Page 1 of 3

Long-term advice

PostPosted: Tue Jan 25, 2005 9:22 am
by Tony
I may be getting ahead of myself, but in a society where they try to pin us down on a career in elementary school, I feel as though I'm on solid ground in asking.
I am an undergraduate student in my first year at university, but technically a Sophomore in standing due to my high school record. However, professors still view me as a Freshie, and I can't seem to find time with the people that could really answer my questions, so I decided to ask the global community at large.
I am working towards my B.S. in Computer Engineering, but ultimately intend to earn my PhD. I have yet to decide on a specialization but am leaning towards neural networks.
To begin with, I want to know what grad school is really like. I've heard that getting in and doing well depends highly on the impression you leave on the right professors as an undergrad. Is there any truth to that?
Once I have my degree, I have questions about where to go from there. What I really enjoy is learning new things. My ideal would be to stay in college forever, taking more and more classes. Is that at all feasible? What would you recommend for someone so bent on learning new things?

Long-term advice

PostPosted: Tue Jan 25, 2005 2:54 pm
by anonymous
trust me, even you will eventually need to take a break from taking classes.

college is great, but eventually all the tests and homework get boring.

grad school is not like taking classes. it involves a lot of self-direction and creativity, and most of it is problem-solving.

many people who are good at 'school' find that grad school is a different animal altogether, and not at all fun for them. if you're the sort of person who loves grades, you will hate grad school because you don't get anywhere near the same amount of feedback. no one can tell you if your answers are the 'right' or the 'best' answers. most of the time they are the only answers, and very few people in the world are even qualified to evaluate them.

grad school is a very different kind of learning. if you really love learning, you will like it despite all the drawbacks. just remember: not all learning takes place in a classroom.

Long-term advice

PostPosted: Tue Jan 25, 2005 2:57 pm
by Bill L.
Hi Tony,

There are two good books that discuss the graduate school experience (reasons to attend, life during each year, mentorship, dissertation, staying forever or getting out, etc.) are "Getting what you came for: The smart student's guide to earning a master's or PhD" by Robert L. Peters or "The PhD process: a student;s guide to graduate school in the sciences" by Dale F. Bloom, Jonathan D. Karp and Nicholas Cohen.

They can probably answer most of your questions.

Be well,

Bill L. & Naledi S.

Long-term advice

PostPosted: Tue Jan 25, 2005 10:45 pm
by John Fetzer
Getting into grad school depends on where you are. The US and Canada are quite different than Surope. In the US, grades mean a little, but your GRE scores are an order of magnitude more important, especially the specialized topical one. I know a 3.25 for a bachelors program (albeit one known for its difficulty) would not have meant much, but GREs in the upper 90 percentiles - and a 99+ in my degree major - had grad schools everywhere chasing after me and offering higher stipends and fellowships. The second major factor is timing. Take your GRES as early as you can and apply early. Before Thanksgiving of your senior year. Grad programs have a limited number of slots and save some plum assistantships just in case, but are freer early on.


Long-term advice

PostPosted: Wed Jan 26, 2005 2:55 am
by Kim
I would disagree. From my experiences, GPA is much more important than GRE. And I have heard the same things from many professors. And in most cases, a person's GPA does agree with his/her GRE performance. Only when GPA and GRE do not agree with each other, a red flag would be raised.

GRE is a one time performance. GPA is the indicator for the achievement during the four years in college. For most professors, GPA is more accurate than GRE.

A low GPA and high GRE may suggest that that person is an underachiever. It may indicates that that person may be very intelligent but somehow neglects his school works, for example, failing to turn in papers. So that person got lower GPA.

A high GPA and low GRE may suggest grade inflation, on the other hand.

Long-term advice

PostPosted: Sat Jan 29, 2005 10:11 am
by Doug
No, Kim and John are BOTH wrong. (Just kidding.) Really, you could easily find examples of every possible combination of GPA, GRE, letters, etc. This thread highlights the act that admission to grad school is extraordinarily variable. At some places, GRE is important, at others, it's grades. My impression is that for most places, how prepared you are to think independently and come up with an interesting project, as attested to by 1) prior research (at the undergrad or MS level) and 2) rapport with potential graduate advisors (who you should be able to converse with, at least via e-mail, even at an early undergrad stage). Good letters of recommendation can be VERY important for both 1) and 2).
Sorry for being long-winded. It's hard to pigeonhole the grad school entrance experience, though.

Long-term advice

PostPosted: Sat Jan 29, 2005 10:41 am
by Mandy Tamer
Another book you can look up on the web is "The PhD Trap" check out this academic site on graduate schools with lots of link. A huge amount of info on web. Try norman matloff as a google search.

Long-term advice

PostPosted: Sat Jan 29, 2005 1:43 pm
by John Fetzer
I was speaking from my own experience. I had a 3.25 or so in a very tough degree program, one that required 50 semester hours in your major - meaning you ended up taking several graduate-level courses. Additionally there was lots of other math and sciences required. Almost everyone getting a BS later got a PhD.

My GREs, especially the special topics on attracted attention as it was in the top percentile. If you get a 99 on your biology, chemistry, biochemistry, or whatever major, you will get attention.

The other point was not waiting. Too many students apply around Christmas, when grad schools have started being choosier. Many very good students do not get accepted or get less assistantship money only because they applied in December or January. Equivalent students applying in October or November get better offers.


Long-term advice

PostPosted: Sat Jan 29, 2005 6:36 pm
by John G. Hoey, Ph.D.
I think Doug gives the best advice....the grad school experience is difficult to generalize. Regardless of what it takes to "get in", you need to also think about what it takes to "get out", i.e., the "exit starategy". I wish I had a nickel for the number of grad students who had everything needed to get into very good grad programs but floundered for years, ultimately either leaving without getting a degree or hanging around so long it became detrimental to their career. More than anything, the one asset you'll need to successfully negotiate your way through the graduate education is "sticktoitness". You will be very much independent----even if you do have a great mentor/advisor (as I did)----and will need to have/develop the discipline to stay the course. Many people who have mastered rote memorization and test-taking needed for succeeding as an undergrad end up on the short end of the stick when it comes to the skills needed to make it through grad. school.

John G. Hoey, Ph.D.

Experience Concerning Graduate School/ PhD

PostPosted: Sun Jan 30, 2005 4:33 am
by Harald
As an undergraduate you learn the basics of your field such as physics and maths. From my experience, I do not see any advantage in trying to skip a year or to take a test without taking a class. In fact, university is a great deal more difficult than high school. Even if you got good grades in school, you might get average or even below-average grades in university.

In graduate school the focus is more on research than on classes. In fact, it is not a good idea to take (many) more classes than necessary in graduate school.

In order to prepare for research, you should have a realizable and concrete goal, you should read recent papers to learn about the state of the art in your field and you need to learn how to use scientific software such as Matlab. You might also collect some material to train and test your neural networks.

In order to perform your research, you design experiments in which you test your own method and compare it to at least one standard method. Then you write your own paper and submit it to a conference or journal to get it published. Note that publishing papers is an absolute requirement for graduating from doctoral course.

Professors and researchers are *extremely* busy with their own research, course work and administrative stuff. They don't really have much time for helping graduate students with research.