,
18 tweets,
4 min read
Read on Twitter
Thread incoming. I've been thinking about this since it was posted this morning and trying to work out how that astonishing statement could be made. All data used by me here is from Ministry of Justice sources and the ONS.
The article states "officials forecast that numbers of trans or non-binary prisoners in England and Wales could rise to 1,000, up from 139 last year." The wording is somewhat deceptive - my initial inference on reading that statement was
"a rise next year" but on reflection I believe there must be a (deliberate?) misreading of the source at work here. I worked in data for use in infrastructural planning for years. Infrastructure plans generally cover at least a decade, often two or more.
In the case of prisons, we can probably assume at least 50 years due to the lifecycle of the assets. I have therefore assumed this period in calculations. To reach the conclusion stated in the article, there are three variables as I perceive it:
1. Greater volume of crimes. Assuming a greater volume of crimes with no rise in the proportion of trans people within the prison population, that then implies an increase in the number of incarcerated people over the next half century, to >600000 people
from the current c. 87000 - a >700% increase where the total UK population is only projected to rise by 19% in the same period, from 0.13% total Uk residents to 0.74%. As a comparator, the USA currently has the highest incarceration rate at 0.65%.
This rise seems implausible, not to mention nightmarish, so I am discounting it. Additionally, recent changes in crime reporting and recording methodology make projection here very difficult - while *reported* crime has risen significantly in recent years,
John Flatley of the ONS rightly points out that the number should be viewed in the context of the overall decline in crime indicated by the Crime Survey of England and Wales.
2. Greater volume of trans people. Given changing societal attitudes, this seems plausible in theory. However, the ONS themselves have no usable data on the number of trans people in the UK - their position paper from 2009 and subsequent updates to it through to 2017 state this
clearly. They add that gender identity information will be added to the 2021 National Census in order to build this dataset. Even then, it would take two further censuses (so in 2031 and 2041) to give three data points - the absolute minimum I would
ever consider using for trending (and even then with massive caveats attached for unreliability). Unless there is another data source in use of which I am not aware, I don't think this can be the source.
3. Greater proportion of criminality requiring incarceration among trans people. Given the extremely low numbers of trans people in prison (139 quoted in the article) and the transient nature of the prison population as sentences are handed down
and carried out, extrapolating a massive increase from this variable seems the most likely. Indeed, the MoJ themselves say the figures "are not yet a reliable reflection of the numbers and location of trans prisoners in the prison estate."
The available dataset does not consider prisoners who have been issued a Gender Recognition Certificate, or prisoners serving sentences too short to benefit from having a case conference to explicitly manage them as a transgender prisoner or people
who have not reported that they identify as trans. It is therefore conceivable that the vast increase quoted in the article has been somehow reached via manipulation/ extrapolation of this variable. Frankly, if this is the case, I call
BAD STATS IN BAD FAITH. Should anyone know the source and methodology of this supposed trans crimewave and be willing to share it, I will be happy to retract, but as it stands, I smell statistical abuse.
"If you torture the data long enough, it will confess to anything." Darrell Huff, "How To Lie With Statistics" (1954).
@threadreaderapp unroll
