Does the SAT Penalise Wrong Answers?

The rule, and when it changed

The SAT once deducted a quarter of a mark for a wrong multiple-choice answer, which made guessing a genuine gamble and made "only answer what you know" sensible advice. That has not been true for a decade, and the advice outlived the rule by a long way — it still reaches students from older siblings, from teachers who sat the test themselves, and from prep material that was never updated.

The current position is unambiguous. Scoring counts correct answers and nothing else. College Board states it directly in its own guidance on the digital test: there is no penalty for guessing, and it is better to answer every question the best you can.

What a blank actually costs

Treat it as expected value. A four-option multiple-choice question answered at random is worth 0.25 marks on average. Left blank it is worth exactly 0. Across ten questions a student runs out of time for, that is two or three raw marks handed back for nothing — and raw marks near the middle of the scale are worth roughly ten scaled points each.

Eliminating options improves it further. Cross out one clearly wrong answer and the guess is worth a third of a mark; cross out two and it is worth a half. This is why partial reading of a question is not wasted work: even a fragment of understanding converts into expected marks, provided the answer actually gets entered.

• Blank multiple-choice question — expected value 0
• Blind guess from four options — expected value 0.25
• Guess after eliminating one option — expected value 0.33
• Guess after eliminating two options — expected value 0.5

Why a blank in Module 1 costs twice

The Digital SAT is multistage adaptive. Each section has two modules, and performance on the first decides which version of the second you receive. A first module that goes badly routes a student to a second module with a different mix of difficulty, and the route taken shapes what the section score can reach.

So a blank in Module 1 does two things: it forfeits the mark, and it makes a lower-difficulty route marginally more likely. That is the strongest reason to enforce a no-blanks rule in the first module specifically — the cost is not one question, it is one question plus a slightly narrower ceiling for the rest of the section.

Turning this into a habit that survives the last minute

Knowing there is no penalty changes nothing unless the behaviour is rehearsed, because the failure is never a decision to leave something blank — it is running out of time with three questions unread. The habit that fixes it is mechanical: mark for review, move on, and reserve the final thirty seconds of every module for entering an answer on everything still empty.

In our Digital SAT groups this is checked on every timed practice, not discussed once. The Math section's type-in questions are the ones students most often abandon, and while a made-up numeric answer is close to worthless, close to worthless is still more than nothing and it costs three seconds to type.