SAT Score Converter

What Is an SAT Score?

Two different numbers travel under that name, and the distance between them is the whole reason a converter exists.

Raw Score vs Scaled Score

College Board's own glossary separates them cleanly. A raw score is "a score on a test that is calculated by counting the number of correct answers"; a scale score is "a score obtained by transforming raw scores", used, in its words, "to facilitate interpretation". The count belongs to one paper. The scaled figure is built to mean the same thing across every paper.

• Raw mark — a count and nothing more. Twenty-eight correct is 28, whatever the questions were, whenever you sat them and however everyone else did.
• Scaled score — a position on a fixed 200 to 800 ladder, rebuilt for each test form so the same figure carries the same meaning from one administration to the next.
• Nothing is subtracted for a wrong answer. On the paper SAT each question contributed one point to the raw total and a blank scored the same as a wrong guess; on the digital test College Board still tells students that "for most students, it's better to guess than leave a question blank".

That last line matters before you use this page. Mark a practice section by counting correct answers only, subtract nothing for the ones you missed, and the number you are left with is the raw mark to type in.

The Two Sections and the 400 to 1600 Total

College Board reports three numbers for the SAT and no more: a Reading and Writing section score, a Math section score, and a total it describes as the "Sum of the 2 section scores".

Hold on to the doubling, because a section score and a total are easy to confuse and the converter will not catch the mistake. A converted 650 is a section result; two sections at 650 would be a 1300 total. Type 1300 into Scaled Score to Raw Score and the answer is 60, not a complaint — 1300 sits past the top of a 200 to 800 ladder, and everything past the top is held there.

A section score is one of two numbers an application file carries. The other is the grade average, which you can work out on the GPA Calculator.

The essay is a separate case again. Where it is still administered, as part of state in-school testing, College Board reports it on a 1 to 4 scale of its own and keeps it off the section scores entirely. Nothing on this page touches it.

Why the Menu Offers Three Sections

The SAT given between March 2005 and February 2016 reported three separate 200 to 800 section scores — Critical Reading, Math and Writing — and the three-way menu here is a survivor of that format. The 2016 redesign folded reading and writing into one Evidence-Based Reading and Writing section, and the digital SAT that replaced it reports the same pairing under the name Reading and Writing.

None of the arithmetic depends on which you pick. A raw 45 returns 650 on Math, 650 on Reading and 650 on Writing; only the Section row beneath the readout changes, from MATH to READING to WRITING. Treat it as a note to yourself about which paper you just marked.

Score Range and Percentile Are Not the Score

A real score report carries two further figures, and both describe uncertainty and company rather than performance.

• Score range — a band printed around each score, "derived from the standard error of measurement", showing "how much a student's scores would likely vary if they took a different administration of the test under identical conditions". College Board's own advice is blunt: "A student's ability is better represented by score ranges than points."
• Percentile — a comparison. A score in the 70th percentile means "70% of a comparison group achieved scores at or below" it. The All Tester Percentile is drawn from the past three cohorts of students who took the SAT in 12th grade, anywhere in the world.

Neither can be derived from a raw mark on its own. A score range needs the measurement error of the form you sat; a percentile needs the results of everyone who sat it. This page holds one table and one number, so one number is what it gives back.

How Do You Calculate an SAT Score?

Inside this page it is a single line of arithmetic, run in either direction. On the real test it is not arithmetic at all, and the last part of this section is about that gap.

The SAT Score Formula This Converter Uses

One straight line, built once per section and identical in all three:

Scaled = 200 + round(Raw ÷ 60 × 600)

Because 600 ÷ 60 is exactly 10, that reduces to:

Scaled = 200 + (10 × Raw)

Raw runs 0 to 60 in whole steps.
Scaled runs 200 to 800 in steps of 10.

The rounding in the first line never rounds anything. All 61 entries were checked one at a time against the code and every one lands on a whole ten — 200, 210, 220 and on up to 800 — so the second form is not an approximation of the first, it is the same table written more plainly. One correct answer is worth ten scaled points anywhere along it: raw 44 returns 640 and raw 45 returns 650.

