't blocked
I get malware-blocked now; during the survey I was also malware-blocked
I get malware-blocked now; during the survey it worked and I wasn't blocked
It works now; I don't know/remember if it worked before
I get malware-blocked now; I don't know/remember if it worked before
There are two new additions to the Gender Census shop today - some grey sparkly handwarmers, and a nonbinary star fidget. (Plus, I added a pair of nonbinary handwarmers yesterday so those are back in stock now too.)
There are two new additions to the Gender Census shop today - some grey sparkly handwarmers, and a nonbinary star fidget. (Plus, I added a pair of nonbinary handwarmers yesterday so those are back in stock now too.)
SmartSurvey have changed their plans! Next year it's going to be EIGHT TIMES as expensive.
Last month it was £70 per month for two months before VAT and non-profit discount (£151.20 per annual survey), and that gave me unlimited responses.
Now, that plan has a 30,000 response limit. The next plan up is Business Teams, which can only be billed annually, and it's £1,080 per year before VAT (£1,296 after VAT), and it has a 75,000 response limit.
I'm putting this on my personal blog (as opposed to the @gendercensus Tumblr or website) because it's not at all scientifically robust. My goal was to collect some data that, while not fully reliable, might at least point me in the right direction - and it did meet my needs.
I wanted to find out more about three things:
Are intersex people more likely than non-intersex people to be nonbinary (or otherwise not-just-men-or-women)?
Do plural/multiple people and systems widely consider being plural/multiple/a system to be a gender identity in itself?
Do otherkin people widely consider being otherkin to be a gender identity in itself?
Context
The wording of the gender identity question in the Gender Census very carefully avoids saying the word gender, to make sure that people who don't experience gender at all aren't immediately short-circuited into leaving the survey. It was guided by the EHRC's "Technical note: Measuring Gender Identity" (2012).
Which of the following best describe(s) in English how you think of yourself?
This intentionally open wording means that I can absolutely understand people responding in the textboxes with terms that refer to aspects of their identity that aren't gender-related, like white or Buddhist or adult. Even if 103% of participants said they were Jedi, I wouldn't add it to the checkbox list because it's not typically a term reflecting a gender identity/experience/presentation. The exception would be if it was clear from their textbox entries somehow that these words were gender-related for a significant proportion of respondents.
Checkboxes are based on popularity within age groups. A few years ago in the Gender Census, enough people were typing in words like lesbian and dyke and queer, and specifying that they were intended as gender identities, that I started to consider them fair game for the checkbox list of identities.
Alongside this, it has been common for participants to tell me to add intersex and plural and (to a lesser extent) otherkin to the identities checkbox list because being these things strongly affects their experience of their gender. This is fine, and my standard response would be to wait until these terms get entered frequently enough and then consider it. However, some went further and said that the majority of intersex/plural/otherkin people are nonbinary or that these are inherently gendered/gendering experiences, and I'm somehow missing that information and omitting these terms from the checkbox list inaccurately.
It's pretty exhausting to respond to these people individually, for three reasons.
I have an FAQ on the Gender Census website for this, but these people often insist that theirs is a special case because their identity is marginalised. (There is no precedent for me adding terms to the checkbox list due to their marginalisation so I don't know why people think that argument might work.)
I ask for research to back up their claims. No one has ever provided any, sometimes saying that there is none, but they somehow know it's universally true.
Even if somehow they were right and 100% of the people identifying this way considered it a gender, the percentages are still not high enough to get these terms into the checkbox list.
So what I wanted was something, anything, that is more than zero information about the gender identities of the people in these groups, as independently of the nonbinary community as possible. I strongly suspect that there is no resource I could cite that would change their minds, but I will feel better knowing I did my due diligence in looking into it and I've got something to back up my instincts. Also, I like the idea of these people finding out that their experience isn't universal, you know?
So, here's what I did and what I learned.
Intersex
I did a survey on Google Forms. I promoted it a lot on Tumblr, Mastodon (and by extension the Fediverse), and Reddit. I also worked out quite early on that I would need to bring in as many people outside of my social media circle as possible, so I decided to try out SurveySwap.io. My goal was to find out whether or not intersex people are more likely to identify outside of the gender binary than people who are not intersex. I accepted in advance that the results would be biased, and at best they would indicate whether my assumptions were totally off and nothing more.
