TARA Scores 2026 | What is a Good TARA Score?

The TARA is scored on a scale of 1.0 to 9.0 for the two multiple-choice modules. Your raw score out of 22 for each module is converted to a scale score using a complex statistical model. The Writing module is not scored by the TARA test administrator but is forwarded onto the universities you’re applying to.

Oxford and UCL applicants will want to know what a good TARA score is and how scoring works. How many raw marks do you need to achieve your target score?

In this guide, we explain what the score conversion process is for the TARA. We also examine the official results data on TARA score distributions, as well as what score you need for a competitive Oxford or UCL application.  

How Does TARA Scoring Work?

The TARA consists of three modules:

• Critical Thinking – 22 multiple-choice questions
• Problem Solving – 22 multiple-choice questions
• Writing – a choice of three essay questions, of which candidates must answer one

Candidates receive a separate score for each of the Critical Thinking and Problem Solving modules on a scale of 1.0 to 9.0, where 1.0 is lowest and 9.0 highest.

Every question in these modules has a single correct answer, worth one raw mark. Your raw score out of 22 for each module is then scaled to your final score between 1.0 and 9.0.

UAT-UK, which administers the TARA (as well as the ESAT and TMUA), does not mark the Writing question. Instead, it passes your answer directly to the university you’re applying to, who will read and review it but not issue you with an official mark. (This is similar to how the LNAT Section B is not marked by the LNAT itself but forwarded to your universities, who may use it however they wish.)

The Scaling Process

Why do you not just receive your raw score out of 22 for Critical Thinking and Problem Solving?

The reason is that not everyone who sits the TARA, even in the same test window, answers the same set of questions. The examiners write more than 22 questions for each module, and different candidates will answer a different combination of these questions.

As UAT-UK explain in their technical report on the design of their tests, “this can lead to small differences in difficulty” across different test versions, “despite the best attempts of the Chairs” to make them “comparable when they are put together”.

It would be unfair for a candidate who scored, say, 15/22 on an easier version of the test to receive the same final score as someone who scored 15/22 on a harder version of the test.

For this reason, TARA scores are scaled using a complex statistical model called the Rasch Item Response Theory (IRT) model. This model accounts for the difficulty of the questions each candidate answered and then places all candidates on a scale between 1.0 and 9.0.

In effect, to go back to our earlier example, this means the candidate who scored 15 on a hard test version will get a higher scale score than a candidate who scored 15 on an easier test version.

UAT-UK state that the scale is fixed so that a score of 7.0 represents the 90^th percentile in the October sitting, while a score of 4.5 represents the 50^th percentile in the October sitting.

This is the same process which UAT-UK use to score the ESAT and TMUA.

TARA Score Distributions

The following charts show the TARA score distributions for both multiple-choice modules in the most recent cycle. There are separate charts for the October 2025 test window and the combined 2025-26 application cycle (including both the October 2025 and January 2026 windows).

These charts are based on official TARA scores data for 2025-26 and October 2025.

Critical Thinking

2025-26

From this chart, we can see that the most common Critical Thinking score in 2025-26 was 5.0, with 11.9% of candidates achieving this, followed by 3.0 (11.3% of candidates). A high number got the minimum score of 1.0 (7.7%) compared to 1.8% who scored a 9.0.

The 50^th percentile score was 3.9, the 75^th was 5.2, and the 90^th was 6.4.

October 2025

There are some oddities to the Critical Thinking score distribution for the October 2025 test window. No candidates scored a 7.5, only 0.2% scored an 8.5, 1.3% a 5.5, and 1.6% a 7.0.

The most common scores were 4.5 (11.4% of candidates) and 6.0 (10.9% of candidates).

This ‘gappy’ score distribution may be due to the fact that the number of candidates taking the TARA in the October 2025 sitting was presumably low. Because in 2025-26, the TARA was only required by UCL, which has a January application deadline, there would have been little reason for candidates to take the TARA in October. With a small sample size, some irregular results may be expected.

