According to graduate students in the sciences, graduate schools are dysfunctional institutions full of excellent individuals: professors, students, and administrators. This is one of the surprising findings unearthed by the On Line Grad School Survey released recently at the University of Texas at Austin (UT) by survey authors Geoff Davis and Peter Fiske. Davis, a former mathematics professor at Dartmouth and currently a researcher with Microsoft Research, and Fiske, a physicist at Lawrence Livermore National Laboratory, launched the survey in April 1999 in the wake of the suicide of a Harvard University graduate student in chemistry.

As Davis puts it, "Why have I spent several hundred hours of my nights and weekends doing this? In part because of that suicide. Graduate education is in desperate need of some basic reforms - nothing complicated, nothing controversial, nothing very expensive. There is a pretty well-established consensus on what needs to be done - it's just a matter of doing it. The problem is, according to press reports, it took a body to motivate the Harvard chemistry faculty. The Harvard faculty had approved some modifications to protect against adverse student-faculty relations - simple things like having a three-person advising committee review a student's research progress each year. One meeting a year. Faculty had approved it, but had not implemented it. They managed to start construction on a building, hire new faculty, but couldn't make this minor reform to give students someone to turn to when things went awry with their advisor. They instituted the reforms one month after this student's death. Something had to be done."

As Davis began to research national publications on the state of graduate education, he found something curious: the reforms advocated in the student's suicide note, and other notions to improve the education of graduate students, had been around for a long time, with some reports dating back to 1969. Sensing the glacial pace of change endemic in the academy, Davis sought an alternative route - empowering the student voice through the World Wide Web. "We looked at these old reports - the AAU, COSEPUP, and others - and took the questions directly from there. We wanted to ask students if their departments were acting on these recommendations. We literally took questions right from those recommendations."

Davis enlisted the help of fellow scientist Fiske, author of To Boldly Go: A Practical Career Guide for Young Scientists. He then constructed the 46-question survey, posted it at PhDs.Org, and implemented some ad hoc publicity efforts, such as requesting HMS Beagle to promote the survey in the Webzine, and utilizing the "viral marketing" technique of having respondents recruit other participants through email. They received about 100 responses a day for 10 weeks, from April 25, 1999 to July 8, 1999. Altogether, 6,533 graduate students and recent Ph.D.s completed the survey.

The duo released their findings on November 12 at UT Austin, joined in a panel discussion by President Larry Faulkner, Dean of Graduate Studies Teresa Sullivan, Liberal Arts Dean Richard Lariviere, Dean of Communication Ellen Wartella, former graduate student assembly chair Denise Gobert, and Associate Dean and Director of the Graduate School Professional Development Program Rick Cherwitz.

Introducing the panel, Cherwitz drew a strong parallel between the aims of the survey and the intent of UT's path-breaking efforts in professional development of graduate students: "The reason for holding this discussion here is that we at UT have also been doing innovative work to help maximize graduate students' educational opportunities. UT's Graduate Student Professional Development Program is the first curriculum of its kind, offering courses in writing, teaching, consulting, ethics, and other cross-disciplinary topics to help students prepare for professions inside and outside the academy. Our Professional Development Program is essentially a graduate-level rhetorical curriculum, centering on the theoretical, practical, and productive arts of a professional scholarly communicator."

Davis agreed, observing, "Communicating is an essential part of good science. Community, collaboration, and teamwork are integral parts of scientific inquiry."

The pair opted to release aggregate data on this initial survey given the relatively small sample size and the potential for individuals to be identified by the specific nature of their qualitative responses. Fiske defended the decision: "Our goals are to spark a national discussion on graduate education, to challenge programs to scrutinize their own educational practices, and to catalyze the implementation of best practices by departments. To make real and meaningful change we need to be both careful and rigorous. We believe that the next incarnation of the survey, to be posted in January 2000, will yield response numbers that will enable us to meaningfully rank departments on these issues and provide a resource for students making initial decisions about going to graduate school."

