Getting into Grad School with a substandard GPA

August 29, 2007 1:16 PM

I'm looking for advice on apply to grad school along the lines of Computer Science/Cognitive Science. Major problem: Low CGPA blues.

I've reviewed some of the previous threads on this topic, and they've been helpful. However, I don't think the overwhelming sense of panic that wells up on a regular basis in me will be assuaged without some personalized advice, so I'd appreciate anyone who knows the ropes on what it takes to get into grad school to weigh in on this. The short of it is that when I finally graduate in two years, I hope that I can get my GPA up to a 2.7. Here's the long of it:

Flew threw high school getting a 90 average without ever doing a shred of work. Got into the University of Toronto.
Went. Recieved a CGPA of .73 in my first year. Dropped out. Moved to British Columbia for two years. Had fun. Came back.

At this point I still didn't know what I wanted to do, and still didn't know how to study. Flew threw another three years of random courses. When it finally dawned on me that I really, really, love academics, that I've assumed all my life that I was going to go to grad school, and that I was royally fucking up, I changed my tune. At this point I had 15 credits under my belt with a CGPA of 1.7.

After having taken courses from roughly 12 different departments, I decided that I would pursue cognitive science and artificial intelligence, which means that, since I do want to graduate before I'm 30, I had to get a computer science degree in three years. Over the last academic year, I scored a 3.9. I stumbled with some summer courses that I was taking, only scoring a 3.0, which is probably what has caused the latest cycle of panic and depression, leading me to post here.

So... some questions. Assuming I can get straight As for the next two years, am I still fucked?

I'm pretty sure I can get stellar recommendations, because I'm always very involved in my classes, and usually come off as knowing exactly what's going in.

I've also been told by some follow students that U of T is notorious for marking very hard, and that some grad schools do note this. I find this somewhat dubious, but I guess, in general, does anyone know if U of T rings a positive note with admissions?

I could go on like this for a while (and am trying very hard to restrain myself), but really I'd just appreciate any information on how best to improve my chances on getting into a good masters program. (Any specific information on the nature of those few cognitive science departments in the world get bonus marks)

posted by Alex404 to Education (25 answers total)

the number one thing you can do is to get involved in a lab group, do some research and decide exactly WHAT you want to research in cognitive science/ai.
posted by noloveforned at 1:30 PM on August 29, 2007

General advice:

Do try hard to keep your GPA up for the last two years. That will help mitigate your early academic flailing, tho it probably won't erase it completely. From your description, it seems like the bulk of your coursework is still to come - that can help dilute the 15 credits of suckage.

GREs are important. Study your ass off.

Don't count on recommendations from professors whose class you've been in. Try to get involved in research with a professor whose interests are similar to yours. It's more impressive to have a couple papers or posters that your name is on than to be known as the smartest guy in the class.

In your application letters, it is usually helpful to identify one or two faculty members who you might be interested in working with. I'm not sure if this is as important for masters only apps, but it lets the school know you are actually interested and did your research. Also, don't hesitate to contact profs at schools your applying to get more information about the program... this helps you determine if it's really a good fit and helps your name stand out come admissions time.

Don't, however, pester them.

Finally, cognitive science is a broad area, covering philosophy, comp sci, psychology, neuroscience etc etc etc. I got my degree through a joint cog sci/psych program, others did CS/cog sci or whatever. It's possible for a comp sci major to get in to a psych program and still get the full cognitive science experience.
posted by logicpunk at 1:39 PM on August 29, 2007

...schools your applying...

Grrr... you're, obviously.
posted by logicpunk at 1:42 PM on August 29, 2007

Seconding the lab, of course. But maybe if you can stay in school longer and bring up your GPA?
posted by k8t at 1:43 PM on August 29, 2007

Also, can you try to slide into your fave undergrad prof's grad program?
posted by k8t at 1:46 PM on August 29, 2007

Getting involved in a lab group might be easier said than done if you don't qualify for an NSERC undergrad research assistant award -- that usually requires at least a B-/B. I'm not sure how U of T handles research assistantship funding outside of NSERC, but even then, they may still be looking for NSERC-admissible qualifications. You might be able to petition for an exception if your weak marks weren't in any of your current major courses. As well, if you're OSAP-eligible, sometimes research assistant positions are offered, with slightly more lax academic requirements.

