SAT raw score conversion
What a raw score is, how it becomes a section score, and the two places this goes wrong on almost every other site.
A raw score is just a count
It is the number of questions you answered correctly. Nothing is deducted for a wrong answer, and nothing is deducted for leaving a question blank. The quarter-point guessing penalty was removed with the 2016 redesign, so on the current test there is no reason not to answer everything.
The official procedure
- Count correct answers in each module.
- Add the two modules together to get one section raw score.
- Look that number up once, in the table for the test you took.
- Read off the lower and upper scaled values.
- Repeat for the other section, then add the lower bounds together and the upper bounds together.
Step two is where calculators go wrong. Converting module 1 and module 2 separately and adding the two scaled results produces a different number, and a worse one: it penalises a student who performed evenly across both modules relative to one who was lopsided at the same total raw score. College Board's guide is explicit that raw scores are summed first.
Why the conversion is a range
Every official table gives two values. A Reading and Writing raw score of 45, for example, converts to a band roughly 20 points wide, and the position of that band moves depending on the test.
College Board explains the reason in the guide itself: the paper scoring method is a simplified and therefore slightly less precise version of the one used on the real test. The range is not padding. It is the published answer.
Why every test needs its own table
Tests are equated so that a given scaled score means the same thing regardless of which form you sat. Since forms differ slightly in difficulty, the raw to scaled mapping has to differ to compensate.
The size of that compensation is easy to underestimate. At a Reading and Writing raw score of 40, the eight official tables put the lower bound anywhere from 470 to 540, a 70-point gap before Math is added.
The full comparison across all eight tables is here.
The adaptive test has no raw score conversion
| Paper practice test | Adaptive Bluebook | |
|---|---|---|
| Reading and Writing | 66 questions | 54 asked, 50 scored |
| Math | 54 questions | 44 asked, 40 scored |
| Conversion table | Published, one per test | None, and none possible |
The real digital SAT is scored with Item Response Theory. Your scaled score depends on which questions you answered correctly and how difficult each of those questions was, so two students with identical raw scores can receive different section scores. There is no table that could express that.
This is the second place calculators go wrong, and it is worse than the first because it is invisible. A raw score out of 54 fed into a table built for 66 produces a confident, specific and badly wrong answer. The full explanation is here.
Where the tables are
College Board publishes 8 of them, for practice tests 4 through 11. Each has its own page here with the complete chart and a link to the source PDF:
- SAT Practice Test 4
- SAT Practice Test 5
- SAT Practice Test 6
- SAT Practice Test 7
- SAT Practice Test 8
- SAT Practice Test 9
- SAT Practice Test 10
- SAT Practice Test 11
Or let the calculator apply the right one and show its working.
Read next
- How many questions can you miss on the SAT The measured miss-to-score ranges across all eight official College Board tables, and why no exact answer exists.
- Bluebook or paper practice test, and which score you can actually calculate Bluebook is 54 and 44 questions; the paper tests are 66 and 54. Why mixing the scales costs 170 points.
- SAT scoring terms that no longer exist Subscores, cross-test scores, EBRW, the essay and the guessing penalty: what each was and when it was retired.
Common questions
- What is a raw score on the SAT?
- The number of questions you answered correctly in a section. There is no penalty for a wrong answer, so a raw score is simply a count of correct responses.
- How does a raw score become a 200 to 800 score?
- On a paper practice test, through the conversion table published with that specific test. On the real adaptive test, through Item Response Theory, which weights individual questions by difficulty and does not reduce to a table.
- Is there a penalty for guessing?
- No. The old quarter-point deduction for a wrong multiple-choice answer was removed with the 2016 redesign. Leaving a question blank and answering it incorrectly cost the same thing, which is nothing.
- Why do the same raw scores convert differently on different tests?
- Because each test is equated to account for small differences in difficulty. At a Reading and Writing raw score of 40, the eight official tables place the lower bound anywhere from 470 to 540.
Last updated