This survey ran from 19th August until 26th August – one week. There were 2,686 participants after I removed a joke submission and a couple of duplicates, which I think is not too shabby!
There were three things I was looking for:
- How popular pronoun sets are;
- Common word choices and spellings, plural/singular verbs;
- How people tend to gender the sets.
This was probably the most complicated survey I’ve run so far just in terms of structure. I had to place a lot of limitations on what people could enter for each question, so there were fewer “other ________” answers than I usually like to provide.
I’ve been wanting to ask about pronouns more accurately for a long time, because in the annual survey of language for nonbinary people I simply list the most popular pronoun sets and then provide a single box called “other” for people to write in sometimes multiple sets of neopronouns. This means that I have to sort through entries like this:
- xe, xem
- ze, v, per, etc.
… and so on. It is really time-consuming and exhausting to separate and group the pronouns to count them accurately, and that’s because there is not one uniform way of presenting your pronoun set. Even if there were, people would assume I know what they meant by “xe/xem”, they might just be lazy, they might not know the “official” way to use the pronouns and leave words out, and probably there are a hundred other ways in which humans are thankfully not computer programs!
This survey was an attempt to get more helpful and accurate information about pronoun preferences overall, as well as about individual pronoun sets and their use.
The survey began with a compulsory question that asked you to confirm that your gender was in some way not binary M/F.
It then asked for your general pronoun preference:
- Any pronouns are fine
- Avoid all pronouns
- I have some kind of pronoun preference, strong or mild
If you selected “any” or “avoid”, you were sent to the end of the survey, to save you being asked about your pronouns when you had none or didn’t care.
If you selected I have a preference, you were sent to the section that lets you enter your pronouns. You could enter up to five sets. For each pronoun, you were given five sentences to complete, plus a couple of extra questions.
- “Ask Sam if ______ would like to buy a newspaper.” (Eg: he, she, they)
- “Sam bought the newspaper so now it belongs to _____.” (Eg: him, her, them)
- “Sam is reading _____ newspaper.” (Eg: his, her, their)
- “Whose newspaper is it?” “It belongs to Sam, it’s _____.” (Eg: his, hers, theirs)
- “Sam bought it for _____.” (Eg: himself, herself)
There was an explanation of what pronouns are at the top of the page. Each sentence had a few examples underneath it from the he/him, she/her and they/them sets, with one exception – I left “themself/themselves” out of the reflexive question because I wanted to see whether people generally preferred “themself” or “themselves” when talking about one person and I didn’t want to influence their response at all.
There was a question to find out whether the pronoun set uses plural verbs or singular verbs, because I wanted to know whether people usually say, for example, “xe is” or “xe are”. Consider:
- he is a teacher – is is singular, so he/him uses singular verbs.
- they are a teacher – are is plural, so they/them uses plural verbs.
And then there was a question to ask how people generally tend to gender this pronoun set. The choices were masculine, feminine, nonbinary, inclusive, or “only I can use it”.
At the end of each pronoun set participants were given the option to enter another set or skip to the end. If someone entered more than one set, they were given some checkboxes that they could check if it applied to them:
- Pronoun preference tends to change depending on context
- I want people to vary which pronoun they use for me frequently
- I have a consistent favourite pronoun
And then there was a feedback box at the end.
I considered providing some of the more common pronoun sets pre-written for people to select, but decided against it. I wanted to see what people wrote when given empty text boxes. In the annual survey any pronoun that is pre-written will get more people choosing it just because it’s right there. Having it pre-written removes a barrier, and it means people who would not normally consider that pronoun set are prompted to do so; both of these things increases the number of people choosing it.
How popular were the pronoun sets?
When reduced to the first two pronouns in each set it looks something like this:
- they/them: 1,952 – 73%
- he/him: 488 – 18%
- she/her: 321 – 12%
- use any pronouns: 294 – 11%
- avoid pronouns: 131 – 5%
- it/it: 58 – 2%
- xe/xem: 57 – 2%
- ey/em: 52 – 2%
- fae/faer: 25 – 0.9%
- ze/zir: 20 – 0.7% // ze/hir: 20 – 0.7%
(Since this was specifically marketed as a survey about pronouns, I expect that “any” and “avoid” options were underrepresented because potential participants with no preference or who dislike pronouns for themselves would be less likely to open the survey in the first place. But that’s just speculation!)
The top ten looks a little different when you include more than the first two pronouns of a set, because after the first two there tended to be more divergence in words and spellings.
A good illustration is the set “xe/xem,” which more than doubles in popularity when you take into account more variations of that pronoun than just the most popular. I’m not suggesting for a second that all “xe/xem” sets are ultimately the same, but I did want to show that a slight difference in spelling radically changes how your write-in pronoun is counted in this survey and in the annual survey. A pronoun set that seems quite popular outside of statistics could seem less popular in surveys simply because a lot of people use it slightly differently, due to regional differences and personal preference and so on.
