Addendum to: Cambridge University discriminates against children from private schools.

[Marchesman]

"From 2013 to 2023 the proportion of UK state-school admissions rose from 61 per cent to 73 per cent. This increase was made possible by undeniable discrimination against another group of students – those who, whether through a choice made by their parents or a scholarship won by their talents, attended fee-paying schools."

For an insider's perspective on Cambridge University's descent into mediocrity see: "Decline and fall: how university education became infantilised" D. Butterfield, Spectator 26th Oct..

@DiscoinFrisco ,

‘When i was at Cambridge the private school students were furious about this because they'd been told all their lives how special and better they were.’

I have to disagree with this. When I was at Cambridge from a London day public school, I had no expectations of what the other students would be like. My friendship group (many of who, I still see nearly 4 decades later) were a mixed bag. What school someone went to just didn’t come up much in conversation.

And, having taught at two private schools, no one tells the pupils that they are special and better. After all, most teachers can’t afford to send their own children private.

(there are some honourable exceptions to this, Eton being the most egregious. I have never met a modest Etonian. I am sure they must exist, but I do think the school exists to perpetuate class superiority, although they do do a lot of good community work).

Different experiences then. Also a day public school is probably more similar than a northern city comprehensive.

What the issue in giving those who have had less educational opportunities but are obviously bright, the chances in somewhere like Cambridge? Sounds good to me.

Many employers now interview blind for graduate positions so where you went to Uni is becoming less relevant.

"You need to be able to identify who should have been admitted by merit to start off with."

Amazingly, the people involved with admissions at Cambridge had the very same idea.

So they identified characteristics that predicted "merit at output" and specifically and repeatedly targeted one with progressively poorer outcomes.

If men got 40% more maths firsts than women, then I'd likewise say that they had overshot on trying to get more women into STEM.

That's an interesting point about the increased risk in contextual offers, and where best to draw the line to catch exceptional students from weaker backgrounds: accepting the cost of some contextual offers not working out. I'd certainly agree that would fall within Cambridge's remit.

It would definitely be interesting to see some more analysis, to understand whether that's happening. How would you define 'truly exceptional'? Postgrad / research / international fame/Nobel prize? It would be interesting to see how the percentages change as the 'truly exceptional' funnel gets smaller, across each of the factors they check, including private schools. If eg private school students were 1.4 times more likely than state school students to get a first, but less likely to get a Nobel prize, I completely agree that would change the picture (and I would accept the discrimination). If on the other hand the discrepancy continued, and they were more likely to achieve at all levels, right up to getting a Nobel prize, then I think you'd have to agree that the discrimination was misguided.

Your argument would be much more persuasive with a series of results from each year showing the same pattern, rather than a single data point.

It's reasonable to assume that merit at output - after 3 years of equal, excellent education - is a good reflection of actual merit at entry (which is what Cambridge should be trying to achieve with contextual offers). Ie if there admission policy works correctly, there should be no difference at output.

I think the issue with analysing the ‘onward funnel’ is that progress through it is to an extent dependent on the same variables that may have affected an individual during school. Poverty / illness within a family nay limit a student’s ability to seek out international postgraduate or post-doctoral opportunities, need provide and natural risk aversion may mean a student (however brilliant) seeks a paycheque rather than a research post etc etc.

Students from private schools were more likely than students from comprehensive schools to be awarded firsts by the following multiples from 2017 to 2023: 1.24 ,1.17, 1.19, 1.17, 1.23, 1.43, and 1.44.

Prior to 2017 the figures are available from the university's website, dwindling to 1.05 (as I recall) in favour of (but not statistically significantly) students from comprehensive schools when the record starts.

Grammar schools? For this to be a state vs private debate, you have to include them.

2019 - 2024?

University absolutely should level things. 3 years of outstanding education is a lot! Surely long enough for anything mystifying to be demystified. All students are in a new situation, and bright, curious Oxbridge undergrads should surely be open to learning whatever new ideas they come across, including different cultural norms. Not many private school kids have experienced dinners in gowns either.

If this were the case, then you'd expect that there wouldn't be any difference in degree results at the end of the 3 years: those with lower A levels due to poor school conditions would have caught up. This is how contextual offers should work - and I haven't seen anyone on these threads arguing against those.