My experience is at another university in Ontario; there, profs just about categorically will not take on undergrad research assistants outside of some structured URA program.
posted by thisjax at 1:55 PM on August 29, 2007

Advice on applying to graduate school from a CS professor at Carnegie Mellon.

Advice on graduate school in CS, including applying.
posted by needled at 1:56 PM on August 29, 2007 [3 favorites]

It has been my experience that teachers, and even admissions boards, place more weight on your improvement of grades than just whatever it ends up being when you graduate. Plus, I think they also put more weight on classes that are in your major. They look at more of the whole picture than what's just described in a simple metric.

When I applied to the university (undergrad) I'm attending now, there were several essay fields on the application. In one of them I was given the opportunity to explain what I had done with my life, including why I dropped out of school, joined the military, and resumed my schooling while I was in the military. I don't know if I barely made it in (to the only university I applied to), or if I was accepted with flying colors, but I'm in now. I'm sure there will be a similar field on your application.
posted by philomathoholic at 1:59 PM on August 29, 2007

Gah, I can't type:

As well, if you're OSAP-eligible, sometimes research assistant work study positions are offered, possibly with slightly more lax academic requirements.
posted by thisjax at 2:02 PM on August 29, 2007

if you are applying to something that lets you write an accompanying letter, plot the grades out, with a big arrow pointing to when they got better, saying "this is when i realised i wanted to do this". seriously. i have included graphics in letters like this and always got compliments (i'm assuming you have some basic aesthetic competence - it should look good).

it makes your point, "explains" the low grades, will get you noticed, and makes a strong case that you really do care.
posted by andrew cooke at 2:04 PM on August 29, 2007 [1 favorite]

If you're applying to any half-reputable place, nobody cares about GPA. It says nothing, and is completely uninformative. Seriously. I doubt anybody would even raise a snuff. However, that you are concerned about it indicates to me that you might not have researched this enough. Talk to professors about grad school right away. Talk to current grad students about it. Be polite, but make yourself known and get information.

First, schools care about you being able to carry on research. Second, they care that you will stick around, finish your Ph.D. and continue research after you finish.

There's only one way to demonstrate this, and that is by doing research and publishing. Join a lab right away. In fact I don't think you should even apply to grad school until you have contributed enough to have your name as an author on a manuscript. If you graduate before your research is done, and you are still interested, see if your advisor can hire you for a year as a research assistant.

This is in your personal interest. First, it will tell you whether or not you really want to go to grad school. By going to grad school, you will sacrifice wealth over your lifetime, and maybe hapiness. Second, having a publication is nearly a requirement for most top programs, so you are closing all sorts of doors without one. Third, unless your recommendations are talking about specific actions related to research, they are worthless, no matter how glowing. Classroom performance is so different form anything in grad school that they might as well be talking about your choice in clothing.
posted by Llama-Lime at 2:16 PM on August 29, 2007

Err, scratch some of the scathingness, masters are slightly different. But you should seriously consider going in on a Ph.D. track and finishing with a masters. It's slightly sneaky, so don't tell anyone, but totally within the bounds of moral behavior in the academic world.
posted by Llama-Lime at 2:19 PM on August 29, 2007

I was in a fairly similar situation but in Computer Engineering. I had around a 2.0 GPA after my first 3 semesters. I had an internship and it was a real wakeup call, I didn't want to end up like the guys in the job I worked at. I came back to school and made nearly a 4.0 my last 2.5 years. And I am now working at the place I originally set my sights on (although I have now changed my goals and am working on my next career).

I only applied to one school for graduate school but a selling point on my essays was how well I had done since my "epiphany". Use the fact that you are passionate about your current major as a selling point in any essay you can.

On the resume front I'd often quote my "In major GPA" which was very high since my freshman year I took mainly general engineering. You might be able to also quote your CS GPA to show that you are very good at what you will be doing in graduate school.

GPA is only one part of your application, make the rest of your profile stick out.
-Be sure to score well on the GRE, especially quantitative. It is not that difficult to get a perfect score, so shoot for that on that section.
-Get to know faculty members! I regret this a lot. Do work "for free" for a professor, it will pay off in recommendations. You won't ever have to feel guilty asking for it and you won't get most professors copy and paste recommendation.
-Get in on research during undergrad. You won't get paid or anything, but you *might* be able to get your name on a research paper which is big bonus points.
-If you're doing AI try to get involved in an opensource project.


