Prof. Kathleen Stock shout out for statisticians re Women's & Equalities Report cited suicide statistics

[R0wantrees]

twitter.com/Docstockk/status/1046446641316843520

"1) Calling statisticians which follow me - What do you think about this use of stats? I could do with some help. Here's a dramatic central claim from the Women and Equalities Trans Equality inquiry Final Report, 2nd para. 'A third of trans adults attempt suicide'.

2. Here's the original report it's taken from. www.gires.org.uk/wp-content/uploads/2014/08/trans_mh_study.pdf

(continues see screenshots)

"4. Does the claim '35% attempt suicide at least once' look credible in this context? (I'm leaving aside other flaws in the report, e.g. it's self-selecting, there's no follow up, it's online, etc). Thanks in advance."

Embedded link refererred to in 5)
www.transgendertrend.com/a-scientist-reviews-transgender-suicide-stats/
see Twitter link for discussion

I am very aware of the sensitivity and responsibility required when discussing suicide, ideation and self harm

Samaritans' information and contact details: www.samaritans.org/how-we-can-help-you

www.samaritans.org/media-centre/media-guidelines-reporting-suicide

I think most people on that Twitter thread are more credible than me.
But a few points I would like to add:
-have they controlled for mental health conditions (i.e. if someone has a diagnosis of clinical depression, this would naturally put them at higher risk for something like suicidal ideation)
-have they compared these stats to those with depression/mental health conditions

Comment from the linked thread :

"I'm a statistician and a retired consultant psychiatrist.

As a statistician, the issue of self selection is extremely important.

Firstly, we're not looking at a representative sample of trans adults but only those who complete the survey.

Secondly, instead of all those who complete the survey they're taking the percentage of those who choose to answer a question on suicidal ideation. The percentage they give is 35% of 581 but of the whole sample of the 889 completing the survey this would be 23%.

As an aside, they can't do simple arithmetic. They claim that of 436 people who had experienced suicidal ideation 48% had made an attempt and that this was equivalent to 35% of 581. But there is no number which rounds to 48% of 436 and 35% of 581."

There is a podcast called 'More or less' looking at "the numbers behind the statistics".
They research and answer listeners questions on statistics they have heard in the media.
Perhaps the question could be forwarded to them. Would be interesting if it is featured on the programme - it's a BBC 4 podcast.

I don't have a Twitter account - can somebody who does advise Prof Stock that she might be well advised to speak to Full Fact about whom they checked the statistics with when fact checking those numbers a few months ago?

twitter.com/FullFact

There is a podcast called 'More or less' looking at "the numbers behind the statistics".They research and answer listeners questions on statistics they have heard in the media.

Unfortunately, the BBC has form for this albeit it was BBC Reality Check who tried to disprove FPFW's numbers:

blogs.spectator.co.uk/2018/08/is-the-bbc-scared-of-the-transgender-debate/

www.mumsnet.com/Talk/womens_rights/3334473-BBC-reality-check-on-trans-prisoners

More or Less have been asked and they're avoiding it like the plague. If you could twist their arm enough they would stick to the BBC party line like glue and come out with Most Oppressed People Ever™ narrative.

If you don't believe me have a look at this link:

www.bbc.co.uk/news/uk-42221629

It is a jaw-dropping misrepresentation of the facts. Unfortunately the BBC will continue to be very partisan and produce propaganda on the issue for quite some time to come.

I was unsurprised to see arranfan got there first. They've also provide a useful mumsnet link that's worth reading.

The really depressing conclusion I've come to after years of hearing Trump say "fake news" is that there is genuinely fake news out there and some of it is coming from progressive liberal media outlets. The BBC, along with other organisations, will, for certain stories, decide firstly what the story is to say and secondly will bend the "facts" around the story to get the result they want.

There's also another independent factchecker organisation.
They comment after BBC Question Time etc
They did something after Paris Lees statement on QT

oops, cross-posting with arranfan

Full Fact is the UK’s independent factchecking charity
fullfact.org/

More or Less can be great, but they wont touch this.

They will continue to stick to the really topical, hard hitting issues like the number of wizards in Harry Potter's universe and the flying speed of fictional giant eagles.

I don’t know about the suicide statistics, haven’t got to that bit, but looking at the report, they say that they have excluded data from anyone under the age of 18.

They then say that they have a dataset of 889 people.

They then say that only 501 of those 889 people answered the age question, meaning potentially 388 points of data in the study are from people under the age of 18 who should have been excluded.

In addition, people from outside the UK and Ireland were excluded, yet of the 889, only 518 answered the Location question.

Why have they included hundreds of people who may not meet their inclusion criteria?

More or Less can be great, but they wont touch this.

Still good for people to request it should they wish them to.

The editorial decision making policies on some BBC programs is going to come under question at some point.

If there are journalists who wish to cover these issues and are being prevented to by editorial decisions, they may be assisted by the numbers of people writing in.

www.mumsnet.com/Talk/womens_rights/3379683-Can-we-just-talk-about-BBC-Womans-Hour

www.mumsnet.com/Talk/womens_rights/3296433-BBC-Bias-Collecting-Examples-here

Having looked at the quote from the government report, I am horrified by their use of this statistic in the way that it has been written.

“High levels of transphobia are experienced by individuals on a daily basis”....” - with serious results. About half of young trans people and a third of trans adults attempt suicide”.

This is clearly attempting to blame transphobia for the attempted suicides. However the Gires report itself says “For the participants in this survey, 65% felt that there were trans related reasons which made them think about or attempt suicide, while 61% identified non-trans reasons as relevant. This suggests that although being trans may be one factor, others are also relevant.”
Same for self-harm “Just under 60% of the participants felt that there were reasons they self-harmed which related to them being trans, while 70% felt there were non-trans related reasons for their self-injury”

It is impossible to blame transphobia for the attempted suicides using the data in the report and yet the government report clearly links the two. That’s dishonest.

And it also confuses correlation with causation. It is possible that people who have mental health issues leading to attempts of suicide are more likely to consider themselves trans, than people who are trans are more likely to attempt suicide. Or there could be a third factor linking the two. There also seem to be a high level of mental health issues and disabilities among respondents unconnected to being trans.

That’s assuming the data is valid. Which is isn’t. Their methodology is horrible, and I can’t actually see where in the data they actually capture that all the respondents are trans. Many of them don’t identify in either the category ‘trans person’ or ‘transgender person’, many didn’t even answer that question. Quite a few have made no attempt to transition and don’t intend to.