Gender Census

lauf-aiya-rson said:

some of the posts don’t have the info in the graphics clearly marked (like on the nonbinary spelling post)? but otherwise yeah, i know you spend a lot of time on this survey so i think the way you’ve been doing it is fair

I’m not making things inaccessible because accessibility is too much work, to be clear! The reason for not putting image descriptions on the identity words results post is because they would take up a lot of space and be unnecessary (because the data is in the Google Sheet), not because I don’t want to put the work in. The same is true for the nonbinary spelling post, which has graphs that as far as I can see have all the labels they need, and the statistical results are in a Google Sheet that’s linked in the start of the post.

There are some limitations with Tumblr as a blogging platform, like that tables are just not a thing, you can’t click images to view them in full size, you can’t have alt text, and you can’t upload images and link to them. So I do the best I can with what I’ve got, but if there is a way I can make something more accessible with a not unreasonable amount of work then please do let me know.

andreashettle asked:

Can you also consider writing a description of the graphs and charts, for people who cannot see them? Blind people do have screen reading software that converts text into speech for them (or braille, via a refreshable braille display), but it only works when the computer recognizes it is looking at text and not an image. Thanks.

I did consider this, but in the end I decided against it, because it would essentially mean writing up a table of the results to go with each chart (some of which have 30 or so words and bars, most of which most people will not be interested in), and the table of statistical results is available in the Google Sheet, which is linked in each blog post.

Each blog post does have a summary of the top 5 words and percentages, and in the case of the “words over time” chart there’s a bit of a description of the notable long-term shifts in popularity.

So overall I feel like visually impaired people have access to all the same information.

The survey took place between 1st February and 25th March 2018, and there were 11,241 respondents, of which 1,528 said they were living in the UK.

After the nonbinary vs. non-binary question, I asked, Which of the following best describe(s) in English how you think of yourself?

There were 23 checkbox options, and participants could check as many boxes as they wanted. You can see the spreadsheet of all 11,000+ responses for this question, as well as the graphs in more detail, here on Google Sheets.

The top five were:

• nonbinary - 60.6% (down 5.2%)
  
• trans - 34.8% (up 4.7%)
  
• genderqueer - 30.8% (down 3.5%)
  
• transgender - 27.9% (up 4%)
  
• agender - 26.1% (down 7%)
  

“Transgender” dropped out of the top 5 last year, and this year it’s back - it’s been hovering between 24% and 31% since the survey began in 2013. “Genderqueer” continues its steep decline in use, and I note that “fluid gender/genderfluid” seems to be on a more shallow but steady decline too. “Enby” continues to gain popularity - it’s now at around 25%.

(Please note the lack of 2014 in the above graph!)

This is the first year that “nonbinary” has taken a definite dip, which I think is probably because of my efforts to reduce bias by removing the words “nonbinary” and “genderqueer” from all of the promotional posts online. I’m hoping this is a good sign and the start of a more representative set of results!

Here’s a graph of all the identity words that were chosen or entered by over 1% of participants:

In all there were 1,885 unique write-ins, of which four were words that were entered by over 1% of participants, which means next year the following words will be added to the survey’s checkbox options:

• queer - 2.9%
  
• genderless - 1.1%
  
• demiboy - 1.1%
  
• demigirl - 1.1%
  
• gender non-conforming - 1.1%
  

People seem to want to specify demiboy or demigirl even when demigender is an option, so I think it might be wise to remove demigender from the list and see what happens. It might be that over 1% of participants will write in demigender, in which case I will re-add it to the checkbox list.

Gender non-conforming was a tricky one to count. I had sorted the list of write-ins alphabetically and I noticed that it was being entered many times but being spelled in a lot of different ways, and therefore wasn’t being counted properly. When I searched the list for “conform” I found 23 unique spellings, 15 of which had been entered only once. When they were combined there were 122 participants entering some variation, with “gender non-conforming” being the most popular, so I will be adding that to the next survey.

Fun facts:

• 78% of write-ins were entered only once, and 429 words were written in more than once.
  
• 37% of participants used the write-in box.
  
• When the write-in boxes were used, on average people wrote in 1.4 new terms each.
  
• 93 people, or 0.8%, used all five boxes.
• The mean number of identity words (including the checkboxes) was 4.1 each, and the most common number of identity words chosen/written was 3.
• 90% of people chose 1-7 words.

Thank you for reading! Stay tuned for the third question. Once I’ve blogged each question individually I will write up a longer report for the worldwide results and a separate UK-specific report.

The survey took place between 1st February and 25th March 2018, and there were 11,241 respondents, of which 1,528 said they were living in the UK.

The first question in the survey was: How do you think this word should be spelled? The options were:

• nonbinary
  
• non-binary
  
• non binary
  
• I don’t know/I don’t care
  
• Other [text box]
  

The first three options, bolded, were randomised. I asked this question first because there was also a question about how people identify, and I had to choose a spelling for the checkbox option for nonbinary, and I didn’t want my choice for the survey to sway the results of this question. I also avoided using the word nonbinary in all promotional materials.

You can see the Google Sheets file of the results for just this question here.

~

The results for the UK compared to the worldwide results were very different!

[Please note that the worldwide results include the UK results, and the “Everyone” results include the responses from nonbinary people.]

I was surprised by this, but in hindsight I think perhaps I shouldn’t have been - a few people commented in the “other” box that nonbinary is the American grammatical convention, whereas non-binary is the British one. This would explain the way UK-based respondents preferred non-binary over nonbinary.

I think it’s interesting that people who are nonbinary are more likely to omit the hyphen, even in the UK.

[In the table above, the percentages add up to more than 100% because some people used the “other” text box to express a preference for two of the options.]

1 in 5 respondents have no preference among those three options. Many said that it should be up to the individual nonbinary person to choose how they spell it, or that they would follow the instructions in a style guide, or that they dislike linguistic prescriptivism and all three spellings are valid. Most of the “other” entries were either blank or people spelling words other than nonbinary, such as enby or transgender.

~

This was the one-off question this year, and I think the results are fairly straight-forward! I asked because I see people spelling the word in various different ways and I was curious to know which was most popular. When being considered for inclusion in English dictionaries the most common spelling is usually the one entered in the dictionary, and in this way the “correct” spelling of a word according to English dictionaries can change over time. (There are some exceptions, such as minuscule.)

I guess ultimately I wanted to know if there was a “correct” spelling that I (and style guides) should adhere to, and I am glad to know that when writing about nonbinary issues for UK resources I should include the hyphen.

~

In previous years I’ve included the results of the “extra” questions in the final report. This year I’m choosing to write a separate report, partly because this question doesn’t really relate to the others in the survey, but also because it’s such a big task that I’m hoping breaking it down into several smaller blog posts might help me.

So I think I will try to write a blog post for each question’s results just to keep the ball rolling, and then at the end I’ll tie them all together into two big reports - one for the UK and one for worldwide results, as usual.

This year’s gender census is now closed!

Thank you to everyone who took part and shared the link around, I really appreciate it. :)

And it helped a lot - there are over 11,000 responses. Wow!

WHAT’S NEXT?

My life chaos continues, so I probably won’t be able to process the results for a little while, but I will let you know what’s going on once I get started.