NSF/NIH (External Funding) Questions

[whybanana]

So I'm looking to apply for either the NSF or NIH grant for PhD studies (commencing in Fall 2019; yes I might be a little early in thinking about this). I was wondering how the grant works exactly. Do they fund the student directly (i.e. stipend) or the institution (to cover the cost of research/tuition)? Is this added to the original/base student stipend or is that decreased if the student receives the grant? What are the other benefits of receiving either grant (besides becoming a more attractive PhD candidate to grad schools)?

Thanks!

[jougami]

As an senior in undergrad (or as a grad student), you can apply for an NSF Graduate Research Fellowship, but the NIH fellowships are restricted to grad students as far as I know. For the NSF GRFP, the grant is paid to the institution as a $34,000 stipend and a $12,000 cost of education allowance for the student, yearly for 3 years. https://www.nsfgrfp.org/general_resources/about

Most of the places I interviewed at stated that if a student receives a fellowship, they get a few thousand added to their stipend as a bonus (so you don't get the entire amount of the fellowship stipend on top of your regular stipend but you're not going to lose anything if your regular stipend is more than the fellowship). In addition to having a nice line on your CV, having your own research funding allows you to be more flexible in the advisor you choose and the research you do since they have to worry less about supporting you financially and can spend that extra money on the research.

[Eigen]

Honestly, the prestige is worth it even with a small pay cut. 

[whybanana]

Yeah, smaller pay cut than I expected but I'm assuming the prestige pays off in the long run, career-wise. Thanks for the advice!

I've seen a good number of people on this forum receive these fellowships - how competitive are they exactly? Is it enough to have a well thought out and interesting proposal? I have a good pub track record and good connections - does that help at all?

[Eigen]

Pretty competitive, but a good track record and a solid proposal gives you a good chance. 

This forum is a poor representation of average- it tends to over-represent both extremes. 

[TakeruK]

Just to confirm...you are not talking about an actual net pay cut right? Are there really schools that would pay you less overall because you won a NSF GRFP? Even if they reduce your university-sourced funding, you will still take home the same or more money than if you had not won the award, right? 

As for the competitiveness question, I believe the funding ratio for these awards is a little over 10%, which is close to the typical funding ratio of most grants etc. They always award 2000 GRFPs and I believe they also name 2000 honorable mentions. A quick google search says that there were 16,500 ish applicants in 2015. The denominator would vary over time, and would probably trend upwards, so a ballpark estimate of 10%-15% funded over all NSF fields.

[Eigen]

I'm talking about actual pay cuts.

When I first got the NSF it was slightly less than my current funding source. 

And even now, since NSF doesn't change for COL, some of the RA/TAships at high COL schools are more than the NSF fellowship, but not all of them supplement the fellowship. For my school, it was an all or none thing- you took the fellowship or declined it, but there was no supplement.

This is especially true of schools with a high tuition. For RA/TAships, the tuition remission is often budgeted in with the funding from the School/University, while an NSF fellowship usually requires the department to come up with other funds to cover tuition. I think it was MIT that had the problem where they started having to limit the number of NSF fellowships they took as it was bankrupting the department- tuition was much higher than the COL payed to the school. 

[whybanana]

I mean, despite the prestige the GRFP comes with, that pay cut kinda sucks. I thought it would at least supplement my funding, either for research or stipend. I guess there's little to no financial benefit that comes with the grant. 

How are schools expected to come up with other funds? Why does the NSF require this? I thought NSF just pays the grant money to the university and then they decide how that's distributed over the student's tuition/stipend/research/department funding.

[TakeruK]

Wow, I didn't realise that there are some places that will fund you on their own system, or have you be on NSF/external funding only! My department views external fellowships as bonus only, so they are generally going to be okay with supplementing the funding since it would cost them less than if the student had no external funding. However, the department is responsible for all tuition costs of students, whether or not they work as TAs or RAs (and the department then charges the supervisor for tuition, I believe). 

8 hours ago, whybanana said:

How are schools expected to come up with other funds? Why does the NSF require this? I thought NSF just pays the grant money to the university and then they decide how that's distributed over the student's tuition/stipend/research/department funding.

I think this expectation is common to lots of national level fellowships, definitely the ones in Canada. The main reason (at least of the orgs. I'm familiar with) is that the funding org. wants to ensure that they are investing in a student that will be properly supported by their institution. The org. can fund more students with the same amount of money if they only contribute something like 50%-75% of the total costs. In addition, if the institution does not commit any funds at all towards an awardee, there is risk that institutions will "take advantage" of the funding org. and their awardees, treating them as "free students" that they don't have to worry about. And, since the funding org. knows their award won't cover all of the costs, they want to see a commitment from the school to cover the rest of it---they don't want the awardee to run out of funding and leave! I think the requirement that the school invest into the student financially means that they will value their student more too. Also, because tuition varies so much from school to school, especially public and private, if NSF sets a standard level of tuition support ($12k is pretty generous), then schools can't just charge higher tuition to get more money from NSF. Schools that want to have higher tuition are responsible for finding other ways to fund it.

