TARA practice test: free sample questions before test day

No — not yet, and not because anyone is holding one back. TARA (the Test for Admission to Reasoning and Academic study) had its first-ever sitting in October 2025, followed by a second in January 2026.

That's the entire history of the test so far.

There is no bank of past papers stretching back years, because there haven't been years — there have been two sittings, and UAT-UK has not published either one as a downloadable past paper the way some long-running exams eventually release retired material.

If you've spent time searching "tara test past paper" or "tara admissions test past papers" and come up empty, that's not a gap in your searching. It's an accurate reflection of where the test actually is in its life cycle.

The honest answer is short: there is currently no official TARA past paper archive, and anyone claiming to sell you one is selling you something else — most likely practice questions written to resemble the format, not real retired papers.

Most UK admissions tests that do have a past-paper archive earned it the slow way — by running for a decade or more before the exam board judged it safe to retire and publish old material without compromising future sittings.

The Thinking Skills Assessment (TSA), which Oxford and UCL used for many of the same courses before TARA replaced it, built up exactly that kind of archive over its lifetime. TARA hasn't had the years to do the same.

It was introduced for 2026/27 entry as part of a wider overhaul of Oxford's in-house admissions tests — confirmed directly on Oxford's own admissions tests page, and covered course-by-course in our TARA vs TSA vs BMAT comparison.

UCL joined the same UAT-UK testing partnership in 2025, as confirmed on UCL's own tests, tasks and interviews page — see our UCL TARA course list for exactly which 11 courses that covers.

So from a past-paper standpoint, TARA is closer to a brand-new test than a mature one, even though the skills it tests are not new at all.

That last point matters more than it might sound.

TARA's Critical Thinking module uses the same seven named question types, the same five-option multiple-choice format, and a near-identical worked-example teaching style to TSA's Critical Thinking component — a directly verifiable structural match, not a guess.

What changed between the two tests is the administration format:

TSA mixed 25 Problem Solving and 25 Critical Thinking questions into one 90-minute sitting ordered roughly by difficulty, where TARA splits the same underlying skills into two separate 40-minute, 22-question modules taken one after the other.

So if a school or family already has TSA-era Critical Thinking material, the reasoning skill being tested hasn't moved — only the shape of the exam sitting around it has. We cover exactly what carried across and what didn't in the TSA vs TARA guide.

None of that produces an actual TARA past paper, though, and it would be misleading to imply otherwise. Two real sittings is not enough material for UAT-UK to retire and publish a paper without risking the integrity of the next one.

Realistically, an official past-paper archive is still some years away.

Given there's no official archive, "practice test" and "mock test" searches for TARA are really asking for one of two things: either a full-length timed simulation of the real thing, or a set of individual questions written to the same seven Critical Thinking types and three Problem Solving types that UAT-UK's own TARA page describes in its specification.

Neither of those requires an official past paper to be useful — what it requires is that the questions are built faithfully to the real question types, timing, and difficulty, and that the reasoning behind each answer is actually explained rather than just marked right or wrong.

That's the honest positioning here, and it's worth being direct about it rather than dressing it up: a growing bank of practice questions, built to the real specification and explained properly, is currently the closest thing to a TARA past paper that exists.

It is not a lesser substitute you settle for because nothing better is available — question quality and explanation depth matter more to actual score improvement than whether a question happened to appear on a real sitting, particularly this early in a test's life when even the two real sittings that exist aren't published for anyone to practice against anyway.

What follows is not a hypothetical description of that approach — it's two real questions from the practice bank, presented exactly as a registered user would see them.

If that sounds better in theory than it does on a page, it's worth just seeing directly — you can start working through the practice bank free before you decide whether to go further. Two real questions from it are right below, no account required to try them.

Both questions below are pulled directly from the GhostPrep TARA practice bank — the same bank a registered account gets full access to — condensed here so you can try them before creating an account.

Read the passage, pick an answer, then reveal the explanation.

The full step-by-step method and common-trap notes for every Critical Thinking question type live on our Critical Thinking guide; what's below is deliberately shortened to keep this page focused on the register-and-try decision.

Those are two of the seven Critical Thinking question types tested on the real exam — Identifying the Main Conclusion and Assessing the Impact of Additional Evidence.

The other five (Drawing a Conclusion, Identifying an Assumption, Detecting Reasoning Errors, Matching Arguments, and Applying Principles) each get a full worked example, method, and trap breakdown on the Critical Thinking guide.

