CEnT-S Scoring Explained: Marking, Normalised Score & Admission
By Velacai · June 25, 2026 · 14 min read
How CEnT-S scoring and admission actually work
CEnT-S scoring awards +1 point for each correct answer, 0 for a blank, and −0.25 for each wrong answer across 55 questions — but the number that goes on your result is a normalised score, not the raw total out of 55. Each university then converts that normalised score into its own ranking metric or minimum cut-off, with program-specific weighting. So the two things that decide your outcome are: how points are earned on test day, and how your university translates the result into a seat.
The CEnT-S — the CISIA English Test – Sciences — launched on 4 November 2025 and replaced the old English-language TOLC tests (English TOLC-I, TOLC-E and TOLC-F), unifying admission to English-taught bachelor's (laurea) programmes at Italian public universities. If you're new to the exam, start with our complete CEnT-S exam guide, and pair this article with the format and syllabus breakdown to see exactly which questions earn those points.
The CEnT-S marking scheme
The CEnT-S is 55 multiple-choice questions in 110 minutes, split into five independently timed sections, each question with five answer options and exactly one correct answer. The marking rule is identical for every question, in every section, regardless of subject or difficulty:
| Outcome | Points |
|---|---|
| Correct answer | +1 |
| Blank (no answer) | 0 |
| Wrong answer | −0.25 |
| Raw maximum (illustrative) | 55 (55 × 1) |
A few consequences fall straight out of this table:
- There is no partial credit — every question carries the same single point, whether it's a Physics item or a Reasoning-on-texts item.
- A wrong answer doesn't just forfeit the point you missed; it actively subtracts 0.25. The swing between a correct and an incorrect answer is therefore 1.25 points.
- Leaving a question blank is "free" — it never lowers your score. That single rule sits at the centre of every guessing decision (and, as we'll see, the math says you should rarely use it).
The CEnT-S is not adaptive and shares none of the Digital SAT's machinery — there's no grid-in/student-produced response, no on-screen calculator app, and no Bluebook-style interface. You simply select one of five options or leave it blank. One important structural note: the test is per-section timed, so you cannot return to a previous section or carry unused minutes forward. That changes when you make your guessing decisions — you must resolve every question within its own section's clock.
Raw score vs normalised score: the distinction that matters
Here's the part most candidates get wrong. The raw maximum is 55, but the official result is a normalised score, not your raw total. Do not fixate on "55."
CISIA applies a per-section normalisation coefficient that adjusts for how difficult each section was on the specific date you sat it. Because the test is offered across multiple dates and macro-periods, two candidates with the same raw count of correct answers could sit slightly harder or easier versions; normalisation exists to make results comparable across dates so nobody is advantaged or penalised by the luck of their test sitting.
In practice:
- You answer questions and accumulate a raw section score under the +1 / 0 / −0.25 rule.
- Each section's raw score is converted using that section's normalisation coefficient for your date.
- Your final normalised CEnT-S score is the sum of the normalised section scores.
So when you see your result, don't compare it to a mental "out of 55" benchmark, and don't directly compare a friend's raw count to yours across different dates. The normalised score is the official, comparable figure — and it's the number universities work from when they build rankings and cut-offs. (Note that because normalisation depends on the whole macro-period's results, your normalised score is published only after the macro-period closes.)
Why the penalty changes your guessing strategy
Negative marking is the single most misunderstood part of the CEnT-S. Many students hear "wrong answers lose points" and conclude they should leave anything uncertain blank. On the CEnT-S, that instinct is usually wrong — and the math shows exactly why.
The expected-value math
"Expected value" is just the average points a decision earns if you repeated it many times. With five options and the +1 / −0.25 rule, a fully random guess has a 1-in-5 chance of being right and a 4-in-5 chance of being wrong:
A blind, completely random guess is worth exactly 0 — precisely break-even, the same as leaving the question blank. Unlike heavily negatively-marked exams (where random guessing carries a real cost), the CEnT-S is calibrated so pure chance neither helps nor hurts you on average.
That's the floor. The moment you can eliminate even one option, the picture turns positive. With four plausible options left and a 1-in-4 chance of being right:
Now the expected value is positive. Eliminate a second option and you're choosing among three:
The more you can rule out, the more guessing pays.
The takeaway
| Situation | Expected value of guessing | What to do |
|---|---|---|
| No idea, can't eliminate anything | Guess or blank — identical on average | |
| Can eliminate one option | Guess | |
| Can eliminate two options | Definitely guess | |
| Down to two options | Always guess |
The rule to memorise: on the CEnT-S, guessing rarely hurts and usually helps. Because a fully random guess is exactly break-even — costing nothing on average — and any elimination tips the odds in your favour, there is almost never a good reason to leave a question blank. The mistake to avoid is the opposite of the IMAT instinct: don't over-skip out of fear. If you can rule out even one option, commit to an answer; if you genuinely can't, guessing still costs you nothing.
A word of caution: this is average behaviour over many questions. On any single guess you might lose 0.25, but the strategy wins across a whole paper. And remember the per-section timing — you must make these calls inside each section's clock, so practising the elimination reflex under realistic timed conditions is what makes it automatic on test day. You can build that instinct with full-length realistic CEnT-S practice that uses the exact five-section timing and +1 / −0.25 penalty scoring, so your mock results behave like the real thing.
How universities use your CEnT-S score for admission
Scoring decides your number; the university decides whether that number gets you a seat. This is where the CEnT-S becomes a competition rather than a pass/fail test — and where the rules stop being uniform.