A couple of weeks ago someone on Instagram sent me 500 words of DM telling me that I should add intersex to the checkbox list, and when I asked for any research or resources to back up their assertion that intersex people were more likely to identify outside of the binary they said, "I am asking YOU to provide this resource." [More on how moral entitlement is driving me slowly up the wall here.] My response was that it's not even that I don't want to provide this resource - I specifically cannot, it must be someone other than me who gathers this data, because the audience I've snowballed over the years are all nonbinary, so any sample I produce will be way more likely than the general population to be nonbinary, to an absurd degree. The bias would lead to the inevitable result that all intersex, otherkin and plural people are nonbinary, because all my followers are nonbinary.
The survey that I ran to find out about intersex people was the absolute best I could possibly do. I didn't mention it on any of the Gender Census social media at all, I promoted it entirely through my personal accounts. I also was careful to post it in places that would more closely reflect and represent the general population gender-wise, such as Reddit and SurveySwap.io. I put the question about whether participants were intersex right at the end, so it couldn't prime them for any gender questions, and provided some clear help text so that people would know what I meant by "intersex". I put a bunch of totally irrelevant questions in there (like location and age) to throw people off regarding the goal of the survey, and did some survey promotion based on those things. I intentionally kept the goal of the survey secret even for people who had completed the survey.
Just in case I would need it, I included a question on whether or not participants had done or heard of the Gender Census. I hoped that this would allow me to filter for a sample that was at least a little more representative of the general population. I wasn't worried about whether this sample would be skewed in a binary direction, because I don't expect that the Gender Census is so famous that excluding Gender Census participants would exclude a significant proportion of nonbinary people.
The survey ran from 16th September to 8th October 2025, and received 1,650 usable responses.
Participation was much slower than I am used to, probably for two reasons. The first is that my personal accounts followership is much smaller than the Gender Census'. The second is that a very generic demographic survey with no exciting niche and a secret goal is so boring and no one cares. (I would like to take this opportunity to thank everyone who participated even though it was so extremely dull and there was zero payoff, I really appreciate it.)
Anyway, back to the story. Within hours of opening the survey, a truly absurdly high percentage of participants were nonbinary, and that was very predictable and also very ugggghhhhhhh. So I kept up the promotion, and even posted a story on my personal Instagram account that I hadn't used since 2022 (private, 54 followers). At one point I fully begged for more participants who weren't nonbinary, and I'm not proud of it. (After that I switched to begging for participants who hadn't heard of the Gender Census, which felt a bit less awful.)
My initial goal had been 1,000 participants, but it soon became clear that even with a sample that big the vast majority would be nonbinary, so I switched the goal to 1,000 participants who hadn't heard of (or weren't sure if they had participated in) the Gender Census, or who didn't answer the Gender Census question at all. For the sake of our reading comprehension, we will call this subset "the Gender Normies".
After all that, what we mostly have is evidence that it's really hard to get an unbiased sample. Of all participants, 54% were nonbinary (the top gender). When limited to Gender Normies (954), 37% were nonbinary (in second place by one (1) respondent). These numbers are so obviously ridiculously high, but just in case you live in a nonbinary utopia and don't have a TV, it's estimated that <0.1% to 0.6% of the general population are any kind of transgender.
Of the subset of the Gender Normies, 1.9% (18) ticked the "I am intersex" box at the end. That's actually pretty much in line with estimates of how many people are intersex in the general population. Here's the intersex Gender Normie gender breakdown:
Man or boy: 8 (44.4%)
Woman or girl: 5 (27.8%)
Another option: 5 (27.8%)
[blank]: 1 (5.6%)
(When this is expanded to include all intersex participants (45), including those who had heard of the Gender Census, the nonbinary cohort rises from 28% of intersex people to 69% of intersex people. For goodness' sake.)
So the first thing I notice is that nonbinary people are not the most common gender among intersex people when I reduce influential variables as much as possible - and that's even in a sample that's so biased that, when it's as unbiased as I can get it, is 37% (!?!) nonbinary.
The next thing I notice is that of Gender Normies, 37% of the non-intersex people identified outside of the gender binary, whereas only 28% of the intersex people did. This indicates that even in my ridiculously nonbinary sample, intersex people are less likely to identify as nonbinary than non-intersex people and the general population.
