The test grade calculator above converts a raw score into a percentage and a letter grade in one step, working either from the number of wrong answers or from the points earned. Alongside the grade it builds a chart showing what every possible number of mistakes would have scored on this particular test, which is the fastest way to see how much a single question is worth.
Arb Digital builds free calculators for the everyday arithmetic that people repeat by hand, and marking is a clear case. A teacher scoring thirty papers, or a student checking a returned test against the syllabus, is doing the same division over and over with a grade table sitting beside it. This page does both parts at once and shows its working.
What This Test Grade Calculator Does
Scoring a test has three moving parts: the raw score, the percentage, and the letter. The raw score is what you counted. The percentage is the raw score divided by the total available. The letter is whatever your institution's scale maps that percentage to, and that mapping is far less standard than most people assume.
The tool covers five scales. The US standard ten-point scale puts A at 90 and each letter ten points below the last. The plus/minus variant subdivides each band into three, which is what most American universities actually use on transcripts. The strict scale, common in some science and professional programmes, starts A at 93. The UK degree classification uses First, Upper Second, Lower Second and Third at 70, 60, 50 and 40. The Australian scale runs High Distinction, Distinction, Credit and Pass at 85, 75, 65 and 50.
One boundary is worth stating clearly. This page scores a single test in isolation. Our weighted grade calculator is the one that combines several categories — homework, quizzes, midterm, final — each carrying a different weight, into a course grade. Use this page for the test; use that page for the semester.
How to Use It
- Choose your entry mode. If you are marking, counting the crosses is faster than counting the ticks, so wrong-answer mode usually wins. If you are reading a returned paper that says 44 out of 50, switch to points earned.
- Enter the total. This is questions on a question-counted test, or points on a points-weighted test where an essay might be worth twelve and a short answer two.
- Enter the score. The tool accepts fractional values, so half marks and partial credit both work.
- Pick the scale that matches your syllabus. The same 88% is a B on the standard scale and a B+ on the plus/minus scale, so this choice changes the answer.
- Read the chart underneath. Each row is a number of wrong answers with the percentage and grade it produces, which tells you instantly how many mistakes separate one grade from the next.
The Formula / How It's Calculated
The percentage is (points earned + bonus) ÷ total × 100. In wrong-answer mode the points earned are simply total − wrong, so the whole thing collapses to (total − wrong) ÷ total × 100.
Work the default example. A test has 50 questions and the student got 6 wrong, so 44 are right. 44 ÷ 50 = 0.88, which is 88%. On the US standard scale that is a B, because 88 sits between the 80 floor for B and the 90 floor for A. On the plus/minus scale the same 88% is a B+, since B+ begins at 87. The gap between those two answers is the entire reason the scale selector exists.
The points-to-next-grade figure works backwards from the next threshold. On the standard scale the next band up is A at 90%, so the student needs 0.90 × 50 = 45 points, which is one more than the 44 they scored. That single question is the difference between a B and an A, and on a 50-question test each question is worth exactly 2 percentage points. Percentage grading scales vary widely between institutions — Columbia University's School of Professional Studies grading policy, for example, places A at 93 to 97.9% with A− at 90 to 92.9%, which would score this same paper a B+ rather than a B.
Why Short Tests Produce Unstable Grades
The number of questions on a test sets the granularity of every grade it can produce, and short tests are far coarser than people realise. On a 20-question test each item is worth five percentage points, so the only achievable scores are 100, 95, 90, 85 and so on. There is no such thing as an 88% on a 20-question test — the grade jumps straight from 85 to 90, straight from B to A.
This has a real consequence for fairness. A single ambiguous question, a misread instruction, or one careless slip moves a student a full grade band. On a 100-question test the same slip costs one point and changes nothing. The measurement is not more accurate on the longer test because the students are different; it is more accurate because the instrument has finer divisions.
The practical takeaway for anyone setting tests is that if you intend to award plus and minus grades, you need enough items for those distinctions to be meaningful. Awarding a B+ rather than a B− on the basis of a 10-question quiz is assigning precision that the test simply cannot deliver. Where a course grade depends on many small assessments, the semester grade calculator aggregates them into something more stable than any single quiz.
Rounding, and the 89.5% Problem
Rounding decisions decide borderline grades, and they are made inconsistently. A student who scores 89.5% is at an A under standard rounding to whole numbers and at a B under truncation. Neither approach is wrong in the abstract, but they must be stated in advance and applied to everyone identically.
The decimal-places control on this tool exists to make that visible rather than to hide it. Set it to zero and you see the rounded percentage that most gradebooks display. Set it to two and you see what the student actually scored. The grade letter here is always assigned from the unrounded percentage, which is the more defensible convention: rounding first and then comparing against a threshold effectively lowers every threshold by half a point.