The 50^th percentile score in October 2025 was 4.4, the 75^th was 5.7, and the 90^th was 6.7. This is a little odd if you recall that UAT-UK state that the 50^th percentile is supposed to be fixed to a score of 4.5, and the 90^th percentile to a score of 7.0. Again, it may be that low numbers of candidates in October 2025 resulted in these irregularities.

2025-26 vs October 2025

This chart shows the cumulative percentage of candidates achieving each score in the Critical Thinking module of the TARA in 2025-26 as a whole vs October 2025.

The two lines do differ fairly substantially especially lower down the scoring scale. Only 22% of candidates in 2025-26 achieved a score above 5.5 vs 31% of candidates in October 2025. This may be explained by the fact that applicants to Oxford and Cambridge are more likely to have taken the TARA in the October sitting (as the Oxbridge application deadlines are in mid-October). We would expect Oxbridge applicants to be stronger on average than non-Oxbridge applicants, hence why a higher percentage of them would achieve the top scores in the TARA.

Problem Solving

2025-26

In 2025-26, the Problem Solving module score distribution for the TARA shows a more conventional bell curve than the Critical Thinking module, with most candidates clustered around the middle of the scoring scale, albeit slightly tilted towards the mid-to-high range.

The most common scores were 5.5 (12.4% of candidates), 5.0 (11.2%), 4.5 (10.8%), and 4.0 (10.3%).

A notable difference from Critical Thinking is that the number of very low scores is much lower. Only 2.4% scored 1.0 in Problem Solving (vs 7.7% for Critical Thinking). 2.6% scored a perfect 9.0.

The 50^th percentile score was 4.5, the 75^th 5.6, and the 90^th 6.9.

October 2025

For October 2025, there is a similar bell curve score distribution for the Problem Solving module. The most common scores were 4.5 (12.1% of candidates), 4.0 (10.1% of candidates), and 5.0 (9.7% of candidates).

3.0% scored a 1.0 and 2.0% scored a 9.0. The 50^th percentile score was 4.4, the 75^th was 5.7, and the 90^th 6.7.

2025-26 vs October 2025

When we examine the cumulative TARA score distribution for the Problem Solving module, we see very little difference between the 2025-26 application cycle as a whole and the October 2025 sitting alone. The two lines track each other closely, unlike for Critical Thinking.

This suggests that there was not a notable difference in Problem Solving ability between candidates in the January and October sittings.

What is a Good TARA Score?

In short, a good TARA score is one that gets you accepted into your university and course of choice. The type of score that will advantage your application will vary depending on where you are applying and for which subject.

Given that Oxford’s acceptance rate is generally lower than UCL’s, we can expect that Oxford will usually require a higher TARA score.

Overall, the median TARA score will be 4.5 in the October sitting, and the 90^th percentile score will be 7.0 in October.

If the TARA follows the same patterns as the other UAT-UK tests, the TMUA and ESAT, the overall 50^th percentile and 90^th percentile scores for the application cycle as a whole will be slightly lower when the January sitting is factored in.

For 2025-26, this was as follows:

	Critical Thinking – Oct-25	Critical Thinking – 2025-26	Problem Solving – Oct-25	Problem Solving – 2025-26
50th percentile	4.1*	3.9	4.4*	4.5
75th percentile	5.8	5.2	5.7	5.6
90th percentile	6.8*	6.4	6.7*	6.9

*Scores data for October 2025 does not match the expected 4.5 = 50^th percentile and 7.0 = 90th percentile. As noted above, this may be due to low candidate numbers in the October sitting. UAT-UK state that 4.5 will be fixed to the 50^th percentile and 7.0 to the 90^th percentile in October 2026.

To score in the top 25% overall, candidates needed between 5.2 and 5.6 depending on the module in 2025-26. To score in the top 10%, this was between 6.4 and 6.9.

