LSE TMUA: compulsory for two courses, recommended for nine

LSE splits its eleven TMUA programmes into two groups, and the split matters more than anything else on this page.

Two are mandatory. LSE puts it as "taking the TMUA is mandatory, all applicants are required to take the test."

Here's how that breaks down:

If you are applying to either, the test is not a decision. Not sitting it leaves the application incomplete.

For these, LSE says the test is "recommended but not mandatory", and that "applicants are encouraged to take the test and a good score may make an application more competitive."

Here's how that breaks down:

Note that the two Actuarial Science entries share a single UCAS code, N321, which explains an oddity on the LSE page covered further down.

The full seven-university course list is in our guide to which universities require the TMUA.

This is the question the page exists to answer, and it can be answered with figures rather than with opinion.

LSE has released admissions data under the Freedom of Information Act. The response below, published in November 2025, covers three of the nine recommended programmes for the 2024/25 cycle.

Here's how that breaks down:

Two conclusions follow, and they pull in opposite directions.

First, recommended really does mean optional. Between roughly a quarter and 43% of offers on these three programmes went to applicants who had no TMUA score at all.

On Financial Mathematics and Statistics, 27 of 63 offers went to candidates who did not sit the test. You are not excluded from consideration by skipping it.

Second, the applicants who did sit it and received offers scored high. Means of 6.6 to 7.4, against a whole-cohort mean of 4.28 and a 90th percentile of 6.7 for 2026 entry.

A mediocre score is not obviously better than no score.

It is worth contrasting LSE with Durham, where a recommended test carries a published consequence. At Durham a score of 5.0 or above makes you automatically eligible for an offer one grade below the standard one.

LSE publishes no equivalent mechanic, so a Durham applicant can calculate the value of sitting and an LSE applicant cannot. See Durham and UCL TMUA requirements for how that works.

One oddity worth knowing about, because it decides whether your course is covered.

The LSE TMUA page introduces the recommended group as "these eight programmes" and then names nine.

The likely explanation is in the table above:

BSc Actuarial Science and BSc Actuarial Science (with a placement year) share the UCAS code N321, so LSE is probably counting them as one programme with two entry routes.

We cannot confirm that on behalf of LSE, so treat it as the probable reading. It makes no practical difference either way: the test is recommended for all nine of the named programmes, and the list above is the one to check your course against.

LSE publishes no threshold, no minimum and no target, so any figure of the form "you need 6.5 for LSE" is unsourced. There is nothing on an LSE page to source it to.

The FOI release is the only real evidence, so it is worth pushing further than the offer counts above.

Three things come out of that, and the third is the one that matters most.

1. Offer holders who sat the test scored high. Means of 6.6 to 7.4, against a whole-cohort mean of 4.28 and a 90th percentile of 6.7 for 2026 entry. On two of the three programmes the average offer holder sat above the 90th percentile of everyone who took the test.

2. Those are recommended courses, and the figures run higher than the Cambridge compulsory one. Cambridge Economics offer holders averaged 5.99 in a comparable cycle. These LSE recommended programmes come in a point above that.

3. Which is the clue to reading the number correctly. When a test is optional, the people who sit it are the people who expect to do well, and the people who sit it and receive an offer are a narrower group again. The 7.4 describes a self-selected group of confident candidates rather than a standard LSE applies.

For scale rather than as a target, the whole 2026 entry cohort had a mean of 4.28, a median of 4.1, a 75th percentile of 5.3 and a 90th percentile of 6.7, across 16,092 candidates.

LSE is in the majority in publishing nothing. Of the seven TMUA universities, only Warwick and Durham publish anything resembling a figure, and only Imperial publishes distributions of what applicants actually scored.

For the benchmark question across every university, see what is a good TMUA score, and for how the scale itself works, TMUA scoring and results.

LSE accepts either window, and says so explicitly, but adds a practical preference: applicants are "strongly encouraged to take the first sitting as this will allow the widest availability of time slots and locations."

There is a second reason to think about October. Across two consecutive cycles the January cohort has scored roughly 0.9 points lower on an October-anchored scale.

That group self-selects, so it is not evidence that the January paper is harder, but it does mean the extra term is not obviously free.

Full dates and deadlines are in TMUA dates, registration and booking.

Automatically, and there is nothing for you to do after the test.

UAT-UK and Pearson VUE match your score to your UCAS application, course by course. You cannot send it selectively, you cannot withhold it, and there is no appeal against the result.

The two groups face genuinely different decisions.

If your course is one of the compulsory two, there is no decision. Prepare for it the way you would prepare for a compulsory test anywhere, and skip to the resources below.

If your course is one of the recommended nine, the question is whether to sit at all, and the honest input to that is evidence rather than nerve. Three steps.

1. Sit a real paper under time before you book. The whole archive is free. One timed attempt tells you more about whether the test will help your application than any amount of reading about it.

2. Compare against the cohort, not against your class. The 2026 entry median was 4.1 and the 90th percentile 6.7. LSE offer holders on the three programmes covered by the FOI averaged 6.6 to 7.4.

3. Decide once, for all five choices. Because matching is automatic, you cannot show the score to LSE and hide it from Warwick. If any of your five courses makes it compulsory, the decision is already made.

For the material, start with the official TMUA past papers from 2016 to 2023. They are free, every released paper comes with worked answers from UAT-UK, and the content specification has not changed since.

That is eight sittings and 320 questions.

Read our guide to Paper 2 first if your time is limited, since it covers the logic and proof material that A-level does not teach, and the complete TMUA guide if you are starting from scratch.