#statistics
- digital Intro to Statistics notes i took today <3
- how is everyone doing with online classes?
If you see that Amazon is a great place to work, you know it was an article paid for by Amazon.
I am still learning how to write polls and survey questions that bias the answers as little as possible. I probably get it wrong a lot. But I like to think that I sometimes do a little better than I used to. (This is in large part down to feedback and criticism, and that’s why I am always careful to include a feedback box at the end of the annual survey and any other non-Twitter surveys I run. I’m so grateful for any feedback!)
And this evening, I found a poll that was so… nonsensical, that I had to ramble about it on Twitter. Maybe if I staple it all together into a coherent blog post it might be a useful reference.
I found this on strawpoll.me after voting in a poll about Pokémon Go. It was called “promoted”, presumably meaning someone has paid to have this poll shown when people vote in other unrelated polls.
[Image description: A poll question. “Are you comfortable with a Transgender person in your bathroom?” Followed by three answers, “yes (40%)”, “no (38%)”, “indifferent (23%)”. It notes that there are 117,072 total votes.]
At first glance it looks like a fairly straightforward poll about a current topic - the issue of trans people and which public toilets we should be using. Should we use the gendered toilet that matches our gender, our presentation, or our birth certificate? And the results are close - only 2% difference between the trans supporters and the transphobes.
Let’s break this down into all the levels of flawed.
~
IT’S AMBIGUOUSLY WORDED
It’s easy to assume that it means “are you comfortable with a transgender person whose gender matches yours using the same gendered public toilet as you?”
This is not what it says.
It doesn’t specify whether the trans person who’s sharing a toilet with you has a gender that matches your own, and it doesn’t specify that the bathroom is gendered or public.
It’sactually asking whether trans people should be able to use toilets. Because it can be interpreted in so many different ways, the results are meaningless. Here are just a few of the interpretations:
- “You’re a trans person. Are you comfortable using the same toilet as other trans people?”
- “You’re a person. Are you comfortable using the same gendered public toilet as a trans person whose assigned gender on their birth certificate matches yours, even though they may pass as a different gender to your own (eg: you’re a man and there’s a passing trans woman in your toilet)?”
- “You’re a person. Are you comfortable with any trans people in the same public restroom as you?”
- “You’re a person. Are you comfortable with your trans friends using your toilet in your home?”
- “You’re trans. Are you comfortable in your own bathroom?”
- “You’re a person. Are you scared of getting trans germs from sitting on a toilet seat that a trans person has sat on?”
~
IT USES LEADING LANGUAGE
The question is decidedly not neutral. It puts the judgement in the question, establishing it in people’s minds before they read the answer options.
A better way might be to ask, “how do you feel about [issue]?” And the answer options can contain the judgements: “Comfortable, uncomfortable, don’t know.”
It also uses the word “your”, evoking a feeling of territory. It doesn’t imply that you and the trans people have equal right to the toilet, by saying that you share it. It says “this is your toilet, you have a right to it, and someone is allowing a trans person to use it too.” I can see that this could sway someone who’s uncomfortable around trans people into a stronger feeling of being violated by a trans person entering their space.
~
IT ASSUMES THE TARGET IS CIS
I’m trans. I use a bathroom on occasion. I’m comfortable with that.
But I don’t think they’re asking me. I think they’re asking cis people. I think it hasn’t even occurred to the writer that trans people might want to have a voice here. It reflects so clearly that the people debating this issue are not considering the humanity of trans people at all.
~
IT SPLITS THE “YES” VOTE
The options are:
- Yes, I’m comfortable (40%)
- No, I’m not comfortable (38%)
- I’m indifferent - I don’t mind (23%)
I think the writer doesn’t even realise what they’ve done. They’ve tried to present a neutral option, a non-answer, with “indifferent”. What they don’t understand is that these options are not positive-negative-neutral, as they intended. Being “comfortable” is the neutral option. If you don’t mind, you’re not uncomfortable, therefore you are comfortable.
To split the yes vote is to hugely skew the appearance of the results. If you acknowledge that comfortable and indifferent are the same, people who are comfortable/neutral are 62% - nearly two thirds. This is a HUGE difference.