Step by Step With This Converter

• Choose a conversion mode. Raw Score to Scaled Score is the one selected when the page loads.
• Choose a section. It labels the answer rather than changing it.
• Type one figure into Enter Score: a whole raw mark from 0 to 60 going forward, or a scaled score above 200 and up to 800 going back.
• Press Calculate. That box starts empty, so the panel holds a dash until you do.

Pressing Calculate on an empty box still computes something, and what it computes depends on the mode. Forward, a blank is read as 0 and the panel returns 200 with Input 0 underneath — a real-looking answer to a question nobody asked. Backward, a blank trips the floor rule and paints the amber notice instead.

Worked Example: 45 Raw Marks on Math

A student marks a 60-question practice section and counts 45 correct. Section set to Math, mode left on Raw Score to Scaled Score:

Scaled = 200 + round(45 ÷ 60 × 600)
       = 200 + round(450)
       = 200 + 450
       = 650

What the readout shows

650

Three rows follow it: Section MATH, Mode raw to scaled, Input 45. Only the headline is calculated — the other three are your own entries read back, so you can confirm what was converted before acting on it. Feed 650 into the reverse mode and 45 comes back, which holds for every multiple of ten on this line.

Running It Backward: Scaled Score to Raw Mark

Going the other way, the converter walks the same 61 entries from the bottom and stops at the first scaled score that reaches or passes what you typed. On the grid the answer is exact. Off the grid it rounds up — up, not to the nearest mark:

Raw = the first entry whose scaled value reaches the number you typed

640     ->  44     640 reaches 640
641     ->  45     640 falls short; 650 is the first that reaches it
650     ->  45     exact
655     ->  46     650 falls short; 660 is the first that reaches it
641.5   ->  45     decimals are accepted on this side of the conversion

Read the answer as the smallest raw mark that would deliver at least the score you asked for, which is the reading you want when you are working from a target. A raw 45 tops out at 650 and cannot produce a 655, so 46 is honest rather than generous.

How College Board Actually Calculates It

None of the above describes the real test, and the difference is not cosmetic. Four mechanisms account for it.

• Equating. Every form is scored against a table built for that form. College Board defines the procedure as "a statistical procedure that accounts for differences in difficulty" used "to make scores interchangeable across multiple test forms", which is precisely what a fixed line cannot do.
• A curve, not a line. When the 200 to 800 section scale was established in College Board's 2014 scaling study, the raw-to-scale transformation was chosen as either an arcsine transformation or a cubic polynomial, fitted so that measurement error stayed roughly constant along the range. Neither shape is a straight line.
• Item response theory. On the digital SAT, College Board tells students their scores "depend on several factors, including whether you answer questions right or wrong, the difficulty level of the questions, and the probability that you might be guessing". Two students with the same number correct can finish on different scaled scores.
• Adaptive delivery. Each section runs as two separately timed modules. The first mixes easy, medium and hard questions, and "based on how a student performs on the first module, the second module of questions will either be more difficult or less difficult".

Two endpoints do survive intact, and they are the two this page borrows. The scaling goals College Board published for the current scale state that "all correct, maximum possible raw scores convert to a highest obtainable scale score of 800" and "none correct, minimum possible raw scores convert to a lowest obtainable scale score of 200". The ends of the line here are real. Everything between them is a stand-in.

The operational tables are not released. College Board's technical manual publishes sample tables only, noting that they "don't represent any operational conversion table that was or ever will be used to score the SAT", and describes the real ones as proprietary. A page that offers you the current official conversion table is offering an estimate under a better name.

SAT Score Chart: Raw Mark to Scaled Score

Ten points a mark compresses the full 61-entry table into something short. Every figure below is this converter's own output in Raw Score to Scaled Score mode, and it is unchanged on Math, Reading and Writing.

The third column is not something the tool prints. It is the arithmetic College Board describes — the total is the sum of the two section scores — applied to a reader who scored identically on both, and it is why the section range doubles neatly into the published 400 to 1600.