[Eigen]

I think saying $12k is generous is... maybe a bit of an overstatement, especially given that it's supposed to cover all required fees (including insurance) in addition to tuition. 

There are very few universities in the US where graduate tuition is under $12k a year ($6k a semester), and quite a few that are significantly over. For comparison, my tuition remission for grad school was around $70k per year, and that doesn't count another $5-8k in fees and insurance. 

[TakeruK]

2 hours ago, Eigen said:

I think saying $12k is generous is... maybe a bit of an overstatement, especially given that it's supposed to cover all required fees (including insurance) in addition to tuition. 

There are very few universities in the US where graduate tuition is under $12k a year ($6k a semester), and quite a few that are significantly over. For comparison, my tuition remission for grad school was around $70k per year, and that doesn't count another $5-8k in fees and insurance. 

Oh, my understanding from my American friends was that the majority of public state universities have in-state tuition rates in the $10,000 to $15,000 per year range. My PhD school was not one of those, but the real cost of tuition is lower than the "on paper" cost. 

I also thought that the NSF tuition funding was not supposed to cover the entire cost and that the school was supposed to cover a good chunk as well as additional fees such as insurance and benefits. I would expect that NSF covers about half while the school should cover the other half, plus fees/benefits, so it is from this perspective (and the numbers above, which you are saying are wrong!) that I said "generous". Sorry if that's not correct though!

[ballwera]

I could see this maybe being an issue for year 1 or 2, but it still doesn't make much sense. Most programs have your funding guaranteed before you even matriculate. I'm not sure how you could be taking a pay cut for winning a prestigious fellowship...   

[Eigen]

45 minutes ago, ballwera said:

I could see this maybe being an issue for year 1 or 2, but it still doesn't make much sense. Most programs have your funding guaranteed before you even matriculate. I'm not sure how you could be taking a pay cut for winning a prestigious fellowship...   

Depends on how university funding is allocated. 

NSF requires that you *not* have any other direct funding source concurrent with the fellowship (i.e., no TA or RA work). Many universities don't just have "slush funds" that they can allocate a portion of to funding the surplus of a student getting a fellowship- that funding is tied to a specific job title and set of duties (TAing or RAing), and can't just be split off to be used at will. 

For schools that get lots of external fellowship winners (NIH, NSF, DOD, DOE) it makes sense that they have a system set up to give partial funding to fellowship winners. For schools where it's not normal, it's very unlikely that HR has a system set up to give funding (payment) to someone who isn't fulfilling a job requirement. 

Generally, when you get an external fellowship, you take that *instead* of a TA or RA position for the time of the fellowship. That's how NSF wants it to work- they are your funding source. 

Schools that "top off" funding have some mechanism in place to fund extra money from some other pool set aside for that purpose. 

[ballwera]

2 hours ago, Eigen said:

Depends on how university funding is allocated. 

NSF requires that you *not* have any other direct funding source concurrent with the fellowship (i.e., no TA or RA work). Many universities don't just have "slush funds" that they can allocate a portion of to funding the surplus of a student getting a fellowship- that funding is tied to a specific job title and set of duties (TAing or RAing), and can't just be split off to be used at will. 

For schools that get lots of external fellowship winners (NIH, NSF, DOD, DOE) it makes sense that they have a system set up to give partial funding to fellowship winners. For schools where it's not normal, it's very unlikely that HR has a system set up to give funding (payment) to someone who isn't fulfilling a job requirement. 

Generally, when you get an external fellowship, you take that *instead* of a TA or RA position for the time of the fellowship. That's how NSF wants it to work- they are your funding source. 

Schools that "top off" funding have some mechanism in place to fund extra money from some other pool set aside for that purpose. 

Just grabbed this from the NSF site, again no idea how exactly this would fit in. 

May I be paid (supplement my Stipend) as a teaching or research assistant on top of my Stipend?
Fellows are expected to devote full time to advanced scientific study or work during tenure. However, because it is generally accepted that teaching or similar activity constitutes a valuable part of the education and training of many graduate students, a Fellow may undertake a reasonable amount of such teaching or similar activity, without NSF approval at the affiliated institution. It is expected that furtherance of the Fellow's educational objectives and the gain of substantive teaching or other experience, not service to the institution as such, will govern such activities. Compensation for such activities is permitted based on the affiliated institution's policies and the general employment policies outlined in The Administrative Guide for Fellows and Coordinating Officials.

https://www.nsf.gov/pubs/2012/nsf12062/nsf12062.jsp#mib

 

[Eigen]

IIRC, that's an old guideline (2012), that has since become much more strict. 

