How we estimate percentiles: South Carolina high school

A clear explanation of the data and the method, so you can judge how much weight to give the numbers.

What the data is

Every number here comes from official public test-results files: the South Carolina Department of Education's End-of-Course Examination Program (EOCEP) data files, one for each school year. They cover Algebra 1 (2018, 2019 and 2021 to 2025), English 2 (2021 to 2026), Biology 1 and U.S. History and the Constitution (2018, 2019 and 2021 to 2026), and English 1 (2018 and 2019). For each school, district and the state, and for each test and year, these files report:

• the percentage of students who earned each letter grade (A, B, C, D and F) and the real number tested. We turn the percentages into student counts, and every count rounds back to the exact percentage South Carolina printed, and
• the group's real average score and its standard deviation, a measure of how spread out the scores are. Few states publish both for every school; South Carolina does, so nothing about the center or the spread of a group has to be estimated.

Who is counted

Each test counts the students who took it during the school year, not only in the spring: South Carolina's files combine the fall/winter and spring testing windows and, in most years, summer. Since 2020-21 every student must take Algebra 1, Biology 1 and English 2 by the third year of high school, even without taking the course. Middle-school students who take Algebra 1 early (or English 1, in 2018 and 2019) are included, counted at their middle school. Students who qualify for the state's alternate assessment take that test instead and are not included.

A course exam is taken by whoever is in the course that year, so the group changes from test to test and year to year: about 64,100 students took Algebra 1 in 2025, and about 61,500 took English 2 in 2026. Each year from 2018 to 2025, between about 240 and 270 schools with no Biology 1, English 2 or U.S. History results, most of them middle schools, report Algebra 1 results. A percentile compares a score with that year's takers of the same test, not with every student of the same age, and an Algebra 1 percentile says nothing about Biology 1.

How the percentile is estimated

Every EOCEP test reports a whole-number score from 0 to 100, and each score matches a letter grade on South Carolina's grading scale (for example, a B is 80-89). The score is not the percentage of questions answered correctly. Given a score, we:

1. identify which letter grade the score falls in;
2. count everyone with a lower letter grade as scoring below;
3. within the score's own letter grade, spread that grade's students along the group's curve (below) rather than evenly across its range. Students bunch up toward the middle of the scale, so an even spread would put too many of them out at the far ends of the lowest and highest grades.

That gives the percentile: the estimated share of students in the group who scored at or below the entered score.

About the smooth curve

The bell-like curve on each chart is a normal curve centered on the group's published average score, with the group's published standard deviation, cut off at 0 and 100. It is not fitted to anything: both numbers are South Carolina's own. It's there to visualize the shape of the group's scores. The letter-grade counts set the headline percentile (steps above); the curve only decides where inside the student's own letter grade the line falls.

Real scores are not perfectly bell-shaped. In every statewide result here, and in most school results, more students land in the A and F ranges than a bell curve with the same average and spread would put there, by a few percentage points. That is why the curve and the bars can disagree a little at the ends of the chart: the bars are the real counts.

When a school's students are almost all in one letter grade, or its published spread is very small (a few selective schools), the curve can't place a score inside a grade. Inside the student's grade the percentile then spreads that grade's students evenly instead (24 of about 9,300 school results), and the chart draws the curve no narrower than one fifth of the statewide spread.

Cut scores

Every EOCEP test uses the same 0-100 scale and the same letter-grade ranges in every year here: A 90-100, B 80-89, C 70-79, D 60-69 and F 0-59, the ten-point South Carolina Uniform Grading Scale in use since 2016-17. South Carolina also describes an F as Does Not Meet, a D as Minimally Meets, a C or B as Meets, and an A as Exceeds; the tool names the levels by letter grade.

The ranges were read, for each year, from South Carolina's Uniform Grading Policy (February 2018 and May 2019 editions), the EOCEP Score Report User's Guide for each school year from 2020-21 to 2025-26, the EOCEP technical reports for 2022-23 and 2023-24, and the grading table printed in each year's data file. Every year's published averages and percentages also agree with these ranges. The ranges for each test and year are on the score guides.

The high-school progress report

The paid progress report puts every high-school test your child took in one table, with the percentile at each level from the school to the state, and charts each subject across the years.

• Each course exam is its own point. Every South Carolina high-school test here is a course exam (such as Algebra 1 or Biology 1), taken by the students in that course that year, a different group for each course, so the report does not join them with a line.
• Earlier years are compared statewide only. You can add your child’s grades 3-8 scores in the same subjects. The tool doesn’t know which middle school or district they attended, so those years are compared with every student in the state in that grade, not with this high school or its district.
• The report opens on the statewide view, the one level with a result for every test.

Limitations to keep in mind

• Percentiles are estimates, not official rankings. Treat a "62nd percentile" as "roughly the top 40%," not an exact standing.
• Small schools have small counts, so their estimates are noisier.
• South Carolina hides the results of any school with fewer than 20 students tested on a test. Those schools don't appear for that test, but their students still count in the district and state results (about half a percent of all students tested).
• Not included: 2020, when no spring test was given (the 2019-20 file has only fall and summer testers, and the state cautions against comparing it with other years); English 1 in 2021, given that year only to 316 students who still needed it; and Algebra 1 in 2026. South Carolina's new math standards put Geometry with Statistics before Algebra 1 starting in 2025-26, so that year only about 8,800 students took the new Algebra 1 test (about 64,100 in 2025), mostly high-school students outside the new sequence and almost no middle-school students. A percentile against that small, unusual group would not mean what it means in other years.
• The tests have changed over time while the 0-100 scale and the letter grades stayed the same. English 2 replaced English 1 as the English end-of-course test (first given in 2019-20), and U.S. History and the Constitution became a new test on new standards in 2021-22. Each percentile compares a score with the same year's takers, so it stays meaningful across these changes.
• The state asks for caution with 2020-21 results because of the pandemic; that year it also dropped the rule that the exam counts for 20 percent of the course grade.
• School and district names are shown as South Carolina prints them; a few long names are cut off at 40 characters.
• South Carolina is compared at the school, district and state levels.

This site stores nothing you type. Scores are used only in your browser to draw your charts.