Easy Grader
The grading chart the plastic wheel used to give you. Say how many questions are on the test and get every possible number wrong, the percentage it leaves, and the letter it earns — the whole column at once, so you mark the stack by reading down it instead of typing thirty scores into a box.
The test you are marking
Every question counts the same. The chart stops at 200 questions — past that nobody reads down a column, and the printed sheet stops being useful.
Optional. Marks one row of the chart and puts it at the top of the result. Leave it blank for the plain table.
Schools disagree. A 93 is an A here and an A− wherever A starts at 94; a 90 is an A− here and a straight A on the ten-point scale.
22 right out of 25. One question moves the score by 4 percentage points, so this test can produce 26 distinct scores and nothing in between them.
- Questions on the test
- 25
- Each question is worth
- 4%
- Scores this test can produce
- 26
- Letter-grade scale
- Plus/minus — A starts at 93
- Score you looked up
- 88% · B+
| Wrong | Right | Percentage | Letter |
|---|---|---|---|
| 0 | 25 | 100% | A |
| 1 | 24 | 96% | A |
| 2 | 23 | 92% | A− |
| 3 | 22 | 88% | B+ |
| 4 | 21 | 84% | B |
| 5 | 20 | 80% | B− |
| 6 | 19 | 76% | C |
| 7 | 18 | 72% | C− |
| 8 | 17 | 68% | D+ |
| 9 | 16 | 64% | D |
| 10 | 15 | 60% | D− |
| 11 | 14 | 56% | F |
| 12 | 13 | 52% | F |
| 13 | 12 | 48% | F |
| 14 | 11 | 44% | F |
| 15 | 10 | 40% | F |
| 16 | 9 | 36% | F |
| 17 | 8 | 32% | F |
| 18 | 7 | 28% | F |
| 19 | 6 | 24% | F |
| 20 | 5 | 20% | F |
| 21 | 4 | 16% | F |
| 22 | 3 | 12% | F |
| 23 | 2 | 8% | F |
| 24 | 1 | 4% | F |
| 25 | 0 | 0% | F |
- Every question carries the same weight. A test where one question is worth five points is a points problem, and one chart cannot describe it.
- Only 26 scores exist on this test, 4 points apart. Any cutoff that falls between two rows moves no paper from one letter to another.
- Letters come from the cutoffs above and nothing is rounded to a whole percent first. If your district rounds an 89.5 up to a 90 before reading the letter, do that yourself.
How the chart is built
Two lines of arithmetic, repeated once for every row. The chart has N + 1 rows because a student can get anything from none wrong to all of them wrong.
score = (N − wrong) ÷ N × 100 • step = 100 ÷ N- N
- Questions on the test, every one worth the same share
- wrong
- Questions missed, from 0 through N — one row each
- step
- The gap between two adjacent rows, and the finest distinction this test can make
The letter comes last. Each row’s percentage is rounded to four decimal places and then read against your cutoffs, highest band first, so the number in a row and the letter beside it can never disagree at a boundary. Nothing is rounded to a whole percent first — that is a district policy, not arithmetic, and if yours rounds an 89.5 up to a 90 before reading the letter you should apply that rule yourself.
The chart stops at 200 questions. That is a readability limit rather than a mathematical one: the table is N + 1 rows and every row is meant to be scanned by eye, and past a couple of hundred nobody reads down a column, they search it. At that length the step is a fifth of a point and the chart tells you nothing the raw percentage does not.
A 25-question test, and the 12-question quiz that cannot be graded with letters
Start with 25 questions. The step is 100 ÷ 25 = 4 percentage points, so the top of the chart runs 100, 96, 92, 88, 84, 80. Against the standard plus/minus scale that is A, A, A−, B+, B, B−. One wrong answer is still an A; two wrong is an A−; three wrong is the 88% that shows in the result panel above, a B+.
Notice what the cutoffs are doing. The A band runs from 93 to 100 and catches two rows, 100 and 96. The A− band runs from 90 to 93 and catches exactly one, the 92. Nothing in the test can produce a 91, a 94 or a 95 — those percentages simply do not exist on a 25-question paper, and neither does the 85 a student will ask you for.
Now cut the test to 12 questions
The step becomes 100 ÷ 12 = 8.33 percentage points. The whole chart is 100, 91.67, 83.33, 75, 66.67, 58.33 and then five more rows of F. Read those against the same twelve-band scale: A, A−, B, C, D, F. That is six letters, and six of the scale’s twelve bands — B+, B−, C+, C−, D+ and D− — cannot be awarded to anybody on that quiz, however they answer.
