Key takeaways
- A numerical reasoning test presents data in tables or charts and asks what follows from it. The arithmetic is deliberately ordinary; the difficulty is selecting the right computation and finishing inside a tight time budget.
- Most items reduce to four operations: percentage change, ratio, share of a total, and rate per unit. Recognising which one an item wants is the skill that separates scores.
- Time budgets are tighter than people expect. The UCAT's Quantitative Reasoning section gives 36 questions in 26 minutes, which is a little over 40 seconds per question including reading the table.
- Percentage growth and contribution to growth are different questions with different answers, and mixing them up is one of the most reliable ways to lose marks.
- Arithmetic fluency is the prerequisite, not the test. Practising calculation frees attention for interpretation, which is the part that is actually being examined.
A numerical reasoning test is a reading test with numbers in it. The arithmetic is deliberately ordinary — percentages, ratios, shares, rates — and the marks are lost somewhere else: deciding which quantity the question is actually asking about, locating it in a table, and finishing before the clock runs out. Candidates who drill calculation and still score poorly are usually solving the wrong problem quickly rather than the right problem slowly.
What the test is asking
ETS describes the quantitative material on its own General Test as covering arithmetic — "properties of integers, operations, exponents, roots, estimation, percentages, ratios, and sequences" — plus elementary algebra, geometry and data analysis, with data interpretation questions "grouped together and refer to the same table, graph or other data presentation". Employer tests follow the same shape at a smaller scale.
Two features make the format hard in a way the mathematics does not. First, several questions hang off one data presentation, so a misreading of the table contaminates a run of items. Second, the clock is tight. The UCAT's Quantitative Reasoning section allows 36 questions in 26 minutes — a little over 43 seconds per question, including the time spent reading. University careers services put the same point in plainer terms: assessors look for both speed and accuracy, and finishing every item is not always expected.
A worked item, and where the time actually goes
The table below reports invented quarterly revenue for three regions of a fictional company, in thousands of pounds, and is the kind of small data presentation a real item would hang three or four questions from.
| Region | Q1 | Q2 |
|---|---|---|
| North | 420 | 483 |
| South | 350 | 329 |
| East | 180 | 216 |
Question 1: which region grew fastest? Growth is a percentage change, so divide the change by the starting value. North: 483 − 420 = 63, and 63 ÷ 420 = 0.15, so 15%. South: 329 − 350 = −21, and 21 ÷ 350 = 0.06, so a 6% fall. East: 216 − 180 = 36, and 36 ÷ 180 = 0.20, so 20%. East grew fastest, at 20%.
Question 2: which region contributed most of the company's growth? Total Q1 is 420 + 350 + 180 = 950. Total Q2 is 483 + 329 + 216 = 1,028. The company grew by 78, so 78 ÷ 950 = 0.082, about 8.2%. Of that 78, North supplied 63, which is 63 ÷ 78 = 0.81, or 81% of the total increase.
Those two questions have different answers, and both are correct answers to their own question. East grew fastest in percentage terms while contributing less than half of North's absolute increase. Items are routinely written to exploit exactly this confusion, and no amount of faster multiplication protects you from it.
Notice where the seconds went. The three subtractions and three divisions take a few seconds each. Reading the table, deciding that "fastest growth" means percentage change from Q1 rather than the larger increase, and resisting the pull of the biggest number — that is the expensive part.
The four calculations that cover most items
Percentage change — the change divided by the original value, not by the new one. If a figure falls from 350 to 329 and then rises back to 350, the fall is 6% and the rise is 6.4%, because the base moved.
Ratio — comparing two quantities directly. In the table above, North to East in Q2 is 483 to 216, which is a little over 2.2 to 1.
Share of a total — a part divided by the whole. East's share of Q2 revenue is 216 ÷ 1,028, or about 21%.
Rate per unit — a quantity divided by some denominator, such as revenue per employee or cost per kilogram. Items that look complicated often just require you to build the right denominator.
Estimation beats exact calculation more often than candidates expect. If the options are far apart, rounding 483 to 480 and 420 to 420 gives 60 ÷ 420, which is about 14% — close enough to pick the answer and move on. The methods in estimation techniques explained are directly applicable, and the anchors in percentages in your head remove most of the calculation load from percentage items.
Why arithmetic practice alone does not fix the score
Work on the speed-accuracy trade-off describes the relationship as continuous rather than a switch: pushing for faster responses means less evidence has accumulated when you commit, so accuracy falls. The published framing is blunt — "Faster responses entail less accumulated evidence, and hence less informed decisions."
Applied to a numerical reasoning test, that has a specific consequence. If calculation is slow and effortful, you will be tempted to shave time off the interpretation step, which is precisely the step that decides whether you are computing the right thing. Faster arithmetic does not raise your interpretation skill; it buys you seconds to spend on it. That is a real gain, and it is a smaller gain than test-preparation marketing implies.
The rest of the exam behaviour — how to pace, when to abandon an item, and what a guess is worth under different marking schemes — is covered in strategy for a timed reasoning test. For the broader question of what any of these instruments claims to measure, what IQ tests actually measure is the place to start.
What arithmetic practice is genuinely for
IQ Test & Brain Training Games is our own app, built by Reign Creative. Its mental math practice drills the calculation layer, and its IQ-style test includes numerical reasoning among the areas it covers. The modes that map onto this material are specific: Percentage Trainer, a 75-second drill with Basics, Tips, Discounts, Advanced and Mixed settings; Ratio Trainer; Estimation Mode, a 90-second drill scored by how close your answer is rather than by exact matching; and Mixed Operations, which interleaves the four operations so you cannot settle into one procedure. The app records score, accuracy, streak, response time and personal bests for those runs.
Its listing does not describe data-table items of the kind above, and it says the IQ-style test's results are estimates based on app activities, not clinical or professional assessments, so neither the drills nor that test stands in for the real table-reading format. The honest claim is that fluent arithmetic is a prerequisite for numerical reasoning rather than a substitute for practising the real format. For a structured route through the methods themselves, the mental math training guide is the cornerstone of that subject.
More on item types, scoring and limits sits in the IQ tests and reasoning hub; the education and brain category collects the arithmetic and memory material; and the app itself is described on the IQ Test & Brain Training Games app page.
Good to know
Frequently asked questions
What is a numerical reasoning test?
How much time do you get per question?
What maths do I actually need?
Will practising mental arithmetic raise my numerical reasoning score?
Sources
- GRE General Test Quantitative Reasoning Overview (opens in a new tab)ETS — accessed
- Test Format and Scoring (opens in a new tab)UCAT Consortium — accessed
- Reasoning tests - numerical, verbal, logical and diagrammatical (opens in a new tab)University of Manchester Careers Service — accessed
- The speed-accuracy tradeoff: history, physiology, methodology, and behavior (opens in a new tab)Frontiers in Neuroscience — accessed
About the author
Reign Creative LLC — The independent app studio that builds the apps covered on this site and publishes every article here. On Google Play it publishes as Reign Collective Apps.
We use AI tools to help research, draft and edit our articles. Each article lists the sources its factual claims rely on, and details about our apps come from their live Google Play listings. No article on this site has been reviewed by an independent expert unless the reviewer is named on it.
Spotted an error? Tell us — how corrections are handled is set out in our editorial policy.