So That's Why They're Leaving

In the past half year, Congress has held hearing after hearing to discuss the so-called American “science pipeline,” which some legislators say is leakier than warped rubber tubing.

July 26, 2006

In the past half year, Congress has held hearing after hearing to discuss the so-called American “science pipeline,” which some legislators say is leakier than warped rubber tubing.

The senators and representatives and the witnesses they've brought before them have offered a rainbow of possible reasons why a smaller proportion of American undergraduates are majoring in physical sciences and engineering than in the past. Some have suggested that biology is the hot science and has lured many top students away from other fields. Others have suggested everything from poor undergraduate instruction to nerd stigma. “Bright girls play dumb and guys can’t get dates,” said Rep. Roscoe G. Bartlett, a Maryland Republican on the House Science Committee.

But, in all those hearings, Congress folk may have missed some of the low-hanging fruit, according to interviews with a range of scientists and experts on science education.

Greener Grade Pastures: Science students get worse grades than non-science students. No comprehensive data for the distribution of grades around the nation by discipline exists, but in 1998 the College Board surveyed a representative sample of 21 selective institutions to find out how students who took Advanced Placement courses in high school were performing in college. The data show that, when students who got AP credit and were taking second-level college courses (as opposed to intro classes) were compared, non-science students got much better grades.   

In English courses surveyed, 85 percent of those high-achieving students that were surveyed received A's or B's. That’s compared to 54 percent of those students in math courses.

Paul Romer, an economics professor at the Graduate School of Business at Stanford University, who has studied the issue, wrote in an article for Stanford Business that “the grades assigned in science courses are systematically lower than grades in other disciplines, and students rely heavily on grades as signals about the fields for which they are best suited.” Thus, he concluded, students usher themselves out of the science track.

Data from the Higher Education Research Institute at the University of California at Los Angeles show that, in 2004, about 9 percent of freshman students nationally planned to major in engineering, and 2 percent planned to major in physical sciences. Those numbers are pretty typical for the last two decades, and what is also typical, according to National Science Foundation data, is that it is not uncommon for fewer than half of those intended majors to stay the course.

It seems that the attrition rate in the physical sciences and engineering is chronically higher than in social and behavioral sciences. According to the NSF, only about 4.5 percent of bachelor’s degrees were awarded in engineering in 2004, and only about 1 percent in the physical sciences. Conversely, depending on the demographic, generally between 8 and 15 percent of freshmen intend to major in social and behavioral sciences, for which degrees made up 16 percent of the 2004 total.

Romer isn’t the only one that thinks unequal grading practices drive students from science. Ronald G. Ehrenberg, director of Cornell University's Higher Education Research Institute and an economics professor there, recalled a student who got an 85 on a test, which was above the mean, coming up to him and saying, “I’m dropping your class, because the best I can do is an A-, and I’m going to Stanford Law School.” Part of the problem Ehrenberg said, is that students who want to keep law school as an option will tend away from quantitative courses because it’s clear to them that disproportionate grade inflation in the humanities and less quantitative social sciences will give them a boost.

With Web sites like, students can instantly find out how “easy” other students think a certain professor is. A 2002 Cornell Higher Education Research Institute study showed that grades in Cornell's science courses are generally several tenths lower than other courses, and a 2005 institute study found that, presented with information on the grading, students will flock to the easier courses, driving grade inflation even more.

In 1996, worried that they were giving lower grades than professors at competitor institutions, faculty members decided that Cornell should publish the median course grades for every course, every semester, so that faculty members could see the distribution of grades, and, presumably, adjust if a particular course’s median grade is too low. Not surprisingly, students started turning to the list, and according to the 2005 institute study, the list started looking different in a hurry, as students migrated en masse to easier courses. By spring 2005, the list shows that, of over 1,300 courses, fewer than 20 had median grades of B- or lower.

Weeding Out: Several experts suggested that the culture of scientists has kept science grades down, while science students at many institutions have watched longingly as humanities grades have drifted up and away like a helium balloon.

“There’s a difficult culture here,” said Daryl Chubin, director of the American Association for the Advancement of Sciences’ Center for Advancing Science & Engineering Capacity. “The culture of science says, ‘not everybody is good enough to cut it, and we’re going to make it hard for them, and the cream will rise to the top.’ ”

Ehrenberg said that some scientists are starting to drop the "weed out" mentality, but Chubin still sees decade old themes. “I took a Ph.D. in 1973,” Chubin said, “and people were saying the same thing then. ‘Look to your left, look to your right, some of you will be gone.’ There’s a joy of attrition; demonstrating your manliness, back then it was all manliness, by failing students.”

Beyond tough guy science education, though, the 2002 Cornell study also pointed out that class size affected grades.

Large and Impersonal: The study showed that, controlling for the effects of discipline, large courses at Cornell had lower median grades than small courses.

The study suggested that large courses might have lower grades because the grading tends to be more coldly quantitative, so students who perform poorly on a major exam may have little recourse to bring their grades up. Whereas, in a smaller course, with more personal interaction, a student might be more likely to get advice on how he or she can improve directly from a professor, and might be rewarded for things like contributing to class discussions. Unfortunately for the science pipeline, many intro science and math courses tend to be large, while English courses, for example, tend to be small enough for a productive discussion.

But grading isn’t the only thing that shoos students from disciplines with large introductory courses.

Math and Science Goes Vertical: Math and science are taught “vertically,” meaning students are often made to slog through two years of large, formulaic introductory courses that teach fundamentals before they get any taste of the hands-on work that makes a career in science attractive to most scientists. In the process, students seldom form any bond with the scientists teaching the course.

Science departments are “daring students to persevere and earn a degree,” Chubin said. He added, though, that there are faculty members swimming upstream.

Richard Losick, a biology professor at Harvard University, said that students can come to think that those impersonal intro courses represent a career in science. Losick goes out of his way to get freshmen and sophomores into the lab. “If a student’s in a lab,” Losick said, “they’re going to get to know a professor and grad students and post docs … instead of a sea of hundreds of students taking chemistry.”

Signs of Hope

Some institutions are attacking disproportionate grade inflation head on. Princeton University has told all departments that they should work toward having 35 percent of grades be A's. Only the natural sciences were around that before the 2004-05 academic year, the first with the new policy. “One of the results we hope for,” said Nancy Malkiel, dean of the college at Princeton, “will be that if the grades in the natural sciences are comparable to the grades in other fields, students won’t be deterred from entering the natural sciences because of grades.” In the first year under the policy, other disciplines made major headway toward reaching the natural sciences.

Losick added that he sees a national trend of scientists, problem solvers that they are, trying to teach intro courses in a more engaging, interdisciplinary way, so, for example, a biology student can learn chemistry while getting some of the bio that he or she is interested in. “We launched a new course last year,” Losick said, “that integrates chemistry and molecular biology.… It has themes in which some biological problem is tackled from multiple points of view.”

If the science pipeline is to be shored up, Chubin said, it’s clear that Congress can only do so much. “The faculty has to own this,” he said.


Be the first to know.
Get our free daily newsletter.


Back to Top