Between the rows the pattern holds without exception: add one raw mark, add ten scaled points. Raw 33 returns 530, raw 42 returns 620, raw 58 returns 780, all three run through the code rather than inferred from the line.

The Same Chart Read Backward

Scaled Score to Raw Score mode on the same line, including entries that miss the grid and entries past the top of it. Nothing at or below 200 appears, because that end of the scale answers raw 0 and lands in the amber notice rather than the readout.

Upward rounding is the row to remember: ask what reaches 655 and the answer is 46, never 45. The last three rows are the ones to distrust, because a total entered by mistake comes back as a plausible 60 rather than an error. The 480 and 530 rows are there deliberately — those two figures are College Board's readiness benchmarks, and what they cost in raw marks is covered further down.

Where the Chart Stops

The table has a floor and a ceiling, and entries outside them divide into two groups: some are absorbed and returned as ordinary-looking answers, others are refused outright. The score box advertises a range of 0 to 800, but the form is submitted with browser validation switched off, so whatever you type reaches the calculation and the six rules below are the only ones that apply.

• A raw mark above 60 is held at 60. Enter 61, 75 or 300 and all three read 800, with nothing to signal that the entry was out of range.
• A scaled score above 800 is held at the top of the table and reads 60. That covers 801 as easily as it covers 1600.
• A blank or non-numeric entry is read as 0. Forward, that returns 200; backward, it meets the floor rule below.
• A negative number is refused in either mode. The notice reads "Invalid score" and no conversion is made.
• A raw mark with a decimal is refused too, because there is no entry between two whole marks for it to land on. Enter 44.5 and the notice reads "Enter a whole raw mark from 0 to 60". Round the mark first. Decimals are fine on the scaled side, where 641.5 converts to raw 45.
• A scaled score of 200 or less is the one answer the readout cannot display. The raw mark that reaches it is 0, and a bare zero headline is how every calculator here signals a bad entry, so the answer arrives in the notice instead: "Raw 0 already reaches 200, the bottom of the scale".

The first three are the dangerous ones, because they come back through the readout looking exactly like a genuine conversion. The last three all paint the same amber panel, and the sentence inside it is the only thing separating them.

How to Read Your Result

The panel returns four lines and only the first of them is calculated.

The Four Lines the Panel Returns

• Converted Score — the headline. Forward it is a scaled score between 200 and 800; backward it is a raw mark between 1 and 60.
• Section — your menu choice in capitals, MATH, READING or WRITING. It records which paper you were marking and has no effect on the figure above it.
• Mode — either "raw to scaled" or "scaled to raw". Worth a glance, because the two modes share one box and answer different questions: 45 typed into the reverse mode is not a raw mark at all, it is a scaled score below the floor, and it returns the amber notice rather than 650.
• Input — the number you typed, echoed exactly. An empty box shows here as 0, which is the fastest way to notice that the 200 above it is an artifact rather than an answer.

Copy result puts the readout and its breakdown on the clipboard in one action, which is the quickest way to move a marked paper into a spreadsheet. The arithmetic itself runs in your browser, though the page loads analytics and advertising scripts like every other page on the site.

When the Amber Notice Appears Instead

Three entries replace the readout with the amber "Check your inputs" panel, and the sentence inside it is the only thing that tells them apart.

The third is not really a refusal. Raw 0 genuinely is the mark that reaches 200 — it is the bottom rung of the ladder — but a bare zero headline is how this site signals a failed calculation, so the correct answer can only be delivered as a sentence. Read it as the answer it is.

None of the three comes from the browser. Validation is switched off on the form, so every one of them is the calculation itself declining to convert.

What the Converted Number Is Worth

A converted score is a rung, and there are two published rungs worth measuring against. College Board's College and Career Readiness Benchmarks are the minimum section scores associated with a 75% chance of earning at least a C in relevant first-semester, credit-bearing college courses.

