A slight break from the norm, there’s no regurgitated email correspondence in this chapter.

At this point in the saga, my friend, and all round good guy, Mica suggested this whole thing was fuelled by my notorious trouble maker personality. To convince her otherwise I decided to crunch some data.

Fortunately UCAS provide data on the people applying for and being accepted for PGCEs. While they break it down into a whole variety of groups, I haven’t found where they break it down by Primary/Secondary AND “ethnic group”. So all this data covers Primary and Secondary PGCEs, I think this might give the appearance that Primary PGCEs aren’t ridiculously skewed towards female white applicants.

There’s a whole heap of shit that makes the UCAS data less than ideal, I’ve listed them at the bottom because they’re pretty boring.

I couldn’t face wading through the data for all 100 or so universities so I just picked 13:

  • The 5 Universities in Wales
  • The 4 Universities within 50km of my house
  • The 5 Universities that accept a similar number of applications as Cardiff Met (±5%)
  • The 5 Universities that receive a similar number of applications as Cardiff Met (±5%)

The mathematically gifted among you will have spotted that makes 19, but there is overlap between the groups.
In my head this Venn diagram made things clearer, in practice I don’t think it does.

Now, I would love to bore you all with the copious amounts of spreadsheets I have but I’ll just hit you with some Cardiff Met centric highlights.

But before that I want to say, this next section is based on my findings and may not be a fair representation of universities’ application processes. Just because they maybe look sketchy doesn’t mean there is discrimination*.

When I saw this, my heart genuinely hurt.
White women under 21 years old are 6.32 times more likely to be accepted on a PGCE at Cardiff Met than Black men aged 25 – 29.

What the actual fuck?

There were only two universities I looked at that were worse:
Liverpool Hope University, where White women under 21 years old are 7.30 times more likely to be accepted than Other ethnicity men aged 40 and over.
The University of South Wales, where White men aged 40 and over were 50.42 times more likely to be accepted than Black Women aged 21 and under.

Disclaimer: I hate these ethnic group categories, and the term “ethnic group” is probably a bit shit too. But they are the terms UCAS use.

So, Cardiff Met hasn’t covered itself in glory there, but congrats to them for making it onto the podium in the “Universities where your protected characteristics have a fucking massive impact on your likelihood to get in” event. That could do with a pithier title.

Maybe it’s not a big deal? Maybe it’s better if we look at it from another angle.

Well I pulled out my trusty Casio fz-991ES PLUS and did a Chi squared test to prove of disprove the hypothesis that Gender and Ethnic Group affect the likelihood of being offered a place on a PGCE. I ignored age for this because maybe being older or younger correlated with experience or something.

Cardiff Met didn’t do as badly here. If you applied in 2014, 2017 or 2018 you probably wouldn’t have been subject to sexism or racism.

Three years without sexism or racism.
That’s… good?

The previously mentioned Liverpool Hope had only two years without an “ism”.
St Mary’s University, Twickenham racked up only one “ism” free year.
The University of East London also scored a single un-“ism”-ed year.
The University of Exeter was without an “ism” once, in 2014.
The University of Wales Trinity St. David also enjoyed its only “ism” free year in 2014.

You’re probably thinking how come Cardiff Met was the third shittiest initially but then only the sixth shittiest?

The first investigation looked at the probability of an applicant being accepted by working through a Probability Tree Diagram.

So in a year the likelihood of a Mixed ethnicity man aged 25 – 29 being accepted was calculated using:

But the second investigation compared the actual number of applicants to the expected number of applicants. From there I looked at whether the difference between actual and expected numbers were different enough to rule out random chance.

I think I’ve explained some stuff there?

So, until next time, if you understood what I was on about and want me to explain more maths, Hire me?


The UCAS data rounds all the data to the nearest 5, so in 2017 when Cardiff Met accepted 5 of the 25 Asian applicants. They might actually have only accepted 3 of 27 Asian applicants or a nicer thought is that they accepted 7 of 23.
There are 3 entry routes, “Apply 1”, “Apply 2” and “RPA”. Ive only looked at “Apply 1” because that’s where the most data was available.
Ethnic Group data was only available for UK applicants so I ignored EU and overseas applicants.

Bits and bobs have been scaled to give a more representative estimate.
I might have made some errors in the maths, but I’m fairly confident the numbers are right.

*If you’re thinking of suing me, don’t it’s all satire and for entertainment.

But, this is a serious issue everyone needs to be talking and thinking about.