This is not reliable info, because 18 intersex people is not a lot. And if the percentages had been closer or if the intersex nonbinary percentage had been higher than the non-intersex nonbinary percentage, I would have taken that with a pinch of salt and accepted that maybe intersex people might still be more likely to be nonbinary in a general population sample. But for it to swing that far in the other direction, with this sample? I am now even more sceptical of the "intersex people are more likely to be nonbinary" claim than I was to begin with.
This is not to say that I actually think that intersex people are less likely to be nonbinary. I cannot stress strongly enough that the data here is not scientifically robust, and the research needs replicating by someone who hasn't been collecting nonbinary survey-lovers for 13+ years. It's just that I tried so hard to verify that intersex people are more likely to be nonbinary, gathered a disproportionately nonbinary sample, and found the opposite. D'you see what I mean?
The next thing I'm curious about is whether intersex people are typing intersex-related words into the gender identity question. The wording of this question was the same very open style as in the Gender Census:
Which of the following best describes in English how you think of yourself?
The answer options were a lot more restricted than the Gender Census question at the first step, because it's open to all:
Always, solely and completely a man or boy
Always, solely and completely a woman or girl
Another option not described here
When people chose another option not described here, they were led to a second page inviting them to select textboxes of the gender identity terms they identify with, which included a bunch of binary and less binary terms, plus questioning or unknown, none / I do not describe my gender, and an "other" textbox at the end.
Google Forms doesn't allow you to have a list of textboxes, so this survey's "other" textbox situation was less elegant - a single, small textbox, where participants were advised to list their custom gender terms separated by commas. Participants could (and did) write any gender terms they identified with that weren't already checkboxes.
Of the 45 intersex people who took part in the survey, 31 chose another option not described here for their gender identity. Of them, 2 people typed "intersex" and a third person typed "intergender" (combined, 6.7% of all intersex participants). Of the 18 intersex Gender Normies, none entered either of these words. I can't find any other words that contain or are obviously intersex-specific in any intersex people's textbox identity terms. I just want to emphasise again: even in a gender identity question so delicately worded that people often enter identities that are not gender-related, only 3 people entered an intersex-related term and they were all prior Gender Census participants.
It really, really seems like intersex people aren't more likely to be nonbinary, and also that only a tiny minority of intersex people identify their gender as being intersex or similar. I feel that adding intersex as a checkbox option to the gender identity question in the Gender Census is not justified at the moment.
If anyone disagrees with me and has anything to back up their opinions, I would really love to see some actually scientific resources with bigger samples that are more representative of the general population, ideally peer-reviewed! Please do send them to me. My Gender Census email address is hello AT gendercensus DOT com.
If you didn't like how scientific this section was, you're going to love the next two sections.
Plural/multiple
This was a Tumblr poll and a Mastodon poll. The question on both was, "This is a poll for plural/multiple/system people only! Being plural/multiple/a system is…" with the following answer options:
Neurodivergence (analogous to being trans, autistic, ADHD, etc.)
Gender (analogous to being male, female, nonbinary, etc.)
A third category of its own
Neither/other (replies welcome)
I'm not plural/multiple/a system
(Thus ensued a debate in the replies about whether being trans is neurodivergent, and the jury's still out. However, I think that the results of the poll are most likely unaffected.)
This graph uses responses only from plural/multiple/system respondents. You can see my calculations on this Google Sheet.
So yeah, I feel confident that the vast majority of plural/multiple people don't feel that it's a gender in and of itself. Participants in the Gender Census are going to have to mostly explicitly state that they are plural/multiple in relation to gender for me to consider adding it.
Otherkin/therian
This was a Tumblr poll and a Mastodon poll. The question on both was, "This is a poll for otherkin people only! Being otherkin is…" with the following answer options:
Neurodivergence (analogous to being trans, autistic, ADHD, etc.)
Gender (analogous to being male, female, nonbinary, etc.)
A third category of its own
Neither/other (replies welcome)
I'm not otherkin
This graph uses responses only from otherkin respondents. You can see my calculations on this Google Sheet.
It's a little more ambiguous, but gender is still the least popular option on the list. So yeah, otherkin/therian people don't seem to widely consider it a gender identity in and of itself, overall.