Each Italian public university converts or thresholds your normalised CEnT-S score on its own terms. There is no single national cut-off and no fixed "passing score." Depending on the programme, your score may be used to:
- rank candidates and admit the top scorers until seats run out, or
- apply a minimum cut-off that you must clear to be eligible, or
- a combination of both, often blended with other admission criteria.
Crucially, weighting is program-specific. A university can choose to emphasise the sections most relevant to its degree and de-emphasise (or even exclude) others.
Example: Politecnico di Milano
For its English-taught engineering programmes, Politecnico di Milano accepts the CEnT-S as an alternative admission test and converts the result to a score out of 100. For engineering admission it excludes Biology and Chemistry from that conversion, effectively weighting the Mathematics, Reasoning-on-texts-and-data, and Physics sections. The lesson generalises: a section that's worth a point on test day might be worth zero in a specific programme's conversion — so the sections that matter most depend entirely on where you apply.
Other confirmed CEnT-S universities include the University of Bologna, University of Milano-Bicocca, University of Brescia, and other major Italian public universities running English-taught STEM and economics programmes. Each sets its own conversion, weighting, and thresholds.
| Aspect | What's fixed (CISIA) | What varies (per university / programme) |
|---|---|---|
| Marking rule | +1 / 0 / −0.25, five options | — |
| Score reported | Normalised score (sum of normalised sections) | — |
| Conversion of score | — | e.g. PoliMi converts to /100 |
| Section weighting | — | e.g. PoliMi excludes Biology & Chemistry for engineering |
| Cut-off / threshold | — | Program-specific, changes each cycle |
| @HOME vs @UNI eligibility | Both modes exist | EU/non-EU acceptance can vary by university |
Treat any cut-off score or weighting you read online as illustrative. They are program- and field-specific and change every cycle. Always verify against the specific programme's admission page for the year you're applying.
Retakes and result validity
You can sit the CEnT-S once per macro-period. CISIA runs roughly four macro-periods per year (approximately Nov–Jan, Feb–Mar, Apr–Jun, and Sep–Oct), so you can take it up to about four times in a cycle, choosing your best result. Each sitting costs €55 (for the Nov 2025–Oct 2026 window) and can be taken as CEnT@UNI (in a university computer lab on provided devices) or CEnT@HOME (remote, supervised).
A CEnT-S result is valid for enrolment in academic year 2026/27 — at least until 31 October 2026. Because you can retake across macro-periods and submit your best score, an early sitting also serves as a high-stakes diagnostic: if the result is below your target programme's expectations, you have later macro-periods to improve.
One reassurance on the broader landscape: the 2025 Italian "semestre filtro" reform — which abolished the entrance test for Italian-language Medicine — does not affect the CEnT-S or CISIA science/engineering admissions. The CEnT-S is fully active for AY 2026/27.
Turning scoring knowledge into a better outcome
Knowing the marking scheme doesn't raise your score by itself, but it sharpens how you prepare and how you behave on test day:
- Stop leaving blanks. Because a random guess is break-even and any elimination is positive, fill in every answer you can't confidently skip. The fear-driven blank is the most common avoidable loss.
- Protect your confident answers. Each correct answer is worth 1.25 points more than a wrong one, so careless errors on questions you actually know are the most expensive mistakes you can make.
- Aim at the right sections. Find out how your target university converts the score — if you're applying to PoliMi engineering, Math/Reasoning/Physics carry the weight and Biology/Chemistry don't, which should shape your study time.
- Simulate the per-section clock. Timed, full-length mocks with the real five-section structure and penalty scoring are the only way to calibrate both pacing and your elimination instinct.
For the bigger picture — identity, format, universities, dates and logistics — return to the complete CEnT-S exam guide. To understand exactly what changed in November 2025, read CEnT-S vs English TOLC. And when you're ready to commit to structured prep, compare options on the pricing page.
FAQ
How is the CEnT-S scored?
Each correct answer earns +1 point, a blank earns 0, and a wrong answer costs −0.25, across 55 questions in five sections. The raw maximum is 55, but the official result is a normalised score — CISIA applies a per-section normalisation coefficient (adjusting for difficulty across test dates) and sums the normalised section scores. Don't fixate on "out of 55"; the normalised figure is what universities use.
What is the difference between a raw and a normalised CEnT-S score?
Your raw score is the straight +1 / 0 / −0.25 total. Your normalised score adjusts each section using a coefficient that accounts for how hard that section was on your specific date, then sums the normalised sections. Normalisation makes results comparable across different test dates, so two candidates aren't advantaged or penalised by the version they happened to sit. The normalised score is the official one.
Should I guess on the CEnT-S or leave questions blank?
Guess in almost every case. With five options and the +1 / −0.25 rule, a fully random guess has an expected value of exactly 0 () — the same as a blank. Eliminate one option and it rises to +0.0625; eliminate two and it's about +0.167. Because guessing never costs you on average and any elimination helps, there's rarely a reason to leave a question blank.
How do Italian universities use the CEnT-S score for admission?
Each university converts or thresholds your normalised score on its own terms — for ranking, a minimum cut-off, or both — with program-specific weighting. For example, Politecnico di Milano converts the result to a score out of 100 and excludes Biology and Chemistry for engineering admission. Cut-offs and weightings vary by university and field and change each cycle, so always check the specific programme's admission page.
How many times can I take the CEnT-S, and how long is it valid?
You can sit it once per macro-period, and CISIA runs about four macro-periods per year (≈ Nov–Jan, Feb–Mar, Apr–Jun, Sep–Oct), so up to roughly four sittings per cycle — you keep your best result. A result is valid for enrolment in AY 2026/27, at least until 31 October 2026, and you can take it as CEnT@UNI (in a lab) or CEnT@HOME (remote, supervised) for €55.