Selective use of statistics in the Guardian

[DipsyLaLaPo]

https://www.theguardian.com/world/2023/apr/19/half-of-lgbt-young-adults-in-uk-are-estranged-from-a-relative-survey-finds

It is just glaringly obvious clickbait (I hesitate to assume a more deep rooted political agenda as lots of these things are usually down to lazy stupidity) when transgender elements of a statistic are highlighted, but they aren't even the most impacted on the graph immediately above?

Almost one-fifth (19%) of transgender people and 23% of non-binary people in the survey said they were not close to immediate family members.

lookat the graph and you'll see three categories with even higher numbers than this. Gay, lesbian and bi.

In fact, nearly 10% of gay men are estranged from their entire family. Trans and non binary are the least affected by total estrangement, according to this graph.

And yet lesbian and gay are the most confident about coming out.

There's a story for you. Sick of seeing stats misrepresented in major national news outlets.

Some stats here. https://www.psychologytoday.com/gb/blog/brothers-sisters-strangers/202212/latest-poll-1-in-4-are-estranged-from-a-family-member?amp

Looks like about about 1/5 to 1/4 of families are affected by some kind of estrangement based on surveys in the UK and USA. The USA rates are higher.

Take a look at AIBU or the relationship board here and you'll see loads of people 'gone/going NC' with a family member

There seems to be an awful lot of narrative around narcissistic parents these days, mostly mothers, full stop. Including on BBC R4 etc. Along with a narrative than anything less than full affirmation is outright rejection and/or hate. A la Jeffery Marsh but dressed up under a veil of mental health.

It is worrying. Really worrying.

In this survey, the first graph shown in the guardian article indicates the "non-LGBT" respondents at about 30-odd% being estranged in some way from family. So the rates for LGBT overall are higher than that (not shown as an overall figure on the same graph, but mentioned in the article).

Apparently the full report on this comes out in June from the charity mentioned. Hopefully it will have some clearer explanation of the methodology.

interpretong statistics is definitely not my forte, but a cogent critique of the piece sent to either readers’ editor of letters page? My favourite misuse of statistics is the high rate of German women getting pregnant by rape, especially in predominantly catholic federal states. Correlation rather than causation showed that getting an abortion was far easier if a woman said rape or coercion led to the pregnancy.

https://hbr.org/2015/06/beware-spurious-correlations