And you only have to get into Harvard once, not every three to five years. This is incredibly depressing:
And the NHLBI:
The NHLBI has established an interim payline at the 6th percentile (inclusive) for investigator initiated R01/R21 research grant applications. Initial paylines will be determined for other funding mechanisms – such as program project grants, training and career development grants, and small business grants – later in the fiscal year.
(NHLBI is hoping that they can get the payline back to ten percent. Good luck with that.)
I’ve been fortunate to have received funding, but the times I’ve been involved with a grant that hit the sixth percentile or better are very few and far between. Not only is this incredibly difficult, but it completely weeds out any controversial grants, since every reviewer has to really like the proposal or it will be dead. With novel grants, there is often one reviewer who has doubts–precisely because it’s novel–and will give it a lower score.
Because what we need are fewer novel approaches and ideas.
“there is often one reviewer who has doubts–precisely because it’s novel–and will give it a lower score.”
I’ve often thought that reviewer panels should work in reverse, so to speak, to avoid exactly this situation. Perhaps each panelist should be allocated a certain number of *individual* funding recommendations, and a limited number of vetoes to nix truly stupid ideas. Then, if one panelist likes an idea that seems rather out there, it still can get funded, even if others are less enthusiastic.
I often explain my frustration with the current system as follows: suppose you right the best grant of your life, working hard on it, such that 80% of the people who read it will give it a “top” recommendation. Sounds great, until you take 80% to the fourth panelist, er, power, and now the best thing you’ve ever written has a 40% chance to actually get top ratings and funded.
This is an oversimplistic calculation, of course, but the way I see it, the more approvals a grant needs, the more random the process becomes.
Oh, and more on topic to this post: the sequester is bullshit and horrible and will hurt science and the U.S. for years, if not decades, to come.