Current Events > Luck in early grants predicts a scientist's success in future grants and also...

(+) Yeah! (+)

...future professorships.

http://www.pnas.org/content/early/2018/04/18/1719557115
https://twitter.com/thijs_bol/status/988495113050640385

Pretty dismal =P
---

Is that because early grant winners are particularly talented/intelligent? Or is this a case of getting more funding the more cited your works are and having that create a feedback loop?

EDIT: Actually, the abstract partially covers it:

Surprisingly, however, the emergent funding gap is partly created by applicants, who, after failing to win one grant, apply for another grant less often.

---

s0nicfan posted...

Is that because early grant winners are particularly talented/intelligent? Or is this a case of getting more funding the more cited your works are and having that create a feedback loop?

No, they examine those just below and above the funding threshold, so the grants were more or less equally good, with luck and subjectivity heavily playing a role in the initial funding/not funding. Secondly, they find that changes of productivity post grant funding doesn't explain the difference in later funding.

It's moreover that having had a good funding track alone increases your ability to get grants.
---

Bumpo
---

They suggest many more smaller grant opportunities, which may blunt this effect by making earlier grants much more likely.

Others have suggested an entirely different way to reduce grant money from pooling in single successful labs, such as selecting a cut off, and then simply selecting randomly from projects that satisfy that threshold (which seems to work in simulations, at least).
---

COVxy posted...

They suggest many more smaller grant opportunities, which may blunt this effect by making earlier grants much more likely.

Others have suggested an entirely different way to reduce grant money from pooling in single successful labs, such as selecting a cut off, and then simply selecting randomly from projects that satisfy that threshold (which seems to work in simulations, at least).

I'm reluctant to support this approach, in part because past success can (and should) impact selection given equally qualified submissions. Going fully random also incentivizes submitting many similar requests to try and game the system. A single successful lab may be sucking up a lot of funding, but if they have significant momentum they can make major breakthroughts that smaller, "starting fresh" labs couldn't.
---

What do you mean by luck? Do you believe grants are given out buy coin toss? Or are they givien out based on how promising a researcher seems to the people giving out grants?
---

s0nicfan posted...

A single successful lab may be sucking up a lot of funding, but if they have significant momentum they can make major breakthroughts that smaller, "starting fresh" labs
couldn't.

This isn't what the data suggests. In fact, if anything, if you look closely, grant pooling creates stagnation of ideas, as a single lab usually only contributes to a single specialty with usually a very particular theoretical stance which biases the research.

From anecdotes of PIs, it seems that a large factor in awarding vs not awarding a grant is simply name recognizability. My PI has literally been told before "yeah, those are problems with the grant, but he's such a big name we should just give him the money anyway".
---

Mal_Fet posted...

What do you mean by luck? Do you believe grants are given out buy coin toss? Or are they givien out based on how promising a researcher seems to the people giving out grants?

---

COVxy posted...

s0nicfan posted...

A single successful lab may be sucking up a lot of funding, but if they have significant momentum they can make major breakthroughts that smaller, "starting fresh" labs
couldn't.

This isn't what the data suggests. In fact, if anything, if you look closely, grant pooling creates stagnation of ideas, as a single lab usually only contributes to a single specialty with usually a very particular theoretical stance which biases the research.

From anecdotes of PIs, it seems that a large factor in awarding vs not awarding a grant is simply name recognizability. My PI has literally been told before "yeah, those are problems with the grant, but he's such a big name we should just give him the money anyway".

There's value in depth over breadth, sometimes. I agree that name recognition is a problem, though, especially if its tainting the selection process.
---

s0nicfan posted...

There's value in depth over breadth, sometimes

A person in my field has published pretty much the same flawed experiment over and over for the past 10 or so years, using millions of dollars of grant money to collect more data on the same flawed point. He's a household name, but there's no reason to continue funding him for the same project over and over.

People pounce on hot topics and run with the cycle, since it works. It's not depth if you do the same thing over and over.
---

COVxy posted...

Mal_Fet posted...

What do you mean by luck? Do you believe grants are given out buy coin toss? Or are they givien out based on how promising a researcher seems to the people giving out grants?

Ok so it's not luck. It is in fact based on how much promise the research group shows to the people giving out grants.

Thanks for your answer.
---

How much does politics/networking play a role a in research grants? By politics I mean both at a personal level as well as at a research level ie some topics of research are valued more above others simply because the leading academics of the field fostered a culture where it pushes research pertaining to certain topics over others, regardless of its academic veracity
---

Mal_Fet posted...

Ok so it's not luck. It is in fact based on how much promise the research group shows.

Thanks for your answer.

Reading doesn't seem to be your strong suit.
---

COVxy posted...

Mal_Fet posted...

Ok so it's not luck. It is in fact based on how much promise the research group shows.

Thanks for your answer.

Reading doesn't seem to be your strong suit.

Critical thinking doesn't seem to be yours. How does the study determine how much promise something has?
---

Kazi1212 posted...

How much does politics/networking play a role a in research grants? By politics I mean both at a personal level as well as at a research level ie some topics of research are valued more above others simply because the leading academics of the field fostered a culture where it pushes research pertaining to certain topics over others, regardless of its academic veracity

A fairly large amount. Name recognition is important. Science is often trendy. You'll find periods where something was hot so people jump on it.
---

Mal_Fet posted...