To put it another way, here were all the unique sets beginning with “n” that were entered, most of which begin with a “knee” sound:
- ne/nem/nir/nirs/nemself (singular verbs) – 7
- ne/nem/nir/nirs/nirself (singular verbs) – 3
- ne/nir/nim/nir/nirs/nirself (singular verbs) – 1
- ne/nim/nir/nirs/nimself (singular verbs) – 1
- ne/nem/neir/neirs/nemself (singular verbs) – 1
- ne/nis/nis/nis/nimself (singular verbs) – 1
- ne/nym/nyms/nyms/nymself (singular verbs) – 1
- ne/nem/nis/nis/nemself (singular verbs) – 1
- ne/nir/nir/nirs/nirself (singular verbs) – 1
- ne/nim/nir/nis/nimself (singular verbs) – 1
- ni/ne/nem/nir/nirs/nemself (singular verbs) – 1
- nin/nim/nims/nims/ninself (singular verbs) – 1
- ny/nym/nys/nys/nymself (singular verbs) – 1
The only thing they have in common is that they are all used with singular verbs: “Ne is a teacher. Ne has a small class today.”
If you just count the most popular it’s 7, but that’s less than half of people claiming pronouns that start with “ne”, so it’s a very limited picture.
Part of what I like to ask in the annual survey is if we are ever approaching any kind of consensus on the more popular neopronoun sets. Do people generally agree on spelling or even on the words themselves? And after several years, I can confirm that the answer is basically no. For most neopronouns, 50% of people or more are using different words in sets that start with the same word. Some sets appear to sound the same but have very different spellings, such as xe/xir and ze/zir. Speaking anecdotally, some sets are spelled the same way but pronounced differently by different people – some people pronounce “xy/xyr” as “zee/zeer” and some “zye/zyer”, for example.
These will settle over time. If neopronouns become more used and people become more fluent when using new sets, the spelling and pronunciation will converge in the more common sets first. I’m thinking that language coagulates with frequent and earnest use.
There were a lot of people taking part who spoke English as a second language, a lot of typos, and a lot of people using unique tricks to circumvent this:
That’s a validation rule to prevent people writing sentences. In the test surveys, people would sometimes write the pronoun, and then an explanation or an alternative, such as “xe or ze”, or “xe but on my girl days I prefer she”. I love those elaborations but this time I needed extremely simple data so that I could process it using spreadsheet queries!
So a lot of the unique pronoun sets entered probably could have been more accurately grouped with others. There were 367 unique pronoun sets entered altogether taking into account verb preferences, which is a new pronoun set for every 7 people. I enjoyed the ones like “he/they/he/they/his/their/his/theirs/himself/themself (singular verbs)”, but this one is my favourite for pure ingenuity:
- [x]ey_xey_or_yey_or_zey/[x]em_or_xem_or_yem_or_zem/[x]eir_or_xeir_or_yeir_or_zeir/[x]eirs_or_xeirs_or_yeirs_or_zeirs/[x]eirself_or_xeirself_or_yeirself_or_zeirself (plural verbs)
That’s five sets combined into one. Each of these could have been entered separately, so it’s possible that the form was somehow not clear or that it was misinterpreted. I did write in a couple of places that you’d be able to enter up to five pronoun sets, and at the end of each pronoun set you had to select “enter another set” or “skip to the end”, so perhaps this is just one set with many interchangeable options? Perhaps they were trying to communicate that each of these sets is equal with no favourite? Or perhaps the participant was just trying to save time. Are xey sad that xeir sets were counted only once together instead of being grouped individually with matching sets?
Popular verb preferences
Folks expressed an interest in knowing which “they/them” sets are more common. There were 88 unique sets entered starting with “they”, and here’s how many people chose the ones entered ten times or more:
- they/them/their/theirs/themself (plural verbs) – 1,445
- they/them/their/theirs/themselves (plural verbs) – 278
- they/them/their/theirs/theirself (plural verbs) – 60
- they/them/their/theirs/themself (singular verbs) – 22
- they/them/their/theirs/himself (plural verbs) – 16
- they/them/their/theirs/ (plural verbs) – 15
- they/them/their/theirs/them (plural verbs) – 13
- they/them/their/their’s/themself (plural verbs) – 11
“Themself” was five times as popular as “themselves”, and plural verbs (”they are”) were overwhelmingly more common than singular (”they is”).
