views:

1173

answers:

8

I'm currently in my second year of a Computer Science degree and all is going well. I finished the first year with flying colours, even whilst spending a couple of months out for physiotherapy on a previous injury, and now the second year's content seems relatively straightforward. If I can keep up this rate of achievement then I'm sure to place first or second in my class (out of 15-20).

Whilst it is still early to be looking around I'm interested in viewing my options for post-graduate education. At the moment a Ph.D appears to be the ideal path for me to go down, as I'd love to go into research and perhaps the option of teaching. However, in my country all the best post-graduate degrees are taught at the top universities and my current university is ranked very poorly in nationwide league tables.

Much of what I know I have taught myself, using my university and the facilities to my advantage. I regularly program in Haskell and Lisp and read books on Linear Algebra and Calculus, even though they're not part of my course. I have relevant work experience in several fields of IT and Business, ranging from Search Engine Optimisation Consultancy for a large real-estate website and working as an Administrator for my universities SU website, not to mention some personal development projects.

The best universities in this country ask for students with a first in their chosen degree, but how much will my previous university stand-out on my application? Does an application from a student that was top in his/her class from a poor university mean less than a student that was 20th from a top university?

In essence I'm still studying Computer Science. I've been taught things like Formal Language, Automata, Computer Systems Architecture, Procedural, OO and Logic Programming and more, but in less detail and not in as much Math as other programmes.

As of now I am looking at opportunities to study for a Ph.D, although a Masters in Computer Science would be fantastic as well.

A: 

This depends on the country and on the school.

For example, in the US certain schools are known to be very "Elitist" so they ask you where you went to but also where your parents went to (in case they went to the school you are applying to, which is a benefit), etc. Other schools (even top-tier) are more relaxed about it. I have known people in top-10 programs who came from unknown universities so not all is lost.

When a school makes the decisions, they are primarily looking for research potential, interest match, and recommendation. In many ways it is safer to pick someone who's already been "screened" by nature of having gone to a top school for undergrad.

If there is someone who publishes at your school (use DBLP or ACM to find out), get close to them and get a recommendation and hopefully do some research with them.

Find a way to do research and publish - good conferences are peer reviewed.

Demonstrate that someone has paid for your research - if you are in Europe and want to go to the US, spending the time on a research masters in Europe may give them some assurance that you know what you want to do with research.

The vast majority of schools care very little about your programming skills or what you've read. There are also a lot of top people in a lot of low-ranked schools (since there are so many unranked schools) so you need to compete well and have something that makes you remarkable.

Uri
Well, I am from the UK and many of the universities here are well-known for their elitism. As of right now I am unsure whether I'd like to study abroad, although a good Masters degree here could set me up nicely if I were to study a Ph.D in America.
EnderMB
There's a lot of good British research from unfamiliar universities. When you're doing a PhD, what really matters is your topic and publications. Find someone whose research you like and go to work with him even if he works for a university in the middle of nowhere.
Uri
I've had a look around some places and the majority of the lecturers and professors there have told me to my face that I would be better off applying somewhere with a good reputation, just on the basis of funding and facilities. I won't name names but I admire his honesty.
EnderMB
A: 

I was privileged to attend an Ivy league school in the US, and I can attest that it was a good form of meritocracy: you were assessed on your abilities and intellect, not your connections or who you know.

If you plan on studying in the US for grad school, you really need to excel on your GREs. For the top 10 schools, it will probably give your application a chance of consideration by the admissions committee but it won't guarantee admission.

One other thing which will pay enormously throughout your software development career: work on your writing skills. You will probably be assessed by your ability to write a research paper explaining your technical work. Since excellent writers are rare and developers who write and communicate well are even rarer still, this will go a long way toward helping others perceive you as intelligent.

Alan
My impression is that the GREs don't matter much, possibly because such a significant percentage of people max out (e.g., 2-3%). GREs may be a requirement but they don't open the door for you.
Uri
I disagre as well. The GRE is not that heavily weighted in CS programs. All the GRE can do is hurt you, it won't help you. That is a good score won't do a whole lot for you, a bad score could hurt you a bit. I think your research statement and grades are the most important factors.
BobbyShaftoe
+5  A: 

Having a degree from a UK university that is lower in the league-tables should not put you at a disadvantage at post-graduate level. A bachelors degree shows you are able to learn to a given standard over a wide range of subjects and work independently (for Honours degree). You don't normally go through UCAS (at least not in 2001!), instead you make an application straight to the research group you wish to join. You should be looking to get a minimum of 2:1 to get funding.