A true positive-negative-neutral set would include comfortable/neutral, uncomfortable/negative, and a positive: being actively happy about the presence of trans people in your bathroom! Like so:
How do you feel about sharing a gendered public restroom with a trans person whose gender identity matches your own?
- I love it!
- I’m uncomfortable
- I don’t mind - the same as I feel about sharing with cis people
~
IT CAPITALISES TRANSGENDER FOR SOME REASON
When people use a capital letter at the start of a noun, it denotes some significance. It indicates that this group of people is special in some way.
Sometimes it expresses respect, like with names of people or places or cultures. This is especially true if the group of people have asserted that their noun should be capitalised, as with the Deaf community.
When the capital letter hasn’t been claimed by the group, it can feel like that group is being made special. Trans people have long been fetishised and alienated, and the capital letter stings a little bit because I think it’s fair to say that most of us just want to be seen as normal, be given the rights and respect and dignity that we deserve as humans, and be left to get on with our lives, instead of having, oh I don’t know, cis people debating our right to basic bodily functions. For example.
In this case, the capital T is condescending, fetishising, and alienating.
Of course, it might be an honest mistake. And if there were any other typos, I might be willing to give it the benefit of the doubt. But there are no other mistakes, and this poll was promoted, someone paid money for this poll to be shown to thousands of people. The typo slipped through the pre-payment spellchecking that even a casual pollgeek would have done when laying some cash on the table. I’m not saying it was deliberate, but there’s more at play here than a complete accident.
~
In conclusion, this poll is basically meaningless. There’s no way of knowing which question anyone’s answering, it assumes trans people are completely powerless and detached, it gives cis readers all the power, and it alienates the trans readers. The answer options themselves make it look like the issue is controversial, when actually there is a clear majority support for trans people being able to use restrooms.
So when you see polls like this, before you make judgements based on the answers, it can be good to think it through.
Imagine finding out 2 days before the test that an effective estimate and sufficient statistic are both statistics
Can’t be me
So I made this summary mindmap beast in an attempt to remember all the formulae for the test tomorrow
Unicorn March is all about pride for the most forgotten and endangered parts of the community.
Sometimes, pride means knowing more about the struggles that people like you face. Pride from within, from knowing what you’re surviving. Pride from other community members and outsiders, supporting you and loving you for what you are, not just what you overcome.
This infographic collects all the current data on ace-spec oppression into one thread. (It exceeds Tumblr’s image limit, so if you’re only seeing the first post, check the notes for the rest.) Feel free to save any of these images to share. Tag @unicorn-march if you can, when you use these in a reply or your own posts; it would be great to see how this info helps people. Image descriptions are in the alt tags.
Links to all the sources, in the order that they appear:
UK Government Equalities Office. (2017) National LGBT Survey. Data from more than 108,000 LGBTQIPA+ people across the United Kingdom. You can review the data and crunch your own numbers by clicking “analyse” on their website! They also have a report which presents some of the data, and their commentary.
Cantor, David, et al. (2015) AAU Climate Survey on Sexual Assault and Sexual Misconduct. Data from more than 150,000 college and graduate students across the United States.
Samuels, Gina E. Miranda, et al. (2019) Voices of Youth Count In-Depth Interviews: Technical Report. This is a 22-county study of 4,139 unhoused youth across the United States. They worked with local agencies, and were careful to include a mix of rural, suburban, and urban areas of varying densities and demographics. This report has a broad overview of their findings and recommendations.
The in-depth data represented here comes from Chapin Hall at the University of Chicago. (2017) Youth Homelessness in San Diego County, California: Findings from the youth count, brief youth survey, and provider survey. Although that’s a San Diego-specific study, Appendix E provides the data from all 22 counties for comparison.
The Williams Institute. (2016) LGB Within the T. This paper crunches the data from the 2011 National Transgender Discrimination Survey, which did in-depth interviews of 6,450 trans people of all orientations.
Borgogna, N. C., et al. (2018) Anxiety and Depression Across Gender and Sexual Minorities: Implications for Transgender, Gender Nonconforming, Pansexual, Demisexual, Asexual, Queer, and Questioning Individuals. This paper crunches the data from the ACHA National College Health Assessment, a twice-yearly survey of (at last count) 67,972 students at 98 schools across the United States.