Note: do not forget that the Writing module, even though it is not scored by UAT-UK, represents a third of the TARA. A very strong Writing performance may make up for a weaker score in Critical Thinking and/or Problem Solving, while a candidate with solid Critical Thinking and Problem Solving scores may be dragged down by a poor answer in the Writing.

For UCL

According to official UCL data released in response to a freedom of information request, these are the average TARA scores of accepted applicants for selected courses in 2025-26:

Course	Average TARA Score
Computer Science BSc	4.4
Computer Science MEng	5.2
Computer Science and Mathematics MEng	5.8
Robotics and Artificial Intelligence MEng	8.2

We can see that there are major discrepancies between courses, with a TARA score of 4.4 (around the average of all test-takers) enough to be accepted for the average Computer Science BSc applicant. By contrast, the average score of accepted Robotics and Artificial Intelligence MEng applicants was an enormous 8.2, which would require a performance in the top 5% of all test-takers.

A good TARA score for UCL could be anything in the range of 4.4 to 8.2, depending on which course you are applying for. However, a score between 5.0 and 6.0 will likely place you in the top 20-40% of test takers. Given that UCL’s acceptance rate is 44% in 2025-26, scores in this range will strengthen applications for most courses.

For Oxford

Because Oxford is using the TARA for the first time in the 2026-27 application cycle, there is currently no data on scores of accepted applicants.

However, we can still make educated estimates of what scores Oxford may require. In short, a good TARA score for Oxford should be 6.0-6.5 or above, and 7.0 or better would represent a very good score. The only exception is applicants for Economics & Management. This is such a competitive course that a score of 8.0 or above may be needed to stand out.

Oxford applicants should consider the data on TARA scores for the October test window rather than the cycle as a whole, as they will all have to take the test in October. Based on this, a score of 7.0 will place you in the top 10% of Oxford applicants.

Two Oxford courses which require the TARA are currently among the courses with the lowest offer rates:

• Economics & Management – offer rate: 5.3%
• Psychology (Experimental) – offer rate: 11.6%

All the other TARA courses have offer rates above 11.6%. A score of 7.0 or above should therefore significantly strengthen your application for all courses except Economics & Management, where a score of 8.0 or better may be needed.

We can also make a comparison with the scores of accepted applicants for Cambridge for the other UAT-UK tests scored in the same way as the TARA – the ESAT and TMUA.

In 2025-26, the average accepted applicant to Cambridge scored:

• 6.06 in the ESAT
• 6.31 in the TMUA

Based on this, a score in the low 6s may be a typical TARA score for successful Oxford applicants.

TARA Score Conversion and Grade Boundaries

The TARA does not use score conversion in the traditional sense. There is no official table of raw scores for each module which equate to a final scale score.

This is because, as discussed above, the TARA uses the Rasch IRT model to produce its scale scores based on the difficulty of questions each candidate answers. The same raw score could therefore produce a different scale score depending on whether the questions were slightly easier or harder.

However, we can still get a rough idea of how many raw marks are needed for each scale score in the TARA. Another UAT-UK test, the TMUA, is now scored in the same way as the TARA. However, it did previously use simple score conversion.

Here are the score conversion tables for the TMUA for the last time they were used, in 2023.

Because the TMUA contained 40 questions, whereas each TARA module contains 22,  we have also added a column showing the equivalent score out of 22 for reference:

Raw Score	/22 Equivalent	Scale Score
0	0	1
1	1	1
2	1	1
3	2	1
4	2	1
5	3	1
6	3	1.5
7	4	1.9
8	4	2.4
9	5	2.8
10	6	3.2
11	6	3.5
12	7	3.9
13	7	4.2
14	8	4.5
15	8	4.8
16	9	5.1
17	9	5.4
18	10	5.7
19	10	6
20	11	6.2
21	12	6.5
22	12	6.6
23	13	6.7
24	13	6.8
25	14	6.9
26	14	7
27	15	7.1
28	15	7.2
29	16	7.3
30	17	7.4
31	17	7.6
32	18	7.7
33	18	7.8
34	19	8
35	19	8.2
36	20	8.4
37	20	8.6
38	21	9
39	21	9
40	22	9

• A score of 38/40 or above (equivalent to 21/22 or above) equated to a 9.0
• A score of 26/40 (equivalent to 14/22) equated to a 7.0
• A score of 14/40 (equivalent to 8/22) equated to a 4.5

Note: score conversion varied from year to year under the old TMUA scoring system. The TMUA is a different test from the TARA, and both tests are now scored under a different system.