A related trap is the difference between rounding a single test and rounding a final course average. Rounding each component before combining them compounds the error across the term. Keep full precision through every intermediate step and round once, at the end, when the number is finally reported. The final grade calculator follows the same rule when working out what a student needs on a remaining exam.
Points-Weighted Tests Versus Question-Counted Tests
Not every test gives each question equal value. An exam with ten short-answer items worth two points each and two essays worth fifteen each has 50 points available but only twelve questions. Getting both essays right and half the short answers gives 40 out of 50, or 80%. Getting all ten short answers and both essays half-right gives 35 out of 50, or 70%. Identical question-level accuracy, one full grade apart.
Enter the point total rather than the question count whenever items are unequally weighted, and enter points earned rather than questions right. The wrong-answer mode assumes every item carries the same value, which is exactly true on a multiple-choice test and quite wrong on a mixed-format paper. If you are unsure which you have, look at whether the mark scheme allots a different maximum to different questions.
Partial credit belongs in points mode too. A student who earned 3.5 of a possible 5 on an extended response has 3.5 points, and the calculator accepts the decimal without complaint. Converting that to "wrong answers" is not possible in any meaningful way.
What a Letter Grade Does and Does Not Mean
A letter grade is a compression of a percentage into a handful of categories, and compression always loses information. A B covers everything from 80 to 89 on the standard scale, so two students with a full ten points between them receive the same letter. That is by design — the categories exist to reduce false precision — but it means the letter alone is a poor basis for comparison.
Letters also carry different meanings across institutions and even across departments in the same institution. The University of North Carolina's explanation of its grading system sets out separate grading schemes for undergraduates, graduate students and several professional schools, which is typical rather than unusual. A percentage-to-letter table you found online is not authority for how your own course is graded; your syllabus is.
Finally, the letter is not the grade point. Converting letters to a GPA involves a second mapping, where A is 4.0, B is 3.0 and so on, sometimes with plus and minus modifiers at 3.3 and 2.7. That conversion, and the credit-hour weighting that goes with it, is handled by our GPA calculator and the CGPA calculator rather than here.
Arb Digital builds free, no-signup calculators for statistics, education, finance and construction. Browse the full library, or get in touch if there is a calculation your team keeps doing by hand.
Browse All Free Tools Contact Arb DigitalCommon Mistakes to Avoid
- Using wrong-answer mode on a points-weighted test — it assumes every item is worth the same, which is false the moment an essay sits beside a short answer.
- Applying a grading scale from another institution — the A threshold ranges from 90 to 93 across common US scales, and UK and Australian systems use entirely different bands.
- Rounding the percentage before assigning the letter — this quietly lowers every grade boundary by half a point for every student.
- Adding curve points to the total instead of the score — bonus points raise what the student earned; they do not change what the test was out of.
- Treating one test as a course grade — a single assessment is a noisy signal, and the weighted average across the term is the number that matters.
Related Free Tools From Arb Digital
Combine several weighted categories into a course grade with the weighted grade calculator, work out what you need on a remaining exam with the final grade calculator, apply a curve to a whole class with the grade curve calculator, convert letters into grade points with the GPA calculator, and handle any other proportion with the percentage calculator. The full free online tools hub holds the rest of the education set.
Frequently Asked Questions
Divide the points earned by the total points available and multiply by 100. A score of 44 out of 50 is 44 divided by 50, which is 0.88, or 88%. If you counted wrong answers instead, subtract them from the total first.
On the US standard ten-point scale, 88% is a B, because B covers 80 to 89.99. On a plus/minus scale it is usually a B+, since B+ typically starts at 87. Always check your own syllabus, because thresholds differ between institutions.
Multiply the total by the A threshold as a decimal, then subtract from the total. On a 50-question test with an A at 90%, you need 45 right, so five wrong answers is the limit. The chart on this page lists the grade for every number of mistakes.
Because the thresholds are institutional choices, not fixed rules. A scale that starts A at 90 and one that starts A at 93 will disagree about every score between those two numbers, and UK and Australian systems use completely different bands again.
The more defensible convention is to assign the letter from the unrounded percentage, then round only for display. Rounding first effectively moves every grade boundary down by half a point, which changes outcomes for borderline students.
Switch to points-earned mode and enter the point total rather than the question count. Wrong-answer mode assumes equal weighting, so it will misprice any test that mixes short answers with essays or gives partial credit.
Enter them in the bonus field. They are added to the points earned before the percentage is calculated, while the total stays as the test was originally set. That is why a bonus can push a score above 100%.
No. This page scores one test on its own. A course grade combines several assessments with different weights, which is what the weighted grade calculator is for.
This calculator is provided for education and planning only. Grading scales differ between institutions, so confirm the thresholds and rounding rules in your own syllabus before relying on a letter grade.