Yulea, Brotto, & Gorzalska. (2011) Mental Health and Interpersonal Functioning in Self-Identified Asexual Men and Women. This is an older study, by comparison, and much smaller, so it’s used very sparingly here.Salway et al (2019). A Systematic Review and Meta‑Analysis of Disparities in the Prevalence of Suicide Ideation and Attempts Among Bisexual Populations. An extremely thorough analysis of the data available in 46 studies on LGBT suicidality, the reasons for different findings in different studies, and the most likely actual numbers.
Bauer et al (2018). The 2016 Asexual Community Survey Summary Report. This is an ongoing annual online survey of major asexual communities; the 2016 survey received a total of 9,869 responses (Ace = 9331 and Non-Ace= 538). As it’s not a peer-reviewed published study, it’s used even more sparingly here; the only data used here from this report is the percentage of cis aces who had considered suicide.
Grant, Jaime M. et al. 2011. Injustice at every turn: A report of the National Transgender Discrimination Survey. Further excellent analysis of the data from those 6,450 trans people.
Kuper et al. 2018. Exploring Cross-Sectional Predictors of Suicide Ideation, Attempt, and Risk in a Large Online Sample of Transgender and Gender Nonconforming Youth and Young Adults. Crunches data from the largest sample to date of transgender and gender non-conforming young people, a geographically diverse group of 1,896 respondents ages 14-30.
Good morning folks, and welcome to today’s statistical map - it’s about something I’ve tracked here for ten years, and about which I am passionate: linguistic diversity (or lack thereof) at Eurovision. ESC was envisioned not just as a peace project but as a showcase of European cultural diversity, and so I personally find the prospect of an almost entirely anglophone contest to be a concerning one.
In 2013, 17 countries out of 39 - some 44% - sang entirely or almost entirely not in English, but it’s a proportion we have seldom come close to since. In the following three years, no doubt partially because so many of the non-English entries did not qualify, we saw annual reductions to a historic nadir of only 15% non-English songs in 2016. We got a modest 1% bump after Jamala’s victory, and a major jump to 33% after Salvador - but in 2019-20, after two English language winners, the number went back down to 22%.This year, the proportion is slightly lower yet - only 21% of songs have no or next to no English lyrical content, whilst 69% of songs have no or negligible (think individual words like in San Marino or Cyprus) non-English content, which is higher than even last year, and the highest since 2016. With even Portugal,who had never before sung entirely in English, breaking that streak, all bets seem to be off the table as to which countries will sing in their native language outside of Serbia, Albania and the Romance-speaking trio inside the Big 5.
That being said, it’s not all doom and gloom this year - we’re getting a prominent language début in the form of Jeangu’s Sranan Tongo choruses, which is also the first time NL sends a song with some non-English content since 2010. The miniature Eastern Slavic language revival continues, with Go_A once again sending a song in Ukrainian and Russia sending their first song mostly in Russian since 2009. Denmarksent their first song entirely in Danish since 1997. Switzerland’sGjon’s tears once again opted for French, and his will be the first francophone song performed from Switzerland since 2010. With France, who have only their 2nd entirely French language song since 2016, also perched at the top of the bookies’ odds, the possibility for an influential non-English win is not negligible.
We also get to hear a little Azeri for the first time ever; a little Czech for the first time since Czechia’s début in 2007; a little Germanfor the first time since the same year. I really hope for a year of success for songs in other languages so that delegations are encouraged to not just use their native languages as a spice to be sprinkled very sparingly over songs, but rather to send songs almost entirely in them in the near future.
Many “gender critical” people on this website like to site a 2011 Swedish study to claim that trans women have the same rates of violence as cisgender men. This claim is a gross misreading and should be challenged whenever seen. Below I will be showing quotes from one of the researchers from the study showing exactly how wrong the “gender critical” claim is.
For starters the study found that women who transitioned from 1989-2003 did not even demonstrate a male pattern of criminality let alone anything close to the same pattern of violence. Which means that the claim of trans women being as violent as cisgender men is definitely false as a general statement for all the women who transitioned from 1989 onward.
The study as a whole covers the period between 1973 and 2003. If one divides the cohort into two groups, 1973 to 1988 and 1989 to 2003, one observes that for the latter group (1989 – 2003), differences in mortality, suicide attempts and crime disappear. This means that for the 1989 to 2003 group, we did not find a male pattern of criminality.
