What a thirty-question test can and cannot express
A thirty-question test, a typical end-of-unit or end-of-chapter paper.
Every question on a thirty-question paper is worth 3.33% of the grade, so the only grades that exist are the 30 multiples of that figure. That is coarse enough that 3 of the thirteen plus/minus bands — B+, C+, D+ — cannot be hit at all: the score that first clears the floor overshoots straight into the band above.
The practical consequence is that a borderline case on a thirty-question test is decided by a single question. Going from 18 correct to 17 moves the grade by3.33% — the difference between 60% and 56.7%, or between a D- and an F on a 60% pass mark. 17 out of 30 is the highest score on this test that still fails.
What each letter needs out of 30
| Letter | Starts at | Lowest score out of 30 | Which is |
|---|---|---|---|
| A+ | 97% | 30 out of 30 | 100% |
| A | 93% | 28 out of 30 | 93.3% |
| A- | 90% | 27 out of 30 | 90% |
| B | 83% | 25 out of 30 | 83.3% |
| B- | 80% | 24 out of 30 | 80% |
| C | 73% | 22 out of 30 | 73.3% |
| C- | 70% | 21 out of 30 | 70% |
| D | 63% | 19 out of 30 | 63.3% |
| D- | 60% | 18 out of 30 | 60% |
Clearing the 93% A cut-off takes 28 out of 30, which allows 2 wrong answers; that score actually lands on93.3%, an A. A pass at 60% takes 18 out of 30, allowing 12 wrong. Those cut-offs are the common US ones and your school's may differ by a point or two — the percentage column is the part that transfers, and the letter grade to GPA page gives the full band table with grade points on the 4.0, 4.3 and 5.0 scales.
Every score out of 30
30 rows, highest first: the score, the percentage it produces, the letter it lands in on a standard plus/minus scale, and how many questions that leaves wrong. Each score links to a page working that grade out in full, including what the next letter up would take and what the same percentage looks like on a test of a different length.