This is the practical consequence, and it is a marking decision rather than a maths one: a short quiz should not carry fine-grained letter grades. The narrowest bands on the US plus/minus scale are three percentage points wide, and once the step is wider than a band, whether that band is reachable at all is an accident of arithmetic rather than something anyone chose. A 12-question quiz measures in eight-point units. Grading it out of twelve letters implies a precision the paper does not have — record it as a percentage, or as a coarse pass/fail, and save the plus/minus scale for a test long enough to earn one.
If you only need the percentage for a single paper rather than the whole column, the test grade calculator takes questions right and questions wrong and gives you one score and one letter. This page exists for the other job: marking a stack against one chart.
What this calculator assumes
- Every question is worth the same share of the grade. That is what makes one chart valid for a whole stack of papers, and it is the assumption that fails first — a test where question 7 is worth five points needs a points total, not a chart.
- There is no partial credit and there are no bonus questions. Half a question wrong has no row, and a 21 out of 20 does not exist here.
- Letters are read off the percentage using the cutoffs you set, with nothing rounded to a whole percent first. The lowest band has to start at 0 so that the bottom of the chart — on a short test, that is most of it — lands somewhere.
- The chart is N + 1 rows, capped at 200 questions for readability. It is not a limit of the arithmetic.
- Nothing here judges whether your cutoffs suit the test. The chart shows you which bands are unreachable; deciding what to do about that is yours.
- Percentages are computed to four decimal places and shown to two, so a 12-question quiz reads 83.33 rather than 83.3333333333.
Easy grader FAQ
How many questions can a student miss and still get an A?
Read down the chart until the letter changes. On a 25-question test with A starting at 93, one wrong answer is 96% and still an A; two wrong is 92% and drops to an A−. So the answer is one. That number moves fast with test length: on a 50-question test the same A survives three wrong answers, and on a 10-question quiz it survives none at all, because a single miss is already a 90.
Why is there no 85% on my quiz?
Because a test of N questions can only produce N + 1 scores, spaced 100 ÷ N apart. On a 12-question quiz that spacing is 8.33 percentage points: 10 right is 83.33% and 11 right is 91.67%, and no arrangement of answers lands between them. An 85 is not a score anyone can earn on that paper. This is the single most useful thing the full chart shows you and the reason it is worth looking at before you set cutoffs.
Should a short quiz carry plus and minus grades at all?
Usually not. On the standard US plus/minus scale the narrowest bands are three percentage points wide. A 12-question quiz moves in 8.33-point steps, so six of the twelve bands — B+, B−, C+, C−, D+ and D− — cannot be awarded to anyone, no matter how they answer. Whether a given band is reachable becomes an accident of arithmetic rather than a decision you made. A short quiz measures in coarse units; a letter scale with fine gradations implies a precision the paper does not have.
Is a 93 an A or an A−?
It depends entirely on the school, which is why the cutoffs here are editable. On the plus/minus scale most common in the United States, A starts at 93, so a 93 is an A and a 92 is an A−. Plenty of schools start the A at 94, which makes the same 93 an A−. On the ten-point straight-letter scale, anything from 90 up is simply an A. Set the scale before you print the chart, because the percentages are objective and the letters are not.
What if some questions are worth more than others?
Then a single chart cannot describe the test, and this is not the right tool. A chart is only valid when every question carries the same share, which is what lets one column serve the whole stack of papers. For a test with a points total — 8 out of 12 on one question, 15 out of 20 on another — work in points and divide once at the end.
Can I print the chart or put it in a spreadsheet?
Yes, and that is what it is for. The print button produces the chart on paper for the desk drawer, and the export menu writes the same table to CSV or Excel, headers and all. Everything is generated in your browser; nothing is uploaded. The chart stops at 200 questions because past that nobody reads down a column and the printed sheet stops being useful.
Sources and review notes
- JCGM 100:2008, Evaluation of measurement data — Guide to the expression of uncertainty in measurement (BIPM) — the standing treatment of resolution as a limit on what any instrument can report, which is exactly what 100 ÷ N is here
- Standards for Educational and Psychological Testing (AERA, APA and NCME) — the reference on not reporting a score more finely than the instrument that produced it supports
- NCES Digest of Education Statistics — the standing federal reference for how grades are recorded across US schools, and the evidence that there is no national scale to default to
A metrology guide is an odd citation on a grading page, and it is deliberate. The arithmetic here is division and cannot go stale. The claim that needs a source is the one about resolution: a measuring device that moves in fixed steps cannot report a value between two of them, and a 12-question quiz is such a device. The cutoffs that turn a percentage into a letter have no standards body at all — your syllabus outranks every source above.