Report urges NSERC to adopt Conference Model to review Discovery Grant applications

A new report to the Natural Sciences and Engineering Research Council (NSERC) is recommending that the process for evaluating applicants to the Discovery Grants Program (DGP) be overhauled. The proposed changes would provide greater consistency across disciplines and an enhanced ability to properly handle research in emerging areas and areas that cross traditional disciplinary boundaries. The report further urges the granting council to base its granting decisions on the cost of research and not the researchers need for funding.

The report's main recommendation is to adopt a so-called Conference Model for the DGP, replacing the existing 30-year-old structure of 28 discipline-based committees with 10 or 12 panels that can be easily configured into highly targeted sections.

"The Conference Model provides the best expert review of grant applications and that means getting specialists for each one. We don't want to keep splitting committees to get more specialized but that was where we were heading and it was already happening in certain areas like electrical engineering," says Dr Adel Sedra, dean of engineering at the Univ of Waterloo and chair of the DGP Structure Review that produced the report. "Further splitting would produce highly specialized silos which are fragmented and have no coherence."

The Conference Model would involve panels with at least 30 members organized along thematic lines. Those panels can be split thematically into topic groups that can decide on specific areas of research.

"This represents a paradigm shift. There will now be flexibility to accommodate emerging areas of research. It's a significant improvement," says Isabelle Blain, NSERC's VP research grants and scholarships directorate. "There's room to improve and add depth in areas where the GSCs are thin and the Conference Model will allow us to add depth to the review of each applicant by mixing and matching committees."

The DGP is NSERC's largest research grants program which dispensed $328 million to more than 10,000 researchers in 2007. It recently received a solid endorsement from an international review panel headed by Dr Peter Nicholson which recommended that it receive additional funding to maintain average grant sizes in real dollar terms (R$, May 20/08).

Sedra says his committee supports the recommendations in the Nicholson report and that the two reports are complementary and mutually reinforcing. Both reports, for instance, support the categorization of DGP applicants into bins ranging from strong recommendation for funding to a recommendation against funding.

Sedra says he's confident that being advocated will help to eliminate the concerns of a significant minority of researchers who feel the committees reviewing their applications do not possess the necessary expertise. A web survey conducted for the review revealed that 14% of funded applicants responded that their GSC did not have the necessary expertise — a percentage that increased when the same question was posed to unfunded applicants.

address concerns over appropriate expertise

When asked whether current or emerging areas were not handled well by the current system, 31% of the 4,500 respondents said yes. The areas of research most frequently cited as examples were bioinformatics, biomedical engineering, biomedical technology, cognitive or neurosciences, environmental sciences, microbiology/ microbial ecology and nanotechnology/ nanoscience.

"We owe it to the applicants to provide the best possible review. That's the beauty of the Conference Model," says Sedra.

If the recommendations are enacted, Sedra says the result will be a system that is more responsive to changes in performance, reducing the inertia that exists in the current system which often bases the size of a grant on the level of support received in the past.

"The system is now highly dependent upon the value of the previous grant and we want to move away from that," he says. "This is different from the track record (of the researcher) which should be considered.

The Conference Model is currently used by a handful of GSCs. Blain says there are two or three other committees that have split and can be merged together again in time for the 2009 DGP competition. The remainder will likely be ready to implement the new model in time for the 2010 competition.

Blain says NSERC will need to add up to 50 more reviewers to the current 330. That offers an opportunity to implement a recommendation from the Nicholson report, which is to bring in more international members and members from outside the Canadian university system. She says decisions on the number of panels and sections should be made by this fall.

Work will also continue to coordinate activities with the other granting councils to better accommodate researchers whose applications fall at the interface between councils.

Sedra concurs with the Nicholson report that the DGP is world class system but adds that constant improvement is essential to maintain that excellence.

"We want to improve it and protect it from stresses that would make it less effective," he says. "The new structure will be much better."

R$