However we also need to look at another key distortion of the facts that the “gender critical” camp manufactured. Pattern of Criminality does not mean the same thing as pattern of violence. Even the trans women in the 1973-1988 group were not shown to exhibit the same rates of violence that cisgender men do. What the study actually does show is that prior to 1989 trans women were being convicted at the same rate as cisgender men, not for the same crimes.
As to the criminality metric itself, we were measuring and comparing the total number of convictions, not conviction type. We were not saying that cisgender males are convicted of crimes associated with marginalization and poverty. We didn’t control for that and we were certainly not saying that we found that trans women were a rape risk.
The idea that trans women are just as likely to rape as cis men is a lie. It is a lie which has become a central tenet of denying trans women access to life saving shelter and community and this lie needs to be confronted where ever it tries to take root. Whenever you see this lie, shine a light on it because it can only grow in the dark.
(source)
I’d like to point out that actual studies have linked trans suicides to constant misgendering as a cause (one of, I should add), whereas affirmation of gender is known to do the opposite. So intentionally misgendering trans people is pretty much legit trying to kill them.
In other news, water is wet.
Where the study/source?
I’m uber busy at work and I don’t have the time to pull up the specific spot it’s discussed. But, look up the work of Dr. Ry Testa. He’s a trans man tackling these issues.
Someone who knows how to navigate resources and has the spoons and time try and find the paper, save trans momma the trouble.
Sounds super legit though. Being misgendered almost always causes some dissociation and social withdrawal form me since it belittles my identity. I don’t get suicidal thoughts but I could sure as hell see how someone who does would be affected by that.Hiya!
I’m hoping I could help somewhat.
This is all the work Dr. Testa’s done.Development of the Gender Minority Stress and Resilience Measure.
Effects of violence on transgender people.
I think this is the one though; The relationship between gender-based victimization and suicide attempts in transgender people.
Misgendering a trans person is a violent act. It’s an emotional abuse that cuts deep into who we feel we are at our core. We have enough to deal with internally, adding external voices to that can destroy us.GBV was associated with both a history of suicide attempts and numbers of suicide attempts in this study. Participants who had been exposed to victimization were almost four times more likely to attempt suicide than those who had not experienced victimization. In addition, over three-quarters of those who had a history of suicide attempt had attempted multiple times. Trans men exposed to GBV were 2.1 times more likely than trans women to have a history of a single suicide attempt and were 3.2 times as likely to have made multiple suicide attempts. A history of suicide attempts was significantly more common among multiracial individuals, and significantly less common among those of higher SES.
From the last article linked to.
Even more relevant is actually The Gender Minority Stress and Resilience Measure article…
In addition to the violence, rejection, and discrimination experienced by both LGB and TGNC people, TGNC individuals may experience an additional distal stressor, which this study has labeled nonaffirmation. Nonaffirmation occurs when one’s internal sense of gender identity is not affirmed by others. For instance, a trans woman may be addressed as “sir” when making a purchase or answering the phone, or might be referred to with her former male name by an individual not yet fully comfortable with her gender identity. TGNC individuals who do not identify as either male or female (e.g., individuals who identify as genderqueer) may also experience a sense of nonaffirmation as people in their life are unable to refer to them in gender neutral ways. Nonaffirmation of one’s gender identity may occur more or less often for any TGNC individual, based on how their gender expression is interpreted by their social context at that particular time (Beemyn & Rankin, 2011; Bockting & Coleman, 2007; Nuttbrock, Rosenblum, & Blumenstein, 2002). In sum, the literature suggests that TGNC and LGB populations experience both common and also unique distal minority stressors.
It also cites these works in relation to the concept:
Beemyn, G., & Rankin, S. R. (2011). The lives of transgender people. New York, NY: Columbia University Press.
Grant, J. M., Mottet, L. A., Tanis, J., Herman, J. L., Harrison, J., & Keisling, M. (2010). National Transgender Discrimination Survey. Report on health and health care (pp. 1–23). Washington, DC: National Center for Transgender Equality and National Gay and Lesbian Task Force.
Hey, thanks. I haven’t had a good mental headspace for reading recently, so this is helpful.