The aggregate data still tell a compelling story. First, the good news:

• 85% of respondents are satisfied with their overall education
• 78% are satisfied with their advisors
• 76% would recommend their programs to others

However, the survey also reveals several areas of concern. Survey respondents indicated that:

• Prospective applicants do not receive sufficient information to make informed enrollment choices
• Training and supervision of TAs is inadequate
• Curricula are narrowly specialized, providing inadequate preparation for anything other than an academic career
• Career guidance and placement services are ineffective
• Important institutional safeguards against adverse student-mentor relations are lacking

A great deal of the discussion focused on student expectations and institutional duty, prompted by Fiske's observation that there are more national regulations governing the treatment of lab animals than exist for protecting graduate students. Are students getting sufficient information up front, prior to enrolling, to make intelligent choices? Are graduate programs focusing on the needs of students? Or are students there to fulfill the needs of faculty?

While the survey showed that most students had good relationships with their advisors, a substantial minority felt exploited:

• 21% of respondents feel their advisors see them as a source of cheap labor to advance their research
• 22% say their advisors expect them to work so many hours that it is difficult for them to have a life outside of school
• 28% believe graduate students in their programs are there primarily to help faculty fulfill their research and teaching obligations

These findings drew thoughtful comments from Faulkner, who spoke from multiple perspectives as a university president, tenured academic scientist, and father of two aspiring graduate students: "We do have a bias at trying to recruit person power because the engine needs to keep running. We are trying to scale the research enterprise and balance that with legitimate educational goals. There are mismatches that are created by trying to match the research needs of the national scientific agenda with the purer aims of an educational institution."

He continued, speaking more as a chemist than a president, "I will say that time to degree has gotten too long and is now unhealthy: it is drawing talent out of graduate programs and into professional programs where time to degree is fixed, like law, business and medicine. Simply, time to degree in Ph.D.s is ridiculous and ought to be brought down by two years or more."

As a parent, Faulkner was concerned about programs providing information up front, helping to set realistic expectations for prospective students. "Young people ought to know what they want from grad school before they start. They need to enter for reasons beyond not wanting to go to work or not knowing what else to do," said Faulkner.

Providing better information to prospective students before they enroll about exactly what they are getting themselves into is a simple way to set realistic student expectations, the panel agreed. Some went even further, asserting that helping prospective students to make informed decisions about their careers is an academic duty. The survey found that universities are not doing a good job of providing that information:

• 37% of respondents report that their programs informed them of where recent program graduates were employed after graduation
• 29% of respondents report that their programs informed them of the percentage of students who complete the program with a Ph.D.
• Only 50% say their programs provided enough information for them to make an informed decision about choosing to pursue a Ph.D.

Davis and Fiske further challenged UT to extend its innovative leadership role in empowering students, as firmly grounded with the Professional Development curriculum, by providing comprehensive information to help students make better choices, and by exerting pressure on national funding organizations (National Science Foundation, NASA, National Institutes of Health) to make this transparency part of the criteria for eligibility to receive grant funds.

Faulkner concurred. "In the U.S., we've got large amounts of government-funded research and the need for folks to do the work, so size is being driven by other elements of the university and the size of the research agenda. We've got lots of research being funded in life sciences, so we need research students, but then there is no market for Ph.D. instructors once they finish."

Vice President Sullivan echoed his comments. "Graduate education should not be a by-product, and it should be a consciously designed activity with a consciously designed size. If this survey is a way for us to improve the quality of education that we are providing for our students, then we want to encourage participation and the reforms implicit in its findings."

Although he left the academy two years ago, Davis still has an attachment to his old line of work: "While the values held within the scientific community are noble ones, the institutions in which we operate do not always do a good job of promoting these values. The result is that our community is not as successful as it could be at living our shared ideals."

Complete survey results are available online. The National Association of Graduate and Professional Students will be conducting a follow-up survey in January 2000, with a goal of generating a 30% response rate, enabling survey authors to meaningfully rank departments and set up an ongoing process to provide comparative data to prospective students.