wahl

The NSF can be deferred for two years, but the NDSEG cannot be deferred to my knowledge. If you win both in the same year, you would take the NDSEG for the first three years, then switch to the NSF for two more years. (You lose a year of the NSF because you cannot defer it for three whole years.)

Someone posted the list of honorable mention recipients for 2008 on another board. My name is on there - it's probably legit!

Yes, these people.

Congratulations! I guess this is bad news for me. When I log into fastlane, I see nothing.

Nope, but if you call and the dude tells you, please let me know :)

Jeeves, I called today and someone at NSF picked up: they expect to notify us next week. Sorry!

I know someone who got into Berkeley after he had already been rejected, probably because of his NSF. (Goes to show that a lot of great applicants slip through the cracks of the admissions process!) Another winner I know didn't manage to have his rejection from MIT reversed. If you win but can't get in anywhere, reapply next year and be sure to mention that you have already won the NSF.

hanurii, don't worry so fast :) In previous years the number of awards has correlated very closely with the number of applications (usually about 1/9 of the total applications). Over the three years I have seen the numbers for, the percent of applicants receiving the award has been nearly constant: 2004 - 11.4% 2005 - 11.2% 2006 - 11.1% Maybe fewer people applied this year.

fuwing, that could be reasonable as long as you find the right housing. The rule of thumb (and one used by landlords before they let you rent) is that you should spend about 30% or less of what you make each month on rent. Assuming you make only $19k for the whole year, that's $475 per month, which means you would need to split a small apartment with a roommate or have subsidized university housing. It could be done in Boston...try the website boston.craigslist.org to find roommates or apt listings. (Since you're from Hong Kong you might not know...craigslist is a free site that lists apartments, jobs, cheap stuff for sale, personal ads, etc. for major cities in the US and abroad.) My suggestion is to find extra work for the summer, maybe an internship. Also, in Boston there are lots of undergraduates looking for tutors and many of them are wealthy - you could get paid $50/hr to help them study if you'll be going to a prestigious school. That's something you could do nights and weekends during the school year to boost your income.

MexEcon, I have a similar situation with a fellowship. In my case the fellowship would provide a bonus on top of the normal stipend; instead of divvying up that extra cash across the biweekly stipend checks, the fellowship money would arrive separately from (and less regularly than) the stipend money. Maybe that's your situation?

Take college courses in the evenings through your local community college (or a better local university, if you can afford it). Take the full math sequence that they offer - linear algebra, multivariable calculus, differential equations, etc. - in the evenings/weekends, during the summers if you're in town, or even during the school day if you can swing it. Learn your way around a machine shop (community colleges offer many classes) and demonstrate your skill by entering design or robotics competitions. Those things will matter more for MIT and your MechE career than tennis or volunteering in a library (although ideally you would do it all!).

fp, you are probably looking for the webpage for the NSF Graduate Research Fellowship Program. You'll need to read the Program Announcement (on the left), esp. Section IV. This seems to be the relevant material on citizenship: Applicants must be United States citizens or nationals, or permanent resident aliens of the United States. (The term "national" designates a native resident of a commonwealth or territory of the United States, such as American Samoa, Guam, Puerto Rico, U.S. Virgin Islands, or the Northern Mariana Islands. It does not refer to a citizen of another country who has applied for U.S. citizenship.) Only applicants to PhD programs (not master's programs) are eligible, and all are required to write a research proposal as well as two other essays. Details are in the Program Announcement. The deadlines have already passed for this year, though.

Rachitbly, Sometimes ETS messes up. It is now late enough that you could reasonably have expected them to arrive by mail, no matter where on Earth you live. I would call their customer service number, tell them that you still have not received your report, and then ask them to resend you a copy. I'm sure they will gladly oblige. Maybe you could get them to tell you your scores over the phone as well (but make sure you get a paper score report at any rate).

Institution Codes Department Codes

I would go with the professor. The reason is that you want to maximize the chances that someone on the admissions committee knows your recommender personally. The research fellow is still young, probably hasn't traveled enough or earned a great reputation like the professor has. Also, there is unfortunately a stigma associated with research fellows. (It might not apply at your university or to this person, but adcomms might assume that it does.) Typically, Ph.D.s start as post-docs, and only become research fellows if they don't find a job (or at least another post-doc position) in a reasonable amount of time. It's like a red flag that says he/she didn't get a tenure-track position fast enough, which is bad news for their reputation when recommending you. I assume that you worked directly under this research fellow. One option is to arrange for the postdoc to write the body of the letter and the PI to sign/send it. That way, you get the glowing rec and the professor recommending you. I would advise against having the research fellow *and* the professor of the same lab both send separate recommendations: it makes it look like you couldn't find anyone else.