How does one go about choosing a graduate school for computer science?

Question

I have a year left until I finish my BS in computer science. I want to go to graduate school, so I am in the midst of trying to decide where I want to go.

I realize, of course, that in the end my decision will be dictated by where I get accepted, and what I can afford.

That said, I would like to apply to places which are relevant to my interests - something that I recognize is continually changing.

So my question is: how do I decide? How do I narrow things down? Has anybody had experiences working with faculty or students (graduate, PhD) from different universities? If so, do you have any insights to lend? ("x has excellent facilities and funding to do research in HCI" or "y is a small school, and their faculty are excellent.")

I myself am interested in computer networking (and I'm currently doing research with a professor in that field.) Of the classes I've taken, I tend to be the most interested in systems-related concepts (rather than HCI or Software Engineering.)

Nearly all schools have research faculty in "computer networking" - but I have no way to tell which projects are the most worth working on, or which program will give me the most depth. I can call the schools and visit them, but what kinds of questions should I be asking to figure out if a given school is a good fit?

[community wiki. this is meant to be in the same spirit as many of the interesting undergrad-experience related questions, only, hopefully, with advice specific for the unique things that grad students worry about.]

Edit: This is a similar discussion, though I think my question is a bit more generic.

Answer 1

Stanford, MIT, CMU, Berkeley ... others.

Answer 2

Here's what you do with grad school:

1. Get into the best grad school you can (where "can" includes "afford" - financial aid is a huge bonus).
2. Get up to doctoral candidacy; take any necessary classes, prepare for any candidacy exams.
3. Find a faculty member with tenure to be your thesis advisor.
4. Pass the candidacy exams
5. Do a dissertation on a topic that your advisor likes and will get behind.

Once you graduate with the union card, you will have a lot of freedom to do whatever you like; you don't have to stick with your dissertation topic.

Answer 3

Are you going just to get a pay raise and do you plan on working in the business world? If so, any local university (not community college) will do. I hate to tell you, but you might end up in the business world doing CRUD whether you like it or not. Also, it's really better to get 2-3 years of experience before getting a graduate degree.

If you are eventually heading for a PHD or a research job, then you want to look at better schools as others have suggested.

Answer 4

I'm in the same boat, i graduate with a BS in Computer Science next year and want to go to grad school most likely Computer Science or Software Engineering.

Rememeber that the name of the school isn't always everything, you have to go somewhere you'll be happy. If you aren't happy, your not going to do as well. You shouldn't go to somewhere like Carnegie Mellon University just because of the name and find out you hate Pittsburg or something like that as an example. If you can, try visiting schools first.

It might also depend on how near or far you want to be from home, east coast, west coast, north, south, if you don't like hot, don't go to a school in the south for example.

now as far as learning and whatnot, you want to try to google schools and see what kind of research they are doing currently for some ideas. Ask some of your current faculty what they might recommend. One of my professors knows one of the professors at Univeristy of Florida, so I have a networking connection right there that could help me out.

Where ever you choose to go, your success will depend mostly on you and not just your facutly advisor, it will also depend on your interest in the topic you are pursuing so try to figure out what you really want.

For some examples, i'm looking at University of Florida, University of Texas @ Austin, Virginia Tech, NC State, UNC, and Georgia Tech, for my "I probably won't get accepted anyway" school i'm looking at UCBerkeley

I'm sorry i don't have a whole lot of info, but that's because i'm in the same boat. This just seemed to work for me so far and was some advice i was given so, i figured i'd pass it on to you. So good luck! (I'm really dreading the GRE's right now... ugh, not looking forward to taking it haha)

again, it is YOUR life, YOUR decision don't let other people tell you that you have to go to a certain school when you feel like you probably won't enjoy it there. Do what will make you happy (I can't seem to stress that enough)

=)

Answer 5

If you can narrow the choices based on things like what the school(s) reputations are, then to the extent possible, visit the departments. Try to talk personally with the profs that you think might have projects you are interested in. Two reasons for this are (1) you very probably will end up assuming a pet project that a prof already has in mind, and (2) you are going to spend a lot of time with, and be very dependent on the good will of your chosen prof.

The outcome of your grad school experience will depend strongly on how you and your chosen advisor relate to one another, so really try to get a sense of the person before you get committed. First impressions count. And if you can't imagine that the project is the 1st thing you want to get to on Sunday mornings, you will want to ask yourself what kind of project would motivate you in that way.

Answer 6

Look at the major labs/projects that universities have in their department. As a graduate student you won't just be in "computer science"; you will be working under a specific advisor or team that has their own research interests, so it's best to find a school that has an active project in a field you are particularly interested in, and a team that you can see yourself fitting into.

For example, if you are interested in HCI, then CMU has the HCII and a variety of projects. Don't just ask which overall university is well rated; ask which of these researchers you'd feel good doing your dissertation under.

Answer 7

some good advice for your question in comments here.

Answer 8

You don't say in your question whether you are considering a PhD or MS, so I'll assume a MS. I'll also assume you're a US student; things are different in other countries.

The university I got my MS from had two options for obtaining the degree: research or exam. For the research option, you take some classes and write a dissertation. For the exam option you take more classes and take comprehensive exams. Both result in the awarding of a Master's degree.

Although I'm still in academia now, I'd have to say that the classes I took in grad school were way more useful in the long run than the research. I'm happy to say that the subject of my dissertation is now obsolete -- it shows that technology has progressed -- but every day I still use the things I learned in those classes.

The place you choose for your graduate school is going to depend on what you want to do with that degree. If you want a good teaching job at a university, it'll be better to have a degree from a good university. But if you want a graduate degree for the doors it will open for you in industry, I think it'll be more important to get a better foundation in the advanced CS topics.