With foreign students making up more than half the pool of US physics graduate students, it’s not surprising that educators are worried by the seemingly disproportionate number who switch, mid-PhD, into areas such as computer science and engineering. Students from the People’s Republic of China (PRC) are the biggest block of foreign students and are therefore the focus of concern over, and efforts to turn around, this budding trend. “I think that we, as faculty, are frustrated because we see students who could have gone on to a physics career,” says Moses Chan, a physics professor at Pennsylvania State University. “And we love physics, so we think it is a waste.”

The scope of the field-jumping is far from clear, but in the past few years, the topic has come up at meetings of physics chairs and within departments. Some departments, prompted by PRC student departures, have changed their PhD programs to improve the retention rates of all students. Others have become more reluctant to admit PRC students. Indeed, their numbers have dropped over the last decade, although PRC students still make up more than one-fifth of the foreign physics graduate student population.

Students from everywhere, including the US, switch fields, of course. And better job opportunities in other fields can draw them away from physics. But professors also suspect that some PRC students arrive with the intention of switching fields. “Departments like physics, math, and chemistry have teaching assistantships available—and it’s easier for students with assistantships to get a visa to come to this country,” says Chan. “I think there are definitely some who come using physics as a ticket.”

These days, about half of the handful of physics students who come each year from the PRC to Penn State don’t complete physics PhDs there, Chan estimates. We also lose a couple of domestic students each year, he says, “but for different reasons.”

“The ultimate problem is that we don’t have enough well-prepared domestic students,” continues Chan. “The worldwide pool of physics students has gone down—even in China it may be dropping.” The physics department at Penn State receives about 300 applications each year—about half from China, and one-quarter from the US.

The situation at Penn State is certainly not unique, but departments differ. Until recently, the out-of-state fee structure at the University of California, Santa Barbara, made foreign students more expensive for the department, so only 10% of their physics graduate students were from abroad. This fee structure is now changing, and the department will try to push that figure up to 20%, says Phil Pincus, a member of the physics department’s admissions committee. Culturally and academically, he adds, “It’s an advantage to have foreign students.”

But choosing the best, brightest, and most dedicated students can be difficult. The UCSB physics department received a couple hundred applications from the PRC this year. “We just don’t know how to deal with them,” says Pincus. A student might score 99% on an English language test, but speak poor English on arrival; students might write the letters of recommendation because their professors don’t speak English; and the graduate record examination (GRE) scores from China are so universally high that they fail to differentiate between students.

In fact, China’s GRE scores are high enough that the Educational Testing Service (ETS), which administers that exam, became suspicious. ETS is now suing, in Chinese courts, the most popular test preparation service there, Beijing’s New Oriental School, for copyright infringement on old test materials. The ETS also alleges that the school has obtained current GRE test questions and passed them on to students. “There was a sharp increase in GRE scores during the time that we allege they were sharing questions,” says Tom Ewing, a spokesman for the ETS.

With unreliable GRE scores and recommendations, admissions committees frequently turn to the graduate students they already have for advice, says Zhang Chen, a physicist at Manhattanville College in New York who experienced this when he was working on his PhD at Columbia University. If there are already students from, for example, a university in Shanghai, then more students will be admitted from that same university. “It’s very random. If students in China knew how it was done, they would be very disillusioned,” says Chen.

The first big wave of physics students from China came to the US in the early 1980s, through the China-US Physics Examination and Application (CUSPEA) program. CUSPEA did not use the GRE exam—it was not yet available in China—but instead created its own test and followed up with interviews in China. Spearheaded by Columbia University’s Nobel laureate T. D. Lee, who was influential in inspiring many Chinese students to go into physics, the program admitted about 100 students each year for a decade. CUSPEA was discontinued in 1989, after other methods of evaluation had opened up.

CUSPEA participants had a high PhD completion rate. “Those students are now making a very important mark on physics in this country,” says Penn State’s Chan. But because many hit the job market in the lean days of the early 1990s, “some are working in areas like Wall Street,” says William Que, one of the first CUSPEA students, who is now a physicist at Ryerson University in Toronto. “And some are working in engineering companies. Actually, not that many are still in academia.”

Bleak job prospects continue to be a major reason for students leaving physics early in their careers. While some PRC students may use physics as a means to get to a US university, Cornell University graduate student Nai Gong Zhang says he doesn’t believe it’s “a giant conspiracy. Young people [from China] were shocked to find it’s hard to get a job as a physicist.”

To retain more students, the University of Southern California, where more than three-quarters of each year’s 10 or so foreign graduate students in physics are from the PRC, changed its PhD program. Tu-nan Chang, chair of USC’s physics department, had noticed that each semester some of the Chinese and eastern European graduate students would drop physics courses for engineering ones. After a couple of years, they’d accumulate enough credits to switch fields and leave with a master’s degree in engineering. Now students are encouraged to start their PhD research early, and take a wider variety of courses later on. They can still earn an engineering master’s, but they more often complete the physics PhD as well.

Chang, who is also president of the Overseas Chinese Physics Association, says interest in introducing CUSPEA-like programs is growing around the US. They would probably run at the institutional level, and be broader than just physics. “You’ll be able to establish a university-to-university relationship [between US and Chinese institutions],” he says.

In fact, that’s what Columbia University, New York University, and the City College of New York have been doing with Fudan University in Shanghai since the mid-1990s. Through this program, more dedicated students are selected, so fewer switch to other disciplines.

“Over the last 20 years, the PRC has produced some of the best students,” says Chan. “After all, China is a big country, with 1.2 billion people. There have to be some people who love physics.”