What the OP is raising is that there is a huge difference in degree results, because state school candidates who are genuinely weaker - not just having lower results due to poor school conditions - are being offered places ahead of genuinely stronger private school candidates.

It is not the same. When high socioeconomic status privately educated student numbers went down they were just replaced by high SES students from state schools. Socioeconomic status has no significant effect on degree attainment.

I think that is over-simplistic - the very specific nature of Cambridge courses and its particular traditions, peer group and lifestyle are less alien to some than to others, and the barriers posed by them can affect output. That’s why I find the bridging course concept really interesting - it aims to break ‘non ability related’ barriers for students who are of sufficient calibre to be admitted to Cambridge. The statistical high performance in finals of a college I know to offer one is anecdotally interesting.

But that could be selection bias. So private schools select the brightest kids who are most likely to get firsts in uni.
So if, for the sake of argument, you take the top 20pc most academic kids in the country at 18, if more than 50pc of those go to private schools (due to bursaries etc) then you'd expect to see them overrepresented in the ensuing results at Uni.
The ones in private schools who didn't get into Cambridge wouldn't necessarily have done better than the state school ones who did go to Cambridge. Just because less state school kids get firsts doesn't mean that they did worse than the private school kids who didn't get in.
Firsts are also banded so not everyone can get one. Firsts are also only a measure of who does best in the exams, it's not actually necessarily the most overall long term academically most successful people. My DB got a 2:2 but now is a successful academic researcher doing very well at a top London university.

I am not talking about degree attainment. A previous poster was interested in attainment post degree, and I was pointing out that life choices post degree are not purely ability-dependent.

Again, a parallel is in male vs female representation at these higher levels - since motherhood has such a dramatic effect on onward career progression, ot is not reasonable to assume that lack of representation of women at the highest levels of academia / academic prizes = women are less able than men and should therefore be selected against at the point if admission.

Students from grammar schools do consistently better than students from comprehensive schools and worse than students from private schools.

In 2023 19.9%, 23.5%, and 28.6% of comp, grammar, and privately educated students respectively achieved firsts.

Since 2017 the narrowest gap between grammar and private was 2018-19: 23.4% vs 25.6%.

Are those all with matched subjects? Or simply the raw data with issues of subject balance retained?

I would also be really interested in proportions of students with sorcial considerations for exams being applied - since many more private school pupils have these for A levels than comprehensive school pupils, with grammars less than half of the comprehensive figure, I wonder to what extent these extra time etc allowances continued into university?

It is wholly attributable to selection bias on the part of the university in favour of applicants from state schools with lower potential.

Raw data, all of the factors identified in the university's 2020 analysis will contribute.

Why do you say they have lower potential? Potential for what exactly? Potential to get a first? You don't know that for sure. Only a random control study would confirm this.

I think you're confusing equity and equality.

Having a higher prevalence of academically able kids at private school than in the general population as you suggest should result in a higher proportion of the kids at Cambridge being from private schools than the 20% who take A levels at private school. And that is indeed what we see - with 30% of Cambridge undergrads coming from private school.

But if the selection is fair, then the 'cut off' for how academically able you need to be to get in should be the same for private and state school. A level grades might be different: but contextual offers should be set such that offers are made to the most intrinsically academically able candidates regardless of school type.

Assuming a similar distribution of ability above that cut off (which I see no reason to doubt) then the same percentage of private and state school students would get firsts. And that's not what we see.

Furthermore all that data shows is a correlation not a causation. It doesn't tell me anything about the people who did not get admitted and it doesn't tell me anything about the factors that may have affected people's ability to get a first. People from different socio-economic backgrounds may have different pressures which may contribute to their ability to study eg in the holidays etc etc.
Telling us that privileged people do better isn't actually news !

Which is why it's relevant that the Cambridge analysis shows that school type is statistically very significant to degree outcome independently to socioeconomic status.

That shows that it isn't caused by having to work, family pressures etc.

Equally privileged state school students are getting lower degree results.

Edited: I originally said that SES doesn't affect degree outcome. There may have been some difference - but it was dwarfed by school type.

Simply reducing it to all state v all private delivers a nonsense data set anyway, even if you take multiple years rather than cherry pick according to your prejudice. Take out a few outlier super-selective schools (Westminster, SPS, SPGS) who on their own account for 10% of the private school Oxbridge intake and see what the degree results of state v private are...