COVxy posted...

Mal_Fet posted...

Ok so it's not luck. It is in fact based on how much promise the research group shows.

Thanks for your answer.

Reading doesn't seem to be your strong suit.

Critical thinking doesn't seem to be yours. How does the study determine how much promise something has?

Again, by examining grants that got scores around the hard threshold, you are looking at what are effectively equally viable grants. For whatever objective measurement exists within grant evaluation, the measurement error is way too high to distinguish between scores so close, making these differences "as good as random", as the paper cited.
---

new and thrilling commentary

luck predicts success

no shit

>pnas

lul
---

COVxy posted...

Kazi1212 posted...

How much does politics/networking play a role a in research grants? By politics I mean both at a personal level as well as at a research level ie some topics of research are valued more above others simply because the leading academics of the field fostered a culture where it pushes research pertaining to certain topics over others, regardless of its academic veracity

A fairly large amount. Name recognition is important. Science is often trendy. You'll find periods where something was hot so people jump on it.

Interesting, thats what I figured, at the end of the day academics are just people with cognitive biases and incentive driven just like the rest of us. Modt human organizations generally trends towards the direction of herd mentality.
---

COVxy posted...

Again, by examining grants that got scores around the hard threshold, you are looking at what are effectively equally viable grants. For whatever objective measurement exists within grant evaluation, the measurement error is way too high to distinguish between scores so close, making these differences "as good as random", as the paper cited.

How do you determine objectively how promising studies are? Are the "equally-promising" studies on the same topic? Do they have the same intent? Do they have the same citations? How do you figure that one study is equally as promising as another unless the studies are carbon copies of one another?
---

Romes187 posted...

new and thrilling commentary

luck predicts success

no shit

Or rather, one instance of luck predicts recurrent success and end career paths.

You'd find many people wouldn't predict this. In fact, some vehemently deny results like this.
---

COVxy posted...

Romes187 posted...

new and thrilling commentary

luck predicts success

no shit

Or rather, one instance of luck predicts recurrent success and end career paths.

You'd find many people wouldn't predict this. In fact, some vehemently deny results like this.

Were you able to secure early grants in your career?

Romes187 posted...

Were you able to secure early grants in your career?

I'm not at the point of applying for grants yet. Still in PhD program.
---

COVxy posted...

Romes187 posted...

Were you able to secure early grants in your career?

I'm not at the point of applying for grants yet. Still in PhD program.

Knowing this information, what will be your strategy? If securing these grants is indeed correlated highly to future success, will you put extra effort into trying to get them?

Romes187 posted...

COVxy posted...

Romes187 posted...

Were you able to secure early grants in your career?

I'm not at the point of applying for grants yet. Still in PhD program.

Knowing this information, what will be your strategy? If securing these grants is indeed correlated highly to future success, will you put extra effort into trying to get them?

It seems initial effort wasn't the primary determinant here. It's hard, without a grant, it's hard to imagine you'd have the resources necessary to keep up productivity and write grants at the same frequency and quality. Might be a good idea to pursue a couple of smaller grant lines with co-PIs during your initial years of start up, so that if the larger grants fail, there is a chance you can still float a bit longer to write a second round on the larger grant for a better chance of success.
---

COVxy posted...

Romes187 posted...

COVxy posted...

Romes187 posted...

Were you able to secure early grants in your career?

I'm not at the point of applying for grants yet. Still in PhD program.

Knowing this information, what will be your strategy? If securing these grants is indeed correlated highly to future success, will you put extra effort into trying to get them?

It seems initial effort wasn't the primary determinant here. It's hard, without a grant, it's hard to imagine you'd have the resources necessary to keep up productivity and write grants at the same frequency and quality. Might be a good idea to pursue a couple of smaller grant lines with co-PIs during your initial years of start up, so that if the larger grants fail, there is a chance you can still float a bit longer to write a second round on the larger grant for a better chance of success.

imo just get a business degree and make phat $ instead of worrying about that

Romes187 posted...

COVxy posted...

Romes187 posted...

COVxy posted...

Romes187 posted...

Were you able to secure early grants in your career?

I'm not at the point of applying for grants yet. Still in PhD program.

Knowing this information, what will be your strategy? If securing these grants is indeed correlated highly to future success, will you put extra effort into trying to get them?

It seems initial effort wasn't the primary determinant here. It's hard, without a grant, it's hard to imagine you'd have the resources necessary to keep up productivity and write grants at the same frequency and quality. Might be a good idea to pursue a couple of smaller grant lines with co-PIs during your initial years of start up, so that if the larger grants fail, there is a chance you can still float a bit longer to write a second round on the larger grant for a better chance of success.

imo just get a business degree and make phat $ instead of worrying about that

I think this is what mathematicians call a degenerate case.
---

I wonder if the degree of difficulty to succeed in academia has anything to do with how they come off as isolated in their ivory towers with sticks up their asses. Or maybe such people are just naturally inclined to academia. Were you always condescending and a wiseass TC or did academia turn you into that?
---

Lol.
---

Up.
---

Im sorry, I assumed you knew, disregard my last post
---

Aaaand yep. the grants are the number one reason Im leaving academia.
---

The only reason to put up with academia is that you cannot imagine yourself doing anything else.
---