TMUA Scoring Explained: How Your Score is Calculated and What It Means

How TMUA Scoring Works

The TMUA uses a sophisticated scoring system based on Item Response Theory (IRT). Raw scores are transformed into scaled scores through a multi-step process to ensure fairness and comparability across different test forms and sittings.

The Rasch IRT Model

Item calibration is performed using the Rasch IRT model. Under this model, the probability of a candidate answering an item correctly is a function of the item difficulty and the candidate's ability. As a candidate's ability increases, the chance of correctly answering the item also increases. Item difficulty and candidate ability are expressed on the same metric, with greater values representing greater difficulty or ability respectively.

Equating and Scaling

Because multiple forms of the test are used, equating is performed to remove small differences in difficulty across forms, placing all candidates onto a single ability (theta) scale regardless of the form they took. The theta estimate is then scaled to generate an easily interpretable score. The TMUA is post-equated, meaning equating is conducted at the end of the testing window, so candidates do not receive an immediate score.

For the 2024-25 cycle, the median candidate theta was fixed to a scaled score of 4.5 and the 90th percentile ability was fixed to a scaled score of 7.0. A regression line between these two anchor points determines the scaling constants, and the resulting scaled scores range from 1.0 to 9.0, reported to one decimal place.

Scaled Score Statistics (2024-25)

The following table summarises the scaled score statistics for the 2024-25 admissions cycle:

What Scores Are Competitive?

Understanding what constitutes a competitive TMUA score requires looking at the percentile data. A score at the 90th percentile (6.5 for all candidates) means you outperformed 90% of test-takers. Key benchmarks:

• Above 6.5 — Top 10% of candidates; highly competitive for the most selective courses.
• 5.2 to 6.5 — Above average (75th-90th percentile); competitive for many courses.
• 4.0 to 5.2 — Around the median; may meet requirements for some courses.
• Below 4.0 — Below the median; less competitive for most courses.

October vs January Differences

There is a notable difference in score distributions between the October and January sittings. The October cohort has a higher mean score (4.58 vs 3.51) and higher percentiles across the board. This is partly because the University of Cambridge requires candidates to sit in October, and Cambridge applicants represent a particularly strong cohort. The January sitting tends to have lower average scores, but the same scaled score scale applies to both sittings, meaning scores are directly comparable.

No Penalty for Incorrect Answers

There is no penalty for incorrect answers on the TMUA. Candidates are advised to attempt every question. If unsure, it is always better to make an educated guess than to leave a question blank, as a wrong answer carries the same weight as an unanswered question (zero) but gives you a chance of gaining credit.