Thread by @fairplaywomen

Baroness Jenkins asks @GEO minister whether she agrees that a small misclassification of sex can have a large distorting effect on gender pay gap in professions where <5% workforce is female /1

She gets a very unsatisfactory answer from the minister:
"no plans to change the guidance...gender pay is not supposed to be a data collecting exercise...to do so would increase the burden on employers" /2

Employers are obliged by law to COLLECT DATA on the gender pay gap and GEO publishes it.

And the point of it is to determine gap in pay between males and female

Of course its a data collection exercise! And of course sex matters! /3

The GEO guidance tell employers "the data you must gather" /4

gov.uk/government/col…

This is what GEO tells employers about the data they must gather:

"Locate employees GENDER IDENTIFICATION using information employees have already provided for HR/payroll" /5

If this information is unreliable employees can ask employees to "confirm or update their gender". /6

Data on 'non-binary' employees even gets omitted from this legally required data exercise /7

Baroness Jenkins questions was an important one about data accuracy. Too often we hear people dismiss concerns over inaccurate data on sex because "the numbers are too small to make a difference". Nonsense /8

Data experts have explained repeatedly why small errors can make a large difference /9

fairplayforwomen.com/small-errors-i…

Many occupations are predominantly male. Fewer than 1% of carpenters & mechanics are female. A small number of males misclassified as female will not affect male totals but will make a big difference to female totals obscuring poor accessibility for women in some careers /10

The same principle applies in other male dominated statistics such as perpetrators of violent and sexual crimes. This was discussed in the HoL last night.
Once against non-experts dismiss legitimate concerns as insignificant /11

Here's Lord @brianpaddick telling us that "he was never any good at mathematics"

Nevertheless he goes on to assert that:

"it's quite clear to me that trans people are not going to make any statistically significant difference to the crime figures" /12

@brianpaddick @brianpaddick then says "Unless you assume, and there is no factual or statistical basis to think otherwise, that trans a more likely to commit crime or particular types of crime".

His logic is flawed /13

@brianpaddick Transwomen are, by definition, people born male. The vast majority of perpetrators of violent & sexual crime were born male (97%). Why wouldn't we assume a person born male has a male-type propensity to crime? There's no factual or statistical basis to think otherwise /14

Please start listening to the experts. Think about the detail. We have and we are getting fed up of continually explaining it time and time again.

Good data on sex matters. Our rights depend on it /15

Here's a stark example how transgender numbers can distort female figures if gender identity is recorded instead of sex.

At the last count MOJ says there are 125 female sex offenders in prison in England and Wales. A tiny number compared to the 14,000 male sex offenders /16

Out of those 14,000 male sex offenders we know that 70 identify as woman. That's just a small proportion of the total male population.

But if those 70 male sex offenders are added to the female total it makes a huge difference. Number rise from 125 to 195. /17

When gender identity is recorded instead of sex we get around a third of all 'female' sex offenders being male.

Its nonsense to say these numbers are too small to make a difference. Stop to think for a moment and its obvious it does /18