New U.S. high school transcript data show stark and growing racial differences in which students progress to the most advanced mathÂ subjectÂ in high school: calculus.Â It appears from this first release of dataÂ that, among black students who started high school in 2009,Â a slightly smaller proportionÂ took a calculus class than four years earlier. Â And even in the earlier cohort, only about 6 percent of black high school graduates took calculus.

Meanwhile, growing numbers of Asians, whites and Hispanics are taking the subject.

Strikingly, 45 percent of Asian-Americans who started high school in 2009 took a calculus class before graduation. That’s more than double the rate of white students, ofÂ whom 18 percent took calculus in high school. Ten percent of Hispanic students and fewer than 6 percent of blackÂ studentsÂ took calculus.

The biggest increase in calculus instruction is taking place in U.S. private schools. Nearly one-third of private school students who finished high school in 2013 took calculus, compared with 23 percent of private school graduates in 2009. At U.S. public schools, which saw giant boosts inÂ calculus instruction between 1990 and 2005, the percentage of students taking calculus seems to haveÂ leveled off. Among all public school students who started high school in 2009, almost 14 percent took calculus.

Calculus course-taking data was part of a larger report, “2013 Update and High School Transcript Study: A First Look at Fall 2009 Ninth-Graders in 2013,” released on June 25, 2015 by the National Center for Education Statistics.Â The Department of Education is following a nationally representative sample of 20,000 students who began ninth grade in 2009 to learn more about the choices students make in high school, and later in college, and about how some students decide to study science, technology, engineering and math (the so-called STEM fields). Following graduation in 2013, high schools submitted each student’s transcript to Washington for analysis.

Calculating how many high schoolers take calculus and who takes calculus is important. Having it on your transcript can be a huge help in gaining entry to the nation’s most selective colleges and universities. Like Latin a generation or two ago, high school calculus is often seen as a proxy for intelligence.Â And it’s more difficult to pursue a science, math or engineering major once you get to college if you don’t already have basic calculus under your belt.

Also interesting is that there are virtually no gender differences in who takesÂ high school calculus. Just as many men as women — 15 percent of each gender — took it. Yet, surveyÂ data from the same report show that more than double the percentage of male studentsÂ (33 percent) planned to pursue a STEM major in college than female students (14 percent).

In addition to race, the study correlated family income and the educational level of a student’s parentsÂ with who takes calculus.Â Certainly, there is a large gap in calculus-taking between the haves and the have nots. Unfortunately, this first report didn’t release figures on family income, but only on what it called socio-economic status, which combines income, parents’ educational attainment and the prestige of a parent’s job. (For example, an adjunct university professor might be low-income, but high prestige and put him and his family in a higher socio-economic status category.) Among students from the top fifth of the socio-economic spectrum, more than 30 percent studied calculus. From the bottom fifth, only 6 percent did.

Having well-educated parents helps a lot. More than 30 percent of students whose parents had a master’sÂ or doctorate degree studied calculus. About a quarter of students whose parents’ highest degree was a bachelor’s did.

AmongÂ parents who were high school dropouts, high school graduates only or held two-year associate’s degrees, fewer than 10 percent of their children progressed to calculus.

The report didn’t speculate on any reasons for the racial and socio-economic differences in calculus course taking. But the report confirms that students are getting sorted into a calculus track before they start high school.Â In order to progress to calculus, most students take algebra I, geometry, algebra II and pre-calculus first. That’s four years of instruction prior to calculus, requiring students to begin with algebra I in eighth grade to get to calculus by 12th grade. As a result, only 4 percent of students who waited until ninth grade to begin algebra I managed to catch up to calculus in four years, a feat that would generally require independent study or summer school.

During this 2009-2013 period, schools around the countryÂ wereÂ experimenting with pushing more students to start algebra in 8th grade. About 6 percent of school districts requiredÂ it, including some low-income districts. But there was alsoÂ concern about pushing unprepared kids into algebra too soon, setting them up for failure. In 2013 California reversed its plan to require algebra in eighth grade.

Still, even among students who started algebra in eighth grade and began geometry in ninth grade, only 36 percent of them made it to calculus. What happened to the remaining 64 percent of them? Did they stop taking math senior year? I’ll leave it to academic researchers to dig deeper into this transcript data to understand the hemorrhaging of our STEM pipeline.

** A note on historical comparisons. Historical data on calculus course taking from 1990, 2000, 2005 and 2009 can be found here. But these older figures and the new June 2015 report are not measuring exactly the same thing. For the older figures, percentages of calculus takers are among high school graduates, not all students who started high school. It excludes high school dropouts. The new June 2015 report included every student who began high school in 2009, including dropouts, of which there were 4 percent. An additional 4 percent of the students were still in high school, but hadn’t graduated yet.Â Â So the change in calculus taking from 15.9 percent in 2009 to 14.9 percent does not necessarily mean there was really a decline in calculus taking. It might have been a slight increase. Â But the fact that calculus course taking among Asian-Americans grew from 42 percent in 2009, excluding drop outs, to 45 percent in 2013, including drop outs, is a sign that more and more Asians are indeed taking the subject.*

[Cross-posted at The Hechinger Report]