UK university league-tables are primarily there for undergraduates. You want a decent research-based university in your topic area - this may not include the "top universities" you mentioned. Background research is key here.

If you want some suggestions on good research-based universities, give me a shout.

Here are a few tips I'd recommend anyone with a [death]wish to do a PhD. Some of this applies to any university, not just UK.

Reseach Topic

1) Think about what you want to research - remember, it needs to be interesting and original enough to produce results that contribute to Human knowledge in your subject area.

2) Find a research group at a research-based university and look at their areas of interest. Look at their publication record in reputable journals and conferences.

3) Write a research proposal (1 page) that targets research group interests.

4) Email research group contacts to see if they are interested in your topic

Funding:

UK Universities have the benefit (for UK and EU residents) of providing virtually full funding for full-time research-based post-graduate studies.

In the UK, the research councils run the Research Assessment Exercise (RAE), basically a ranking of UK research-based universities. This ranking is important, since it determines what funding a university gets from research councils, and therefore how much funding you can expect to get, e.g. EPSRC funding.

The next RAE results will apparently be released mid-December 2008. 2001 results are here.

Doing Research

Be motivated, because you won't get the lovely new title until you have finished your thesis, done the viva, completed corrections, filled in the forms, and printed so many copies of your thesis you are sick to death!

Career Prospects

There are plenty of career prospects, although from my own experience, only a small percentage of post-grads take up a career at university beyond being a research fellow - just not enough money there ;-)

HTH

Rowland

Thanks for the fantastic response! I've got a couple of good research universities in mind and I feel that post-graduate education is where my future lies. I do have a question about funding. How easy is it to get funding and what will funding cover, just the PhD or living costs as well?
EnderMB
A: 

Admissions at top graduate schools are about research, research and research. You need to convince them that you're capable of doing independent research. Your letters of recommendation count the most -- how positive they are and (unfortunately?) who's writing them.

Things like class rank, grades, gres etc count for little. Show capacity for genuine, creative research, and that's the way to go.

Ying Xiao
Certain schools such as MIT require your professor (who's giving a recommendation) to list if you're the best student that he's seen in the last X years and other BS like that. Class ranking matters to get you considered. Research gets you admitted.
Uri
Definitely. However,t his is not just true of the Ivy League. Honestly, in the U.S., the CS rankings aren't that big of a deal. All right, sure an MIT or Carnegie Mellon will jump off the resume, however there are big papers being churned out at state schools as well.
BobbyShaftoe
A: 

As for applying to an American university, I know that many of them offer limited admissions/financial aid to international (as opposed to domestic) students-- a disadvantage, though they would be less likely to compare you with others from your country.

Dylan V
+1  A: 

From a UK perspective: I would definitely recommend doing an MSc in Advanced/Research Computer Science before commencing a Ph.D. (I did this myself).

This type of Masters is a much higher level than undergraduate, will get you studying/involved in the latest research in the CS community, will help you move from a lower standard institute to a higher standard one, and would also help you focus on (i.e. find) a research area for your Doctorate.

graveca
+2  A: 

I realize your question has already been answered but I just wanted to point you to the best article I've read on getting into and surviving a Computer Science Ph.D:

So long, and thanks for the Ph.D!

"Everything I wanted to know about C.S. graduate school at the beginning but didn't learn until later."

Simucal
+1  A: 

I'm in agreement with your chosen answer, and also the comment about doing an MSc--generally you pay for the privelage of doing one, and when money's involved that usually means better universities aren't so picky. You can then use that as a stepping stone.

My one piece of advice would be to pick a 3rd year project which is research focused, and not merely a software engineering project. That will make it far easier to make a decision about likely research topics, you'll be more conversant with academic jargon, and it will look better on your CV.

As for funding, I'd say the vast majority of home students are on EPSRC funding. Generally exceptionally low pay--£12k ($20k ish) plus fees but it pays the bills.

Andy