NSF GRFP 2011-2012

[TXTiger2012]

Y'all are either very bored or very crazy. And please, I don't mean that in a derisive way at all! More of a friendly, teasing, "My god, I wish I had the time to graph that all out" kind of way

[delta16]

Well we now know it might be early April based on Facebook....

http://www.facebook.com/NSFGRFP

[illini2015]

woweeeeeeeeeee can't believe the NSF is using facebook to announce something so important. It's not even on fastlane. Thanks for the heads up!

[guttata]

It's not on Fastlane but its the same generic statement as they have on nsfgrfp.org. Newsflash: Late March is "around Early April." No new information.

[deepbreath]

My vote is on the morning of Tuesday, April 3. Previous years indicate a preference for the first Tuesday in April. I need to stop over-analyzing this stuff....

[rockhopper]

woweeeeeeeeeee can't believe the NSF is using facebook to announce something so important. It's not even on fastlane. Thanks for the heads up!

I agree it seems strange that they would use Facebook to announce dates. Well of course now I've got to go and Like this page...

[svh]

I like the time-series analysis up there. Very nice.

Edited March 23, 2012 by svh

[maath805]

Here are some extra data points!

2008 - March 31

2007 - March 26

2006 - March 30

2005 - April 9

[Jimbo2]

Sat, Thu, Mon, Mon, Fri, Tue, Tue

[svh]

I'd like to see a plot of your residuals.

[anthro.fish]

More Fellowships this year?

From the NSF Fiscal Year 2011 Budget request to Congress:

"The Graduate Research Fellowship (GRF) program, (16 percent increase to $158.2 million); an

Administration priority, supports the development of the Nation’s future scientists and engineers. FY

2009 marked the beginning of a growth trajectory to triple the number of new awards made each year

to 3,000 by FY 2013."

http://www.nsf.gov/a...view_fy2011.pdf

(Overview Page 3)

Edited March 24, 2012 by Anthroman

[Calvin_xc1]

More Fellowships this year?

From the NSF Fiscal Year 2011 Budget request to Congress:

"The Graduate Research Fellowship (GRF) program, (16 percent increase to $158.2 million); an

Administration priority, supports the development of the Nation’s future scientists and engineers. FY

2009 marked the beginning of a growth trajectory to triple the number of new awards made each year

to 3,000 by FY 2013."

http://www.nsf.gov/a...view_fy2011.pdf

(Overview Page 3)

"

- $10,500 cost-of-education allowance to the institution (anticipated to increase to $12,000 for 2012)"

Most likely means more funding as well as more offers. If you think about it, since funding is for 3 years, and there's ~2,000 applicants who receive it, an increase of $1,500 per applicant per year would cost an extra $9,000,000.

Edited March 24, 2012 by Calvin_xc1

[maath805]

The graph looks curvilinear more than anything. I don't think a general linear regression is quite appropriate :-/

[syn2012]

I can't believe that we may only have a week left to wait. Incredible.

[maath805]

^^

Here's hoping it's just a week!

[ejmurphy]

From a fluid mechanician's point of view, your graph looks like the result of a statistically stationary random process, the mean is sufficient.

[Agradatudent]

I had a dream that NSF released the winners... but I woke up to find that is not true.

Edited March 25, 2012 by toypajme

[illini2015]

I had a dream that NSF released the winners... but I woke up to find that is not true.

I also had a dream that the results were released Saturday night. It was actually a nightmare. Read the reviewer comments and got another HM

[uksawfly]

I haven't dreamed about the results yet - I've been buried under a pile of other small grant applications - but I'm sure it's not long in coming. I also just found out some of my family is visiting during the timespan NSF is likely to announce the GRFPs, so on the bright side they can take me out to dinner to A.) drown my sorrows again (more likely) or B. ) celebrate.

Edited March 26, 2012 by uksawfly

[koopahtroopah]

So, even after obsessively reading endless pages in this year's forum (and from previous years)...I'm still a bit confused about certain "factors" that might contribute to your chances of getting an award. I mean, besides the obvious factors including the strength of your proposal, how you address broader impacts, your academic background, PUBLICATIONS, and letters of rec, there are also other determinants including sex, race, the state you received your high school diploma, and whether you wear boxers or briefs...can anyone confirm?

Edited March 26, 2012 by koopahtroopah

[Dynamom]

I also had a dream that the results were released Saturday night. It was actually a nightmare. Read the reviewer comments and got another HM

This wait is seriously crazy-making. Sigh! I keep telling myself I'll be ok as long as I at least get an HM since I have one more year to apply... oh, and that as long as the reviewer comments are constructive, I can even take rejection... any other "it'll be okay" mantras out there?

[Jimbo2]

my "it'll be ok" mantra is a bit two sided... i think my proposed research essay sucks now, which means the one i would write next year should be much better, but then i'm really not sure how this year's application will go. an HM would be ok, but to be honest the $100K and funding to upgrade from a M.S. to a Ph.D. program that's the difference between the award and the HM would be huge for me.

[deepbreath]

I tell myself it'll be okay if I don't get it because I can reapply next year, but I don't think my attempts to calm myself are working. I'm still thinking about the results more than I should be. This morning I even checked fastlane for messages about being down before I checked my email. x_x

Edited March 26, 2012 by deepbreath

[tlevine]

It appears that results have been released on the first Tuesday of the month for the past few years, so I predict that the results will be released on April 3. They also sometimes come out on Fridays, so my second guess is April 6, which is the Friday of the same week.

Also, here's a frequentist estimate that ignores the day of the week. Among other invalid assumptions, we assume that the last six years are a random sample of years,

#!/usr/bin/Rscript

release_dates_text <- c('Apr 17', 'Mar 23','Apr 1', 'Apr 10', 'Apr 6', 'Apr 5')

release_dates <- as.POSIXct(release_dates_text, format = '%b %d')

conf.int <- as.POSIXct(t.test(as.double(release_dates))$conf.int[1:2], origin = '1970-01-01')

print(mean(release_dates))

print(conf.int)

and we get a 95% confidence interval of the population mean of March 27 to April 14 (much wider than the credible interval that qlathrop estimated). And if we had to guess based on this analysis, we'd say April 5.

But again, this estimate is stupid because it doesn't consider domain knowledge; considering day of the week, I predict that the results will be released on April 3 or maybe April 6.

Edited March 26, 2012 by tlevine

[Jimbo2]

haha very nice. i'm hoping for a leak like there was a couple years ago, but the folks at NSF have probably fixed whatever the glitch was.