What does my TARA score mean?
Critical Thinking
How this result is worked out
Published cut points (UAT-UK technical report, 2026 entry): 5.5 is between the 75th and 90th percentiles of all 2,626 candidates for 2026 entry (October 2025 and January 2026), and between the median and the 75th percentile of the 976 October 2025 candidates, the sitting the scale is anchored to. UAT-UK publishes only these percentiles, so no exact percentile is shown.
The published distribution has two clear concentrations, around and . A single mode would hide that shape.
Oxford: Economics and Management
Sitting: October 2026 required.
How your course uses this score
This module result sits between the October median (4.5) and the October 90th percentile (7.0). That is a wide band, not a hidden course cutoff. Oxford publishes no course cutoff or offer-holder TARA benchmark.
Read both module scores. The Writing Task is separate, and its use depends on the university and course.
A past score describes one sitting. It does not predict your next one.
How this works
Two modules, two separate signals
UAT-UK reports Critical Thinking and Problem Solving separately. There is no official combined TARA score, so this tool never averages them.
The first published distributions are not smooth bell curves. Critical Thinking has visible concentrations around and , while Problem Solving peaks around . That is why the same scaled score can sit in a different published band in each module.
Why raw marks produce a range
UAT-UK does not publish a raw-mark conversion table. It uses a Rasch model to account for item and form difficulty after the sitting, so a practice-paper mark cannot be turned into an official score in advance.
So the raw-mark mode gives no scaled score or percentile. It shows your mark out of with UAT-UK's published measurement error for that module, which is all a practice mark can support.
Real items are not equally difficult, and live forms are equated, so the same practice mark can become different scaled scores. Treat the range as orientation for practice, not a result prediction.
There is no hidden safe score
UAT-UK publishes no pass mark or grade boundary. Oxford and UCL publish no TARA course cutoffs. UCL says the score sits alongside the rest of the UCAS application; Oxford requirements remain course-specific.
A high percentile is evidence of a strong relative test result. It is not an offer forecast. A lower percentile is a weaker signal, not an automatic rejection.
The official numbers: all 2026-entry candidates
UAT-UK's first technical report covers all 2,626 candidates for 2026 entry. In Critical Thinking the median was 4.0, the 75th percentile 5.4 and the 90th 6.7. In Problem Solving the median was 4.7, the 75th percentile 5.8 and the 90th 7.0. These describe the whole cohort, not the level any course needs.
| Module and sitting | Candidates | Mean | 25th | Median | 75th | 90th |
|---|---|---|---|---|---|---|
| Critical Thinking, All | 2,626 | 4.21 | 2.8 | 4.0 | 5.4 | 6.7 |
| Critical Thinking, Oct 2025 | 976 | 4.46 | 2.9 | 4.5 | 5.8 | 7.0 |
| Critical Thinking, Jan 2026 | 1,650 | 4.06 | 2.8 | 4.0 | 5.2 | 6.6 |
| Problem Solving, All | 2,626 | 4.72 | 3.5 | 4.7 | 5.8 | 7.0 |
| Problem Solving, Oct 2025 | 976 | 4.66 | 3.5 | 4.5 | 5.8 | 7.0 |
| Problem Solving, Jan 2026 | 1,650 | 4.76 | 3.5 | 4.8 | 5.7 | 7.0 |
October is the anchor: UAT-UK fixes each cycle’s October median at 4.5 and 90th percentile at 7.0, so the January figures, and next year’s, can differ.
What a mock cannot predict
Not precisely. UAT-UK's report puts the standard error of a raw mark at about 2 marks out of 22, so the same person could score two marks higher or lower on another form, and live scores are equated to each cycle's October cohort after the sitting. A mock's scaled score is an estimate, never an official number.
- Raw-mark standard error: 1.81 to 2.12 marks in Critical Thinking and 1.92 to 2 in Problem Solving, depending on the form.
- Scaled-score standard error: about 0.87 to 1.24 in Critical Thinking and 0.94 to 0.98 in Problem Solving. A 5.2 and a 5.8 are not reliably different.
- Reliability was 0.59 to 0.79 across forms, below the 0.80 rule of thumb, which UAT-UK attributes to the short 22-question modules.
Where marks go: time, especially in Problem Solving
The median candidate used 39:49 of 40 minutes in Critical Thinking and all 40:00 in Problem Solving. 8% of candidates left at least one Problem Solving question unreached, against 1% in Critical Thinking, and 28% either left one unreached or answered one in under five seconds (4% in Critical Thinking). There is no penalty for a wrong answer, so never leave one blank.
Have a practice mark rather than a score? Review your mark and pacing instead of converting it: the review separates errors from questions you ran out of time for.
Next: a free drill on the module that cost you most
The free TARA reasoning quiz needs no account.