It used to be (when I first got my fellowship) that it was allowed, and up to the advisor/CO. It changed over the course of my fellowship to no longer allow TA/RA compensation as a general rule, with some rare exceptions granted.

[ballwera]

2 minutes ago, Eigen said:

IIRC, that's an old guideline (2012), that has since become much more strict. 

It used to be (when I first got my fellowship) that it was allowed, and up to the advisor/CO. It changed over the course of my fellowship to no longer allow TA/RA compensation as a general rule, with some rare exceptions granted.

Gotcha, just out of curiosity, what field are you in?

[Eigen]

Biochemistry.

Was editing in more details as you posted, so I'll include them here:

Generally, the 2016 language is similar, but the emphasis is on the fact that teaching assistantships *must* be to the benefit of the students education and development, and not in service to the institution. 

For instance- if teaching is required as part of a pedagogy class, an opportunity to teach a class you wouldn't otherwise get to teach, etc. 

But if you're performing TA duties that the department needs, that's considered a no-no. It's mostly left up to the CO of each institution, but some are more experienced than others.

[ballwera]

2 minutes ago, Eigen said:

Biochemistry.

Was editing in more details as you posted, so I'll include them here:

Generally, the 2016 language is similar, but the emphasis is on the fact that teaching assistantships *must* be to the benefit of the students education and development, and not in service to the institution. 

For instance- if teaching is required as part of a pedagogy class, an opportunity to teach a class you wouldn't otherwise get to teach, etc. 

But if you're performing TA duties that the department needs, that's considered a no-no. It's mostly left up to the CO of each institution, but some are more experienced than others.

are you in a medical school setting? For my program at least ,we are paid directly with dept. funds (some of which comes from a T32) for the 1st 2 years and the following 3 years are funded directly by the PI, usually in the form of a grant. So in this case those in our program would be fine, actually our handbook states that we get x% of any money that we bring in through a fellowship or F31. 

[Eigen]

I'm not, but I'm also no longer a grad student. I'm faculty now. I was not at in a med school setting when I was in grad school, however.

And yes, there are plenty of cases where there's funding that's fine in addition to the fellowship. That said, you may run into problems with NIH and NSF funding at the same time, but I think a T32 training grant should be fine since it's to the institution and not the individual. 

That said, the department may not want to apply funds from other sources to you if you have a fellowship that will support you- that's funding that can be used to bring in another student.

[aberrant]

On 7/9/2017 at 4:40 AM, whybanana said:

So I'm looking to apply for either the NSF or NIH grant for PhD studies (commencing in Fall 2019; yes I might be a little early in thinking about this). I was wondering how the grant works exactly. Do they fund the student directly (i.e. stipend) or the institution (to cover the cost of research/tuition)? Is this added to the original/base student stipend or is that decreased if the student receives the grant? What are the other benefits of receiving either grant (besides becoming a more attractive PhD candidate to grad schools)?

Thanks!

Don't think people have discussed much about NIH extensively, so just to fill in some blanks, if any.

You are probably looking at NIH fellowship(s) as opposed to a grant per se. Typically grants are for individual PI or an institution/facility/consortium, even if that is a training grant where students can be directly benefit from (i.e. finacial aspect).

For a NIH fellowship, you basically need sufficient preliminary data to be even worth considering, besides the significance and originality of your proposed research. Hence, it isn't common for someone to apply and received a NIH fellowship before starting graduate school (nor would I expect someone can get it in their first year of grad school). A fellowship does not have indirect cost. They fund the student directly, but, to my understanding, the money still went through the office of research and distribute to you, per your school policy.

Common practice, whether that be NSF or NIH, to my knowledge, is that a stipend from fellowship does not add to your original stipend offered by your program from your department at your school. In the event where your fellowship stipend is less than what your school is offering (i was in one of those situation), your school/PI would make up the difference, typically without requiring you to become a TA because most, if not all, of these fellowships require the awardee to devote 100% of their time and effort on the proposed research. In short, you will always received at least what your program's offer letter wrote, as your minimum stipend (if not more).

That being said, you are unlikely to receive anything from NIH before you start grad school. Benefits of having a fellowship, however, is pretty obvious if you plan to have a career in academia.

[Cervello]

If you win a NSF grant, does it make sense to defer the grant for a year or two so that you'll have more flexibility in choosing a lab head who may not otherwise be able to fund you? Or do most ppl have the grant immediately start when they enter grad school, which would benefit the program over an individual PI? Pros/cons to either decision?

[Eigen]

I got mine at the end of my first year, and deferred one year, and wish I'd deferred two. 

Using it later gives you more funding when you need it- use available funding your first year if you have it. 

First year is the best time to TA if you need to as well, at least in terms of research productivity. 

[Cervello]

@Eigen Thank you, belatedly, for your response. Much appreciated.