Conclusion
I have yet to see any evidence that intersex is a term that is widely applied to gender identity
I have yet to see any evidence that otherkin people widely consider otherkin to be a gender in and of itself
I have yet to see any evidence that plural/multiple people widely consider plural/multiple to be a gender (or genders) in and of itself
Even if any of them were widely considered to be genders, they are not popular enough to be on the checkbox list
Even if any of them were widely considered to be genders, I haven't been presented with any compelling reason to bypass the usual checkbox selection system for these terms and add them to the checkbox list in spite of them not being popular enough
I have decided not to add these terms to the gender identity checkbox list of the Gender Census until both I see evidence that they are widely considered gender identities and they are popular enough to oust something else from the list.
I will not reconsider this decision unless someone can present me with compelling supporting evidence that might shake this decision, such as a minimally biased quantitative survey report, with published info about its design and a larger sample size, showing that a clear majority of participants considers the term to reflect a gender identity. It doesn't have to be peer-reviewed, but "most of my friends are [identity adjective]" isn't going to cut it.
Resources
Intersex survey responses on Google Sheets
Plural/multiple and otherkin spreadsheet on Google Sheets
I have yet to see any evidence that intersex is a term that is widely applied to gender identity
I have yet to see any evidence that otherkin people widely consider otherkin to be a gender in and of itself
I have yet to see any evidence that plural/multiple people widely consider plural/multiple to be a gender (or genders) in and of itself
Even if any of them were widely considered to be genders, they are not popular enough to be on the checkbox list
Even if any of them were widely considered to be genders, I haven't been presented with any compelling reason to bypass the usual checkbox selection system for these terms and add them to the checkbox list in spite of them not being popular enough
I have decided not to add these terms to the gender identity checkbox list of the Gender Census until both I see evidence that they are widely considered gender identities and they are popular enough to oust something else from the list.
I will not reconsider this decision unless someone can present me with compelling supporting evidence that might shake this decision, such as a minimally biased quantitative survey report, with published info about its design and a larger sample size, showing that a clear majority of participants considers the term to reflect a gender identity. It doesn't have to be peer-reviewed, but "most of my friends are [identity adjective]" isn't going to cut it.
hey, you mention in the summary that Magister was invented about a year ago, but its in at least 3 years of the data and multiple people mention it being a historical word with etymology back to the Latin Magister
as well as Magis coming from Latin for more which has also been present in a few years back
though the tumblr post does acknowledge this.
pronunciation was suggested to be mage
but personally I'd expect people reading Mg to say Mag or Maj
anyways, thanks as always for this incredible project
Thank you, I have updated the report to not say that that one viral post writer created Mg - just that its popularity is probably because that post went viral.
crunched some numbers from the 2025 @gendercensus data for Finland. Identity words, pronouns (from checkboxes), and number of pronouns. This is 354 responses out of the 43,096 total, so it's uhhhhh. a pretty small set. I'm comparing percentage of finns to percentage of the whole set (finland included), but the results are hardly reliable w the sample sizes being so different LMAO
Still. I was interested in the identity words and pronouns here, because Finland is NOT an english speaking country, but most finns speak english well, and we are disproportionally represented on the internet, based on population. and in this census, also. Finnish pronouns are also gender neutral. So, it's neat to see how finns describe their gender in a second language.
Nonbinary is also the most popular choice for finns in this census.
A larger % of finnish participants described their gender as "questioning/unknown", "a person/just me", and "none" compared to the % for all the responses. (Could be a trait for finns in general, or related to participants not having a translated identity word.)
For pronouns, "They/them", "It/its" and "avoid pronouns" were chosen by a larger percentage of finns, compared to all the responses. She and he were less popular for finns, as well as most of the neopronouns. (harder to learn for a second language?) The equivalent of "it/its" is a common pronoun in casual spoken finnish :]
Finnish participants were slightly likelier to choose under 3 pronoun sets than all participants, similar to the graph shape of participants under and over 30 yrs old.
i did these pretty quick and didnt check too much so there could be mistakes. ok thats all byeee <3
Does anyone know of an ad-sharing system that's a bit like the old skool Project Wonderful or the newer, still active ComidAd network?
I would really like to share the nerdiness of the Gender Census with the wider nerdy internet with some kind of benevolent ad system, but ComicAd is limited to creative and comic projects!
