Interested in reading more?

It Pulled The Plug On Two Engineering Centers, Provoking Debate Over Its Program Goals The National Science Foundation’s announcement in 1985 that it hoped to set up a network of up to 25 university-based engineering research centers set off a frantic scramble to snare a center—and a roiling debate about the value of the idea. After all, Director Erich Bloch’s vision to spend a half-billion dollars over the next decade on projects intended to improve both U.S. industrial competitiveness and the training given the next generation of engineers raised the pulse of academics accustomed to NSF’s strong support of individual researchers. Three years later, the passion hasn’t abated. But disappointment and disillusionment have replaced the initial excitement. In March, the NSF decided to withdraw its support over the next two years to two of the six original centers, saying they were "doing a poor job." The centers, in turn, have charged that the fault lies with NSF’s management and that politics played a major role in the decision. NSF officials admit they are still searching for the best way to evaluate the program, and outside observers say it is impossible to determine objectively whether the two centers were performing well. In fact, many scientists believe that such interdisciplinary centers on university campuses (NSF currently supports 14 sites) represent a novel approach to basic research that should not be judged by the usual standards. "The Engineering Research Centers have been oversold," says Don Kash, research professor of science policy at the University of Oklahoma and chairman of a National Research Council committee that has studied such centers. "And they are being overevaluated. Put simply, no one yet knows how to make them a success." Congress has also become embroiled in the debate. On March 22, the House Science, Space and Technology Committee discussed the centers as part of a review of several NSF programs, and budget hearings have also focused on the program. In September, the General Accounting Office is expected to deliver a report to the Senate Commerce, Science and Transportation Committee on NSF’s management ot the program. And the National Research Council is weighing a request from NSF to form a successor to its committee on crossdisciplinary research that has been advising the foundation for the past three years. The latest controversy stems from a decision by NSF to award new five-year grants to four of its original six centers and to phase out those at the University of California, Santa Barbara, ,and the University of Delaware. The two centers were actually in quite different situations: Santa Barbara’s Center for Robotics Systems in Microelectronics had been created in 1985 with a promised $14 million in NSF money over five years, while Delaware’s Center for Composites Manufacturing Science and Engineering was started back in 1974, and only expected to receive $7.5 million from NSF. But both centers reacted to NSF’s decision the same way—angrily. "They needed to make cuts for two reasons," charges George Munson, director of industrial relations at the Santa Barbara center. "They’re facing budget pressures, and they felt their image in Congress required them to show due diligence in their oversight of the program." Not true, counters Nam Suh, who oversaw the program as assistant director for engineering at NSF before returning to MIT’s department of mechanical engineering in February. "We weren’t trying to guess how Congress would react, and we weren’t playing politics. My job was to worry about the overall health of the ERC program, and to run it in such a way that the taxpayers would be willing to support it." NSF made its decision as part of an extensive and ongoing review of the program. Last fall, teams of NSF program directors and a half-dozen outside reviewers visited each of the six sites, and their reports all recommended continued NSF support. A committee of the National Science Board chaired by Warren Baker, president of the California State Polytechnic University, received a report from NSF officials that essentially agreed with the outside evaluations. The report suggested that the board defer a final decision until the winter of 1989. This would allow time to monitor steps that the centers were taking to resolve questions of focus and research quality raised by the site review teams. But the committee decided to take immediate action. It recommended to the overall board that funding for both centers be phased out. NSF officials say the results of the evaluations are clear-cut. The work performed by Santa Barbara scientists, "was less than excellent and failed to meet an acceptable level," says Baker, Suh is even more emphatic: "There was a clear consensus that four of the centers were doing a great job and that the other two were doing a poor job." Naturally, the centers disagree. Munson contends that Santa Barbara answered all criticism fully by giving the site review team data on publications, presentations, and collaborative efforts that demonstrate the high quality of the work being carried out. Delaware officials, preferring to speak anonymously, are no less outraged. "I don’t think that NSF really knows what the criteria should be," says one administrator. "They’re still trying to decide, and the way they will do it is to adjust their criteria after each round of renewals and cuts." The ruckus raises all sorts of questions about how to evaluate interdisciplinary centers. "We know what’s good within a particular discipline," says Kash. "But in crossdisciplinary fields, the question of the appropriate yardstick to use in measuring the quality of research is inherently difficult. The use of an evaluation scale that varies according to budgetary and other external factors could doom the centers, said Kumar Patel of AT&T Bell Laboratories in testimony before the House panel that discussed the centers. Budget Pressure "There’s incredible pressure to evaluate," he said, "but you must strike a balance between allowing an experiment to continue and being a tough, cost-conscious manager. Budget pressure makes it worse. It’s easy when budgets are expanding, but it’s hard to know where to cut when the money’s tight." But can the centers even do what NSF envisions? Kumar echoed the concerns of many scientists when he told the House panel that quantity is no substitute for quality. "Some people have the idea that the way to advance science is to bring together a group of individuals for a common purpose," he said. "But two second-rate engineers don’t equal one first-rate engineer." The next step is a decision on a new round of centers, expected to be announced in June. NSF offi- cials originally wanted to add four more, but will have to settle for two because NSF received much less budget support than it had re- quested for fiscal 1988. But two new centers won’t end the debate. Kash told the House science panel that a minimum of 25 engineering centers, NSF’s original goal, is needed to create a network that can eventually overcome institutional hostility to such interdisciplinary work. However, few people expect NSF to receive the 20% increase it has requested for fiscal 1989, which begins October 1. ‘A Quality Issue’ The process continues this fall, when the five centers created in 1986 face their own third-year review. Baker doesn’t think the board’s actions held out any lessons for other centers. "It’s primarily a quality issue, and not so much the type of work you’re actually doing," he says. But Kash does offer up one piece of advice. "Our committee thought it perfectly acceptable for NSF to say ‘no’ to a center that wasn’t doing what it had said it would do,he says. "But remember that this is an experiment. The process of trial and error means there will be mistakes. The worst thing you can do is to evaluate them too early, before they’ve had the chance to show what they can accomplish."

Ruckus Over NSF Grant Reversal

May 15, 1988

Interested in reading more?

Become a Member of