ITHACA, N.Y. — ITHACA, N.Y. -- Alison Amsterdam, disappointed with the less-than-perfect grade she got in an introductory computer course, dropped out of engineering classes at Cornell University after her first semester and began searching for a major in the humanities.
Similar circumstances worked against Brandon Coors. Unable to understand her professor's foreign accent or see the benefit of spending her weekends in the lab dissecting electrical currents, she gave up her Air Force scholarship and, with it, her dream of becoming a biomedical engineer.
Both women are bright, and both say they were "in love" with math and science in high school, where they were A students. But in college, they said "no" to engineering.
Women represent the largest pool of desperately needed new science talent for the nation's universities and industries. But as the proportion of white men and overall interest in doctorate degrees decline, women are not rushing to fill the gap.
Among the reasons: the fear of being labeled a "nerd"; a sometimes tedious curriculum that doesn't allow for much of a college social life; subtle messages in the classroom that women don't belong there; and, as researchers are beginning to discover, fragile self-esteem.
Although women have been earning more than half of all undergraduate degrees since 1982 and have had major success entering medicine and biology, where they now account for half the degrees, they remain outsiders in chemistry, physics and engineering. About 7 percent of the engineering work force is female.
Nationwide, the percentage of women choosing to major in the natural sciences and engineering has stalled or is dropping. At this rate, 1 million jobs in engineering and science will go unfilled by 2003.
Even at Cornell, a leading producer of undergraduates who go on to stock those research labs, a peek into the engineering classrooms raises this question: Where are the women?
The situation has so alarmed the National Science Foundatio that it has rushed $30 million to a dozen of the country's top engineering schools -- among them Cornell, Stanford, Penn State, and the University of Maryland College Park.
Around the country, experiments to attract women to science range from a residence hall for female science majors at Rutgers University in New Jersey to a drive to train 1,800 Girl Scout leaders in three states to help their 71,000 troops earn badges in science.
If attracting students like Ms. Amsterdam and Ms. Coors is becoming a national obsession among groups like the NSF, the mostly male faculty in these disciplines have been slow to make women feel welcome.
Three weeks ago in a private tutoring session, Cornell sophomore Chrystal Brooks says her male teaching assistant told her that he doesn't encourage women to major in math or physics because, he said, women don't have the aptitude.
"We'll see about that," she said she told him.
The roots of the problem are deep.
"Probably every boy has received [the game of] '1,001 experiments you can do as a kid,' " said Rita DiPaulo, a College Park physics major. "My brother got that. I went out and bought it last summer."
If the crisis is to be overcome, those working in the field say educators first must learn why women are leaving.
What turned off Ms. Amsterdam and Ms. Coors is what turns off countless talented women needed to build the nation's bridges, keep its industry competitive and clean up the environment:
* Initial grades that paled beside their achievements in other subjects and made them question -- more than their male peers -- whether to continue.
* An inability to see the long-term social relevance of years of grueling and sometimes tedious coursework.
* A feeling that they didn't belong in a field dominated by men.
Like many college-age women enamored of science, they started less well-prepared in math than many of their male peers. In their introductory math classes, they found mostly male and often difficult-to-understand foreign teaching assistants and professors.
L In the end, they concluded the extra effort wasn't worth it.
"I had no social life. I'd study three days for a test and still get a zero," said Ms. Coors. She has since joined a sorority, signed up for classes in international affairs, and doubled her grade-point average. Now, she says, "I'm very happy."
Without any special effort, Cornell has attracted more female engineers than most campuses -- roughly 20 percent in recent years compared with a nationwide average of 15 percent. But a drop in enrollment to 17 percent last year prompted alarm. The College of Engineering has a goal to increase the number of women to 35 percent.
"We've long had programs for minorities," says William B. Streett, dean of engineering. "We realize now we need to do more of that for women."
Beyond the sometimes subtle societal messages that say women can't or shouldn't do science -- and perhaps because of them -- something else is working against women, according to experts who are beginning to study the phenomenon.
It's called internalizing failure.
Inevitably, says Jane Daniels, a counselor at Purdue University in West Lafayette, Ind., "when a woman student gets a C grade, it seems she takes it personally, where a lot of our male students get C's and go merrily on their way."
Internal doubt is why women decide more quickly than men that science is not for them, she said. Ironically, some of those who leave science are just as smart or smarter than the men who stay.
A 1975 study at Cornell found that women attracted to engineering tended to be smarter and more academically motivated than the men. The dropout rate for women engineers at that campus is 35 percent, down from 43 percent in 1983. The comparative rate for men is 27 percent.
"You tend to get these superstar women who question themselves," says Margaret Palmer, who teaches science majors about the issue of self-esteem in her classes at College Park.
About 21 percent, or 1,324 students, in Purdue's engineering school, one of the largest in the country, are female, up from 2 percent a decade ago. They are admitted on the same competitive basis as males, and they graduate at the same rate.
With money from the NSF, Dr. Daniels of Purdue has been traveling nationwide to help universities attract women. The most successful programs, including her own, include outreach to high school girls and feature counseling sessions on self-esteem.
Compounding the problem is a lack of female role models, and nowhere is that more apparent than in physics.
Cornell, College Park and the Johns Hopkins University each have two female physicists on their faculties. Most universities have none, and at the current rate, each university in the country would get a female faculty member every 19 years. Of 4,600 professorships at major research universities nationwide, 100 are held by women.
After getting more women in the pipeline, "I think self-esteem is the most important issue," says Susan K. Watson, who will earn her doctorate in physics from Cornell this year.
Ms. Watson, the top undergraduate in her class at the University of California at Berkeley, says she overcame doubts in the first two years of graduate school by joining a group of Cornell female graduate students and faculty who meet weekly to share research, debate science questions and discuss issues of gender.
Karen Miller, president of the Cornell Society of Women Engineers, is working to change the image of engineers by making the group more of a social club. "One of my goals is to make it more fun," she said.