Independent preparation for the numerical reasoning test in the Wizz Air Pilot Academy selection process. A data panel — a table or a chart — and one sentence claiming something about it. You decide: true, false, or cannot say. Twenty questions, ten minutes, calculator on screen.
Every question shows a data panel — a table or a chart — and one sentence claiming something about it. Three answers sit under it in the same order every time: True, False, Cannot say. That third option is what separates candidates: in about a quarter of questions the panel simply does not settle the claim, and any arithmetic you do on it is beside the point. Twenty questions in ten minutes, with a calculator on screen.
Nothing here is harder than the four operations and a percentage, and there is a calculator on screen. What is actually measured is whether you can find the right number quickly, pick the right operation, and — above all — notice when the panel does not answer the question at all. Ten families produce every question, four claim forms tell you how exact to be, and one answer in four is “cannot say”. The five steps below are the order in which they are worth learning.
Every question is the same shape. A data panel sits at the top — a table about six times in ten, otherwise a bar, line, stacked or pie chart — and under it one sentence claims something about it. Three answers follow in the same order every time: True, False, Cannot say, selectable with A, B and C. Panels run from three to six rows and two to five columns, so the most you ever scan is thirty cells, and the panel numbers follow the language you choose — 1,234.5 in English, 1.234,5 in Turkish. Read the title first — it names the scenario and, crucially, the measure. A calculator is on screen, Alt+H opens it.
None of this is difficult arithmetic: the four operations and a percentage, with a four-function calculator on screen. What is measured is something else — whether you can find the right figure in an unfamiliar panel, choose the right operation, and recognise the questions where the panel does not answer the claim at all. Twenty questions in ten minutes is thirty seconds each, and a blank scores exactly what a wrong answer does.
The third option is the whole test. In one question in four the panel does not settle the sentence, and the answer is Cannot say — not because the arithmetic is hard, but because the sentence names a category, a period or a measure the panel never carried. Reading evidence for what it does not establish is the skill being scored.
Panels run three to six rows by two to five columns, so at most thirty cells, and they arrive in five shapes: a table six times in ten, otherwise a grouped bar, a line, a stacked bar or a pie. The title carries the scenario and the measure — and the title is the line candidates skip.
A four-function calculator is always there — the button, or Alt+H — so nothing is testing your long division. What is tested is picking the operation: a share divides by the total, a percent change divides by the earlier period, and an average over a row is not the average of two of its cells.
In the timed modes you can move freely between the twenty questions and flag any of them, and a numbered strip shows what is answered and what is still open. Run a first pass, come back to the flags, and finish with a sweep that leaves nothing empty — an unanswered question scores exactly what a wrong one does.
Every Numerical Reasoning session you finish is broken down automatically. See more than the right–wrong ratio: which of the ten families cost you, whether the “cannot say” questions are where you lose them, whether tables or charts slow you down, and where in the ten minutes the time actually went.
Every question’s duration and your median against the thirty-second budget, so you can see where the ten minutes went — and whether one stubborn panel ate three easy ones.
How many questions each of the ten families produced and how many you got right. Comparison, total and share are 56% of the test between them, so a weakness there is worth far more than one in a family you may not even meet.
Your accuracy split three ways, which is the number that actually diagnoses this test. Someone at 90% on true/false and 40% on “cannot say” does not have an arithmetic problem — they have a reading problem, and it is worth a quarter of the paper.
Every question can be reopened afterwards with the correct answer and the reason written out as a sentence — and for a “cannot say” it names which of the three things the panel was missing. Finish the test, then read the review; that is where the learning is.
This module has just been published. Verified reviews from CADETPRO students appear here automatically as they come in — we do not write them ourselves and we do not display placeholder ratings.
✓CADETPRO course data — real student reviews, read live from the course record. Names shortened for privacy.
Each question shows a data panel and one sentence claiming something about it. The panel is a table about six times in ten, otherwise a grouped bar, a line, a stacked bar or a pie, and it runs three to six rows by two to five columns. Three answers sit under it in the same order every time — True, False, Cannot say, chosen with A, B and C — and exactly one is right. Under exam conditions there are twenty questions in ten minutes, with a four-function calculator on screen, free navigation and flagging.
Because one answer in four is Cannot say, and those questions do not look hard: the sentence asks for familiar arithmetic and the panel is full of numbers, so you calculate something and the something is beside the point. The panel can fall short in exactly three ways — the sentence names a measure the panel does not carry, which is half of them, a category it does not have, or a period it does not have. Finding the sentence’s category, period and measure in the panel before doing any arithmetic ends a quarter of the questions in five seconds.
CADETPRO ships a module for each of them and they ask for different things. Deductive logical reasoning fills a cell so a grid obeys stated constraints. Inductive logical thinking and inductive logical reasoning classify figures against a hidden rule. Logical thinking asks which one object of nine breaks a rule. Abstract logical reasoning continues a transformation across a row of panels. Diagrammatic reasoning gives you a matrix with a hole in it. This is the only one of the seven built on numbers: a data panel, a claim about it, and a verdict on whether the panel supports the claim.
Yes. The player and the walkthrough panels load the same module you get after purchase, straight from hub.cadetpro.com. The free EASY and MEDIUM modes are open without buying anything, and both write out the reason after every question — including, for a Cannot say, which of the three things the panel was missing. That is the fastest way to learn where the quarter hides.
EASY is free: 6 questions, no time limit, with the answer explained after each one. MEDIUM is free too: 12 questions, still untimed, with the same explanations. HARD and EXAM are PRO: 20 questions in 10:00, with free navigation, flagging and a numbered strip showing what is answered and what is still open. HARD and EXAM are identical in content — only EXAM feeds the leaderboard, the badges and your best score, so HARD is a full dress rehearsal that cannot spoil your statistics.
Nothing beyond the four operations and a percentage, and a four-function calculator is always on screen — the button, or Alt+H; Esc closes it and C clears it. Because the digit keys belong to the calculator at all times, the answers are chosen with A, B and C or the up and down arrows, while left and right page between questions. Two questions in five need no arithmetic at all: the claim is qualitative — higher, lower, highest — and only asks you to compare.
The result screen shows the raw number correct and a performance percentage, and the percentage is not linear. Over the twenty questions, sixteen correct scores 76, twelve scores 51 and ten scores 38, while a single correct answer scores 10 because there is a floor. Speed does not raise the score, a blank counts exactly as a wrong answer, and any test with a mistake in it is capped at 98.
No. CADETPRO is an independent preparation platform and has no affiliation with, endorsement from or partnership with Wizz Air or any assessment provider. The module reproduces the publicly described format of the test so you can practise it; it is not the assessment itself, and no result here guarantees any outcome in a real selection process.
Create your account to continue securely to payment.