A registered account gives you the rest of the bank beyond those seven examples, growing on an ongoing basis rather than capped at a fixed set.

Problem Solving is the second timed module, sat separately after Critical Thinking with no going back:

22 multiple-choice questions in 40 minutes, scored on its own 1.0–9.0 scale rather than combined with the Critical Thinking score.

UAT-UK's specification, laid out on its own TARA page, names three question types within it — Relevant Selection, Finding Procedures, and Identifying Similarity — and calculators and dictionaries are banned throughout, the same as in Critical Thinking.

There's no negative marking here either, so an educated guess on a question you're unsure of costs nothing.

The maths the module actually draws on is deliberately non-specialist.

UAT-UK's own Problem Solving appendix covers simple fractions, place value, percentages, the four rules of number, percentage operations, calculating an average (mean), time/money/measures, unit relationships within the metric system (1km = 1000m, 1m = 100cm, 1cm = 10mm, 1kg = 1000g — no imperial conversions), the area, perimeter, and volume of a rectangle or box, and reading information out of graphs, charts, and tables.

None of it goes beyond roughly GCSE-level content; the difficulty sits in how the numbers are wrapped inside a problem you have to work out how to attack, not in the arithmetic itself.

For the full Problem Solving picture — all three question types worked through with real bank questions, tables and all, plus the complete maths appendix — head to the Problem Solving guide.

A registered account gives you practice questions across the module as the bank grows, and if you'd rather see one of these question types in action than keep reading about them, register free and try one yourself.

If you've compared options for TARA preparation, you've likely come across exams.ninja's TARA.Ninja platform, a paid product built specifically around TARA practice and mock testing.

It's a genuine option worth knowing about, and this isn't an attempt to talk it down — the fairer way to put the choice is what each approach actually optimises for.

A paid-first platform is built around the assumption that you commit to a subscription before you see much of what's inside it, which can work well if you already know TARA prep is worth paying for and just want to pick a provider.

GhostPrep is built around the opposite assumption: that you should be able to try genuine, real practice questions — not marketing samples — before deciding whether to pay for anything at all.

That's the entire reason the two questions above are sitting on this page in full, revealable answers included, rather than gated behind a signup wall.

The honest differentiator isn't that one platform is good and the other isn't. It's breadth and cost structure.

GhostPrep's practice bank is free to start and growing on an ongoing basis, with a paid tier only if you want the full library beyond the free portion.

A paid platform's core value is committed upfront, in exchange for content that may already be more built out on day one.

Which matters more to you depends on whether you'd rather test the quality of the explanations first, or you're confident enough in a provider already that you're happy to commit before trying.

Here's how that breaks down:

That last row matters regardless of which platform you're comparing: no provider, free or paid, can currently offer you a genuine TARA past paper, because none exist. Be cautious of any product that implies otherwise.

Until a timed, full-length official mock exists, you can build a reasonably faithful simulation yourself using the real structure UAT-UK publishes and a working set of practice questions.

It's worth doing at least once before test day, because the module structure — three separately timed sections, taken in order, with no going back once a module ends — catches people out more than the question content does.

A useful routine looks like this. Set aside two hours without interruption.

Sit Critical Thinking first, strictly timed at 40 minutes for 22 questions, and stop the moment the clock runs out even if you haven't finished — that's what the real test will do to you.

Take a short break, then sit Problem Solving under the same conditions:

40 minutes, 22 questions, no calculator, no dictionary.

If you want to simulate the full sitting, finish with a 40-minute Writing Task under one of the three prompt types, capped at 750 words, even though it isn't separately marked the way the two MCQ modules are.

Afterwards, the part most people skip is the part that actually improves a score: going back through every question you got wrong, or guessed on, and working out why the correct answer is correct — not just noting that it was.

That's the specific gap a bare past paper wouldn't close even if one existed, because past papers alone don't explain their own answers.

A practice bank with reasoning built into every question closes it directly, which is the main reason we built ours that way rather than as a plain answer key.

Repeat that cycle a handful of times in the weeks before your sitting, rotating through fresh questions rather than redoing the same set, and treat your own timing discipline — not just your accuracy — as something to track improvement on.

Both matter separately: a candidate who gets every question right but runs out of time on six of them scores worse than one who is a little less accurate but finishes the module.