Oxford to drop PAT, MAT, TSA, etc

[Muu9]

https://www.ox.ac.uk/admissions/undergraduate/applying-to-oxford/guide/admissions-tests

Thoughts? I think it's a bad move for physics, as the time-pressured nature ESAT simply isn't very good at spotting the deep thinkers most physics programs look for.

In the past, it was thought that Cambridge interviews a higher % of candidates post ESAT than Oxford has done post PAT ie the ESAT was just to weed out the weakest of candidates who had no chance of recovering via strong interviews, whereas the PAT result was carried forward and was a more significant determinant in receiving an offer. I wonder if this will now change. Something to ask at open days!

Interesting thought, @Notanorthener

I've read elsewhere about teachers' children being overrepresented in the most selective higher education in countries other than the UK. It's intuitively not terribly surprising. IIRC in France the most desirable parenting couple to be born to from a selective HE perspective is a father who is a senior executive (for the monetary resources) and a mother who is a secondary school teacher.

@livelifeandenjoyit having fewer tests might encourage equity- but the MAT was free and TMUA is expensive. You can apply from a bursary - but that’s not that easy and I’ve known students not do that because of needing to book a place at a local centre

Unless to interview more applicants, scrap the MAT might kick back.
www.telegraph.co.uk/news/2026/01/29/oxford-ditching-traditional-tests-bad-brilliant-students/

I don't think anyone but the designers of the tests (TMUA, ESAT, TARA) is qualified to comment on their efficacy. Standardized tests such as the SAT and GMAT (whose academic content is less specific than that of TMUA, ESAT or TARA) show demonstrable efficacy. The widespread adoption of the UAT-UK tests will have the major advantage of creating sufficient data with which to analyze with statistical significance the relative performance of applicants from different schools and countries.

We can all have opinions, @Ceramiq .

I agree that the data will eventually tell whether the ESAT is meeting Oxford’s goals, which are theirs to determine.

The SAT etc (GMAT is merely one of a myriad of similar exams for American PG study) have so much data because they are used at such large scale. The SAT in particular is supposed to a test of aptitude or potential for success in undergraduate study and has been taken by many millions of pupils.

Of course there is a correlation with background; however many WP applicants are encouraged to take it because even at present results can be interpreted contextually. Sadly this message isn’t always being heard.

@poetryandwine I think we agree: the only measure of efficacy of TMUA/ESAT/TARA is whether they meet the goals determined by the universities that decide to adopt them.

Ironic that we are talking about tests for entrance into our best universities and someone thinks that it isn't legitimate to have a discussion about proposed entrance tests. It seems a very reasonable subject for discussion to me.

My especial concern is that the TMUA was administered extremely badly last year. We spent months literally wrangling with Pearson about how to get my son registered for the exam. I think that if Oxford is going all-in with Pearson there needs to be some assurance that there is going to be proper back up. (I am considered writing yet more letters)

I don't think that we cannot have a conversation about entrance tests. Indeed, the conversation about the delivery of the tests is important in the light of so much apparent chaos.

However, I don't think that the efficacy of the tests is up for discussion: England is unusual in investing hugely in research into teaching and assessment and what and how is being measured and to what ends is the business of the universities. These are not public examinations with a requirement for transparency.

The efficacy is a question for each university using the exams, @ceramiq. No one on this thread has implied otherwise.

But anyone with subject knowledge and content knowledge for teaching can appreciate that the MAT, PAT, TMUA, ENGAA and of course STEP each have much to recommend them, and each is getting at something a bit different from the others.

The format of the ESAT as I described it above means it inevitably gets at something else, something less sophisticated. The move towards equity (in spite of the real financial barrier mentioned by @ninja) probably does trump other concerns. But for those of us concerned with the diminishment of critical thinking, it is also a loss.

It will take a minimum if five years even to start getting good data.

Data collection has to start somewhere. Useful longitudinal data with, potentially, predictive value of degree outcomes will obviously not be collected and analyzed for a few years but indicatively useful data comparing students across schools/countries/courses will emerge sooner.

I read, not very long ago, one of Cambridge's admissions reports that highlighted that the predictive value of degree outcomes of A-levels and admissions tests was pretty poor, once a certain threshold had been met eg one student might arrive at Cambridge with 5 A-stars at A-level and another with 2 A stars and an A and it was impossible to tell which of the two would do better in their degree. So perhaps admissions tests for some courses do not need to test skills above a certain threshold of difficulty? Indeed, to do so might be counterproductive?

