How your score is calculated
We show you a range, like 1210–1290, instead of a single number. That is not us hedging. It is the only honest thing we can report, and this page explains exactly why.
Why does Preptics show a score range instead of a single number?Link to the section: Why does Preptics show a score range instead of a single number?
Converting raw answers into a 400-1600 score requires equating data calibrated on a large reference population, and the College Board does not publish it. No practice platform has that data, including this one. Reporting a single number would hide the uncertainty rather than remove it, so we report the estimate as a range with its confidence stated.
Turning a raw count of correct answers into a score out of 1600 requires equating: a statistical mapping calibrated against a large reference population who sat the same material. The College Board has that data for their tests. We do not have it for ours, and neither does any other practice-test company.
The difference is that we say so. When a practice platform reports you a 1250 to the point, that precision is invented — the underlying uncertainty has not gone anywhere, it has just been hidden from you. We would rather show you the uncertainty and let you plan around it.
How is this different from how practice scores are usually reported?Link to the section: How is this different from how practice scores are usually reported?
Most practice platforms report a single scaled score and leave the uncertainty behind it unstated. We report a range, say where the conversion tables came from, narrow the range as you sit more mocks rather than keeping the precision fixed, and publish what would make our numbers wrong. The table below sets the two side by side.
| On the practice score | Common practice | Preptics |
|---|---|---|
| The number you are shown | A single scaled score, reported to the point — 1250. | A range with its confidence stated — 1210–1290, low confidence. |
| Where the conversion comes from | Usually unstated. | Our own calibrated tables, and we say plainly that we do not have the College Board's equating data. |
| What happens as you sit more mocks | The reported precision does not change. | The range narrows on evidence: ±60, then ±45, then ±40. |
| Second-module routing | Often a fixed linear test scored as if it were adaptive. | Two modules per section, the second routed by the first, with a separate conversion curve for each path. |
| Being wrong | Not addressed. | A published account of what would make our scores run high or low, and a commitment to say so here when a recalibration moves them. |
How is a Digital SAT practice score actually calculated?Link to the section: How is a Digital SAT practice score actually calculated?
In four steps. Reading and Writing and Math are scored separately from 200 to 800 and added together. Which second module you were routed into selects the conversion curve, because the same raw count means different things on the easier and harder module. A lookup table turns raw correct answers into a section score. Then the estimate is deliberately widened into a range.
Two sections, scored separately
Reading and Writing and Math are scored independently, 200 to 800 each, and added to make your 400 to 1600 total. They are not averaged and they are not weighted against each other — a strong Math score cannot rescue a weak Reading score, exactly as on the real test.
Which second module you got, matters
Each section gives you a first module, then routes you to an easier or harder second module based on how you did. Getting 40 questions right on the harder module is not the same performance as getting 40 right on the easier one, so the two have different conversion curves. A test that ignores this — or that is not adaptive at all — cannot produce a number that means anything.
Raw count to section score
Each second module has a lookup table from raw correct count to a section score in 10-point steps. It is a table rather than a formula because the relationship is not a straight line: the difference between 30 and 32 right is not the same number of points as the difference between 50 and 52.
Then we widen it, on purpose
This is the step almost nobody else does. The number that comes out of step 3 is an estimate built on a conversion table we calibrated ourselves, not on the College Board's equating data — which is not public and which we do not have. So we report a range around the estimate rather than the estimate alone.
How wide is the range, and when does it get narrower?Link to the section: How wide is the range, and when does it get narrower?
It is ±60 points on your total after one mock, ±45 after two, and ±40 after three or more, with ±30, ±25 and ±20 on each section. The range narrows as you supply more evidence, not as you improve. The real SAT's own standard error is roughly ±40 with full equating behind it, so a practice test deserves a wider band than that, not a tighter one.
| Evidence | Range on your total | We call it |
|---|---|---|
| After 1 mock | ±60 | low confidence |
| After 2 mocks | ±45 | moderate confidence |
| After 3 or more | ±40 | good confidence |
For comparison, the real SAT's own standard error of measurement is roughly ±40 points on the total — and that is with full equating behind it. A practice test without that data deserves a wider range than the real test, not a narrower one, which is why we start above it rather than below.
How should I use a score band when I plan?Link to the section: How should I use a score band when I plan?
Plan against the bottom of your range rather than the top: a band of 1210-1290 has not reached a target of 1300. Watch the trend across several mocks instead of any single score, because three overlapping bands drifting upward is stronger evidence than one high result. And treat the band as our estimate from our test, not as your SAT score.
Plan against the bottom of your range, not the top. If your band is 1210–1290 and your target school wants 1300, you are not there yet even though the top of your band nearly touches it.
Watch the trend, not any single score. Three mocks with overlapping bands that are drifting upward is much stronger evidence than one high score.
Do not treat this as your SAT score. It is our best estimate of how you would do, from a test we wrote. Only the College Board can tell you what you actually scored.
What would make these score estimates wrong?Link to the section: What would make these score estimates wrong?
Our conversion tables rest on our own item-difficulty estimates. If our questions are systematically easier than the real test our scores run high, and if they are harder they run low. As real response data accumulates we recalibrate and the ranges tighten. If a recalibration moves scores materially we will say so on this page rather than quietly adjusting the numbers.
Our conversion tables are built from our own item difficulty estimates, and those are estimates until enough people have answered each question for the real statistics to emerge. If our questions are systematically easier than the real thing, our scores run high. If they are harder, they run low.
As response data accumulates we recalibrate against it, and the ranges tighten because they have earned the right to. If a recalibration moves scores materially we will say so on this page rather than quietly adjusting the numbers.
What is already running is the groundwork. Every question’s difficulty is re-estimated from real answers, and a question graduates to a proper statistical model only once it has the evidence — 150, 400 or 800 usable answers depending on the model — with a weekly job refitting them from answers read without names or accounts. Those estimates are used today to find questions that behave badly. They do not yet feed the conversion tables above; when they do, this page will say so first. How question difficulty is estimated.
If you have sat the real Digital SAT and your actual score fell outside the band we gave you, we want to know — that is the single most useful piece of feedback anyone can send us. Tell us here.
SAT® is a trademark registered by the College Board, which is not affiliated with, and does not endorse, this site. All practice questions and reading passages are original works created by Preptics and are not actual SAT® questions.