However, this table may provide a very rough rule of thumb for TARA score conversion. You may only be able to afford to drop one or two marks to get a perfect 9.0, while a 7.0 may equate to getting roughly 60% of answers right. Once again, scaling will vary from module to module.

How is the TARA Writing Scored?

UAT-UK do not mark the TARA Writing module. Instead, they send your answer to the universities to which you are applying which use the TARA.

It is then up to these universities how they mark or otherwise use your Writing answer.

UCL

While UCL assess your essay qualitatively, they do not give it an official numerical mark or grade. Admissions tutors will read your essay and review it holistically alongside the other aspects of your application – your scores for the other TARA modules, your predicted or achieved grades, your personal statement, and your reference.

Oxford

Oxford have not announced how they will use the TARA essay. They may choose to follow the UCL model of reading your Writing answer and assessing it qualitatively without giving it a formal mark.

Alternatively, Oxford may give numerical marks to this module. They did so with the TSA (which the TARA is in effect replacing) and also do so with the LNAT essay.

The TSA essay was marked according to this official mark scheme:

A single mark on the scale 1–100 should be given, to reflect the overall quality of the essay, and specifically whether the candidate has:

i. addressed the question in the way demanded
ii. organised their thoughts clearly
iii. used clear, concise, compelling and correct English
iv. used general knowledge and opinions appropriately

Mark	Criteria
70–100	Clear construction, interesting and effective argument, excellent use of English.
65–69	A well-written and well-structured essay addressing the question as required.
60–64	A reasonably well-argued and appropriate answer. There may be some weaknesses in the use of English, the scope of the answer, or the coherence of the ideas.
50–59	A weak answer failing to address the question appropriately, or with significant weaknesses in argument, structure, or use of English.
49 or less	A very poor answer strongly suggesting that the candidate is not suitable.

The LNAT essay, meanwhile, is marked as follows:

LNAT essays are marked by comparative marking, rather than being given individual marks. This produces a ranking of essays and these are then assigned scores to reflect that ranking.

Essays ranked highest will display the following features “high mark features”, each of which is related to a general admissions criterion:

1. Application: close attention to the question(s) asked and sustained and focussed treatment of the issues.
2. Reasoning ability: well-drawn distinctions, a keen eye for relevance, awareness of more than one possible line of argument, and an element of independent critical judgment.
3. Communication: clear and fluent writing and notable clarity and appropriateness of structure and argument

Essays ranked lower will display the following features, “low mark features”, each of which is related to a general admissions criterion:

1. Application: poor attention to the question(s) asked, no sustained and focussed treatment of the issues.
2. Reasoning ability: poorly developed arguments, a preponderance of irrelevant points, few or no well-drawn distinctions, a lack of awareness of more than one possible line of argument, no evidence of independent critical judgment.
3. Communication: lack of fluency and clarity and no clear or appropriate structure or argument.

Get a TARA Tutor

Dukes Plus offers the highest standards of Oxbridge admissions test support for students preparing for their applications, including TARA tutoring.

Our packages of 1-1 teaching from subject-specialist, Oxbridge-graduate tutors are designed to enhance your performance – whatever your current level of preparation. Our students improve their scores by 20% on average after just four hours’ tutoring.

Book a TARA tutor today to enhance your preparation.

Sources

TARA Explanation of Results, 2025-26

TARA Explanation of Results, October 2025

ESAT Technical Report 2024-25

UCL TARA scores data

Oxford LNAT essay mark scheme

Oxford TSA essay mark scheme