• 480 on the reading and writing side — the benchmark for literature, history, social science and writing courses, set on the section College Board then called Evidence-Based Reading and Writing. On this converter's line, 480 is raw 28.
• 530 on Math — the benchmark for college algebra, statistics, precalculus and calculus. On the same line, 530 is raw 33.
• Both were calculated on students who sat the pre-2016 SAT and matched to their actual college transcripts, then carried across to the current scale, so they are outcome-based rather than a passing mark somebody chose.

Movement is easy to price here and hard to price on a real form. On this line one more correct answer is worth ten points and ten more are worth a hundred, flat, everywhere: climbing from 620 to 700 is exactly eight extra marks, raw 42 up to raw 50. On an adaptive, equated paper the value of one more correct answer depends on which question it was and how the rest of the section went, so take eight marks as the shape of the gap rather than its price.

Turning a gap that size into hours you can actually sit down and do is a scheduling problem rather than a scoring one, and that is what you get from the Study Planner Generator.

One thing the number will never be is a verdict. A converted score has no percentile attached, no score range around it and no knowledge of the schools you are applying to, and the benchmarks above describe readiness for introductory coursework rather than admission anywhere.

Limits: When This Converter Does Not Apply

This tool does one job accurately and several neighboring jobs not at all. The full list, so nothing here has to be guessed at:

• It is not an official conversion table. College Board equates each form separately and does not publish its operational tables; this is a flat line standing in for their average shape.
• It does not convert to or from the ACT. There is no ACT field, no ACT output and no concordance table anywhere in the calculation.
• It does not report a percentile, because that needs the scores of everyone else who sat the test.
• It does not total your sections. There is no 400 to 1600 output and no way to enter two sections at once.
• It does not treat Math, Reading and Writing differently. The three menu entries read one shared table.
• It does not score an essay, a subject test or a PSAT, and it has no setting for the older paper format.
• It does not change between visits. There is no test date, no difficulty control and no curve for it to change with.

What it is good for is the job it does: turning a marked practice section into a figure on the familiar 200 to 800 scale, and turning a target back into the number of marks that would reach it. The two subsections below are the reasons to hold that figure loosely.

The 60-Mark Denominator Is the Biggest Distortion

The table assumes 60 marks per section. No SAT section has 60 questions. College Board's published structure gives Reading and Writing 54 questions across two 32-minute modules and Math 44 across two 35-minute modules — 98 questions in 134 minutes, and neither figure is the one this line is built on.

• 45 correct out of 54 on Reading and Writing is five-sixths of the section. Entered as a raw 45 it returns 650. Rescaled to the tool's 60-mark denominator it is exactly 50 marks, and 50 returns 700. Same paper, 50 points apart, from the denominator alone.
• 40 correct out of 44 on Math is a little over 90% of the section. As a raw 40 it returns 600. Rescaled to 60 marks it is 54.5, and the nearest whole mark, 55, returns 750 — a spread of 150 points on one performance.

If the number you actually need is the mark a target grade demands on something you have not sat yet, that is a different calculation entirely — try the Exam Score Predictor.

Neither figure is right, because a percentage correct is not a scaled score either. The point is narrower: the answer moves by more than a hundred points depending on which denominator you assume, and this one assumes a number no current SAT section uses. Where a practice test came with its own scoring report, that report is the authority and this page is the estimate.

A Raw Count Cannot Describe an Adaptive Test

The deeper problem is that the digital SAT has no fixed paper to count against. Both sections run as two separately timed modules; the opening module carries a mix of easy, medium and hard questions, and "based on how a student performs on the first module, the second module of questions will either be more difficult or less difficult". Two students who answer the same number correctly may not have answered the same questions, or even the same second module.

College Board's description of the scoring is not a count at all. Scores, it tells students, "depend on several factors, including whether you answer questions right or wrong, the difficulty level of the questions, and the probability that you might be guessing". That is item response theory, and it has no raw mark to take as input.

The clearest evidence is what a real score report leaves out. The current Understanding Scores document College Board sends home with SAT School Day results does not use the word raw once: what arrives is a Reading and Writing score, a Math score, a total, a score range and a percentile, with no count of correct answers anywhere in it. This page converts a number the SAT no longer reports — which is exactly why it earns its place on a practice paper and none at all on a real result.