@Ceramiqi believe the STEP exam is an excellent predictor of success. A Levels less so. There used to be data also about whether AS grades were a good indicator but that’s largely irrelevant now

Yes, @Needmoresleep cited the predictive ability of STEP above.

I am not surprised. But I imagine constructing the STEP papers and marking rubrics is highly laborious. The ESAT papers will be at the other extreme. (Not that thus is inherently bad)

@ninja Absolutely and Cambridge has not abandoned STEP.

From what I’ve been told by students, this is a good move. The PAT etc seem to have very clunky and amateur interfaces with a built in calculator that is hard to use and wastes time. But the main reason is that they have been beset by data leaks, including students getting other student contact details and results. Apparently the gossip has been that it’s run by a bloke in his garden shed. Probably not because it is, but because it’s that bad.

It is well known that the PAT has always been written by the Physics Dept at Oxford. It is known for being difficult, which of course pupils moan about.

In 2024 Pearson VUE (surprise!) took over its administration. They initiated use of a new, clunky built in calculator. Many Pearson-administered university admissions tests have been beset by data leaks. Why would PAT be different? Why will ESAT be different?

Interesting about Pearson. I didn’t know that. I was simply reporting what students say where I work. ESAT doesn’t seem to have a reputation for data breaches and I think participants use a physical calculator. The quality of the test as a measure of aptitude is a different matter, but presumably Oxford have decided that if it’s good enough for Cambridge and Imperial, then it’s good enough for them.

ESAT does not allow a calculator.

On the whole I support the move. I wish the test was different but I agree it is for the universities to decide.

As you can see, I’m not a STEM teacher. I was just reporting what kids say in a casual setting. I’ve not heard any moan about how tough the tears are. They know to expect that. But I do know the built in calculator not responding reliably has caused some feelings of unfairness. Pearson is big enough to have a smooth interface.

They are quite different. Cambridge interviews much of the applicants, Imperial made much more offers (35% took offers 2024). Screen out 67% applicants by Oxford using EAST probably too harsh)

Your pupils’ reaction is important, @Genevieva

That level of screening based on one Pearson test requires that everything run smoothly if it is to be fair and useful, @DEI2025 Pearson’s record is not encouraging.

Yes, the interesting question is going to be whether Oxford increases the number it invites to interview, given that ESAT is no PAT. They claim they will base their decision on the whole application and not have an ESAT cut off score, but this first year will be a bit of a guinea pig year.

It is important to take into consideration that competitive admissions tests are not infallible and that it is not necessarily those who score most highly on competitive tests at age 17 who will perform best four years later in their degrees. There seems to be a strong attachment on this thread to the idea of ESAT being less demanding than PAT and that therefore Oxford entrance standards are being lowered. I dispute this, conceptually. Many applicants to Oxford will have been educated in systems that are unfamiliar to admissions (who, despite all their protests to the contrary, don't fully grasp any high school leaving exam systems beyond A-levels, IB and, perhaps, APs). If the UK's highest ranking universities want to attract and recruit the very best students from all over the world, then they need a standardized test with which to compare them as fairly as possible as a basic filter and that, apparently, is the goal of the new TMUA/ESAT/TARA formats.

As you said earlier, @Ceramiq , time will tell.

You’ve inserted a key caveat: If Oxford wants …. and we will need to see which way this goes. Imperial took a deliberate, financially based decision to recruit a highly international student body. Oxford’s wealth means it doesn’t need to do that; and at the UG level it will need to figure out what it wants to do.

(Yes, we all know the official line, but we also know that rightly or wrongly a significant portion of highly able Home candidates are put off Imperial by the international nature of the student body, and some Home students find it difficult)

This circles back to the role of state funding in UG education. The greater the funding, the more legitimate I think it is to reserve a portion of places for Home students. Right now that funding model is broken, so the legitimacy is questionable. It will be interesting to see which way Oxford goes.

I agree a good exam suitable for candidates from different backgrounds would be an ideal filter. If ESAT does what Oxford wants, that’s fine.

Pearson has messed up enough over the last few years that yes, I wish this were administered differently, but that is a separate question.