A parent finishes marking a twenty question science quiz, counts seventeen right, and types it into a grade calculator. Eighty-five percent. Then comes the pause. In a classroom that number sits inside a spread of thirty others. At a kitchen table it sits alone, with nobody to ask whether the quiz was harder than it should have been.
This is the part of home education that catches numerate parents off guard. The arithmetic is trivial. The interpretation is not.
The stakes are low in third grade and high in tenth. What a family records early is what it has to work with later, and turning course grades into a transcript is far easier when the numbers were kept with that in mind from the start.
What a percentage means with a sample size of one
Classroom grades carry two kinds of information. There is the raw proportion of items answered correctly, and there is a second channel nobody writes down: the class average, whether the teacher thought the test was brutal, whether two thirds of the room missed question fourteen. Thirty wrong answers to one question are evidence about the question, and teachers adjust.
At home only the first channel exists. Eighty-five percent means the student answered seventeen of twenty items on one assessment, chosen or written by the same person now reading the result.
Small samples compound that. On a twenty item quiz, one careless sign error moves the score five points. A run of 90, 80, 95 and 85 is noise, not a trend. The useful move is to read the items instead of the total. Three wrong answers across three unrelated topics mean something different from three that are all the same operation with fractions. The second is a specific gap with a specific remedy, and the percentage hides which one you have.
Deciding what actually counts toward the grade
Weighting is a decision a classroom teacher inherits from a department and a homeschooling parent makes alone. Most families settle on a mix of daily work, tests, and projects, and the temptation is to weight daily work heavily because there is so much of it.
There is a good reason not to. Daily work at home is done with a parent within arm’s reach. A worksheet completed while someone says “check that sign again” measures the pair, not the student. A test taken alone measures what the student produces unaided.
A defensible shape for a high school course puts tests and exams at roughly half, projects, labs, or major writing at a quarter to a third, and daily work in the remainder. The exact split matters less than choosing it in September and not moving it in May.
Mastery based redos complicate this. Many families let a student correct a failed test and re-earn the points, since the goal is knowing the material rather than being marked forever for not knowing it in March. The catch is that if every score converges on 100, the grade stops carrying information. Recording two numbers solves most of that. The first attempt says what was there independently. The final one says whether the gap closed.
Feedback and record are two different jobs
Most assessment in the elementary years is feedback. A parent who sat through the lesson knows whether the child understood it, so writing 85% on the page adds nothing. The useful record in those years is descriptive: what was covered, what was hard, what had to be repeated.
Grades earn their place when somebody outside the household reads them, and that is a narrow set. Admissions offices, scholarship committees, a state or district official reviewing a portfolio, a school taking a transfer student. None of them can see the work.
That distinction decides which numbers need to survive June. Knowing which is which is what separates a stack of marked papers from something a stranger can actually read.
When families start keeping grades at all
Most families begin recording grades formally around ninth grade, sometimes eighth for a high school level course like Algebra I. The reason is external. Ninth grade is when the record starts being read by people who were not in the room.
Before that, what a family must document varies considerably. Homeschooling in the United States is regulated state by state, and requirements take different shapes: a notice of intent, a count of instructional days or hours, a portfolio reviewed by an evaluator, standardized testing in certain grades, or almost nothing. Letter grades satisfy very little of it. Attendance logs, work samples, and test results do more.
High school is different because the transcript becomes the artifact, and a course grade has to be defensible on its own terms. That means a stated scale, so that 90 to 100 is an A and every reader knows it, and a credit definition, commonly a full year course or roughly 120 to 180 hours of documented work. Families assembling a transcript usually find the GPA arithmetic is the easy half.
What has to survive for the numbers to mean anything
A percentage with no denominator cannot be reconstructed. “Chapter 7 test, 88%” is a dead end a year later. “Chapter 7 test, 44 of 50, October 14, stoichiometry” can be rebuilt, reweighted, or corrected. Keep points earned and points possible, and recalculate the quotient as needed.
The same holds at the course level. A B in American Literature answers nothing by itself. Set beside a book list, a count of essays, and the weighting that produced it, it becomes a claim a reader can evaluate. The families who find transcript season painful are rarely the ones who graded too strictly. They are the ones who kept percentages and threw away what explained them.
A grade calculated at home is a statement a family is making about its own student, and it is read that way. That is not an argument against grading. It is an argument for keeping each number narrow: this assessment, these items, this weighting. A number that says exactly what it measured holds up. A number standing in for a year of judgment does not.