Pretty much everything else was singular verbs, including sets that sound a little bit like or rhyme with or are based on they/them:
- ey/em/eir/eirs/emself (singular verbs) – 1%
- fae/faer/faer/faers/faerself (singular verbs) – 0.6%
- ae/aer/aer/aers/aerself (singular verbs) – 0.4%
Preferences for multiple pronoun sets
841 people (31% of all participants) entered more than one pronoun set. Here’s how they wanted people to use those pronouns out of an incomplete list of preferences, of which participants could select more than one or none at all:
How do people tend to gender the pronoun sets?
For this step I used just the first two words in each pronoun set, and did some colour-coding with conditional formatting.
The darker the box, the higher the percentage. Any grey “0%” on the right are answers where no one chose that option at all, and the black “0%” are positive numbers that have been rounded down, such as “0.1%”. Percentages are of the people who claimed that pronoun, not of the entire population surveyed.
Here’s a bit of a summary:
- They/them was considered the most inclusive, and the least “nonbinary” of anything other than he/him or she/her.
- He/him was overwhelmingly considered masculine, and she/her was overwhelmingly considered feminine.
- People were more likely to characterise he/him as inclusive or nonbinary, and less likely to characterise it as masculine or feminine, compared to she/her. This might account for people being 6% more likely to be okay with he/him overall? This might also be influenced by the style guide push in recent decades for he/him to be the standard non-gendered pronoun when talking about someone of unknown gender. “Masculine as neutral” is a prevalent stance even among nonbinary people, anecdotally speaking, with feminine still being seen as a deviation from a masculine default, and the statistics here reflect that. It could also be influenced by gender assigned at birth, which I didn’t ask about.
- I note that the results for he/him and she/her are quite different from the annual survey’s usual pattern. Here he/him was 6% more popular than she/her, whereas in the annual survey (which has pre-written options for the pronoun question) the two are more similar, sometimes getting the same percentage when rounded to the nearest whole number. Checking a box is much easier than having to type in a whole set and then answer questions about it, so perhaps people who are okay with she/her don’t like it enough to type it out?
- Of all the pronouns other than he, she and they – only two pronouns were ever considered binary gendered. It/it was gendered masculine by 2% of people who use it (but never feminine), and fae/faer was gendered feminine by 12% of people (but never masculine).
- The pronoun set that was considered most nonbinary-exclusive (used for nonbinary people, and not anyone else) was xe/xem.
- A huge 29% of participants who chose it/it said that it/it could only be used for them. Originally this answer option was intended for people whose neopronouns were based on their names or otherwise exclusive to them, but I think perhaps what people were trying to express here was “this is not a pronoun you can use for just anyone.” When I’ve heard people talking about it/it pronouns, folks tend to be very blunt and up-front about the fact that a lot of people find it/it upsetting, it’s often used to insult or dehumanise visibly trans people, etc. So folks using this pronoun are perhaps warning others to be very careful about how to use this pronoun – but that’s purely speculation.
If we were going to extract a “nonbinary pronoun” from this data (as if he/him was a masculine pronoun and she/her was a feminine pronoun), it would be xe/xem. The most common use of this pronoun was:
- xe/xem/xyr/xyrs/xemself (singular verbs)
“This is Sam. Xe is a teacher. Xyr daughter often gets a ride to school with xem, though sometimes when she’s sick Sam drives to work by xemself. If Jenny leaves her lunch in the car by accident… well, that lunch is xyrs now!“
It’s worth mentioning, though, that this pronoun set as it’s written above was only entered by 22 people – less than 1% of >2,500 nonbinary participants. Almost 1 in 4 participants preferred some form of singular they/them, most of them preferring “themself” rather than “themselves”, and plural verbs.
Tiny personal rant
I’m going to take this opportunity to say that plural verbs are not weird. You already use plural verbs when talking to/about individuals all the time. They do not mean that you’re talking about more than one person, nor is it grammatically incorrect. Regardless of how old they/them is and how many famous writers have used it and whether or not they were right to do so, if you always say “you are” and you never say “you is” even when you are talking to one person, complaining about singular they but not singular you is a double standard.
If you’re reading this then you probably already know that. 😉
I ran this survey to get more accurate information about pronoun preferences and pronoun sets, and I think I succeeded! I find the results interesting and helpful. I hope it’s interesting to others too, and I am really grateful to everyone who took part and helped to promote the survey.
I’m not sure whether I will run a pronouns-only survey like this again, but I may incorporate something like it into the annual survey. It made counting pronoun sets so much easier.
Thank you everyone!
If you like what I do and you want to express some gratitude, please consider checking out my Amazon wishlist!
And if you liked how you were asked about your pronouns, you might be interested to see Starfriends – it’s a truly gender-inclusive social network that me and Andréa have been playing with, that actually collects and uses your neopronouns properly omg.