Just go for it and don't worry about it. Stay clear on your goals (get into a *good* grad school) and use that to stay enthusiastic.
posted by wolfkult at 2:24 PM on August 29, 2007

Recieved a CGPA of .73 in my first year. Dropped out. Moved to British Columbia for two years. Had fun. Came back.

I made it into a good grad school and my freshman GPA was in this ballpark. A professor I know did something like this as an undergrad and is now a really prominent researcher in his field. So it happens, and more regularly than some might think. However, I did pull significantly up after my freshman year, by retaking classes, and I later added a second major (the field I am in grad school for now), in which I had a 4.0. I also had really good recommendations (or so I have heard) from really good people, and recommendations are probably the most important thing (incidentally my main one was from someone who knew my academic history and knew me pretty well; I don't know whether she addressed it in the letter but she may have, so you should probably talk about this issue with your letter writers.) Also, just doing well in a class doesn't necessarily mean getting the best letter you could -- the better someone knows you, the better a job they can do on a recommendation. If you still have a year (I can't tell when you want to apply) you should try to do something (work in a lab, do a senior thesis, try to get some kind of research assistantship, even unpaid -- cog.sci. people often need lab techs to run experiments) so as to get to know some particular faculty member well.

I can answer more questions over email (in profile) if you want.
posted by advil at 2:41 PM on August 29, 2007 [1 favorite]

Response by poster: Wow. Thanks for everyone's replies. I'd mark them all as best answers, but that would be silly.

Please, if anyone has anything else they'd like to offer, don't hesitate.

It seems there's a consensus on the importance of real lab work though, so I will definately make that my goal for next year. I'm nearly certain that I am ineligable for an NSERC grant, but I will take a much harder look at the work-study programs that I'm eligable for due to OSAP.
posted by Alex404 at 3:55 PM on August 29, 2007

Response by poster: Also, just to clarify, I've got two years left.

Also, Llama-lime, I don't quite understand what you mean. I assumed that I should start off with a masters, because it would be easier to get in, and then I would move on to *the idea PhD program*. What's the logic of doing a PhD and then a masters?

Also (again!) I've investigated grad schools to an extent, but part of the problem is that I don't know who to really talk to. I've talked to my own schools guidance councillers, and I've contacted two admissions people at different schools. Everyone basically keeps telling me the same thing, "try hard. Do your best. You've still got a shot." Which is nice, but not particularly specific. I want something to *do*, like try and do work for professors.b
posted by Alex404 at 4:04 PM on August 29, 2007

What's the logic of doing a PhD and then a masters?

Many programs have Ph.D. track programs for which a masters is not required. Usually you earn a masters degree in the course of your Ph.D. edumacation, or you skip the masters altogether. It can also be awarded as a consolation prize for people who hate graduate school.

Really, if you're going to possibly maybe want to get a doctorate, try to get into a Ph.D program rather than a masters. Doctorates take long enough in north america without the additional time spent in a masters prog.
posted by logicpunk at 4:19 PM on August 29, 2007

nthing everyone's advice about getting involved in research and seconding advil's advice that you should volunteer if you have to. while it may be possible to get into computer science grad programs without research experience, it's increasingly difficult to get into a cog sci program if you haven't done at least some undergraduate research. you will be competing with people who have already given conference presentations and have publications - that will hurt you more than your low GPA. everyone's already said good and constructive stuff about GPA, so I'll just talk about the reputation of U of T from a US perspective. given that it's one of the best schools in Canada, it has a good reputation here as well. however, it's not considered as highly as, say, Harvard or Yale. but in general I would encourage you not to put too much emphasis on the prestige of your undergraduate institution. in my (cognitive science, Ph.D.) program we have students from all kinds of educational backgrounds - highly ranked colleges, big state schools, small colleges in the middle of nowhere. what seems to matter most is what you do with the options you have. a student from a school like U of T, which has stellar research programs and many opportunities for undergraduates to get involved, would be expected to take advantage of those opportunities as much as possible. on the other hand, students who go to small liberal arts colleges, which generally lack such research programs, are not expected to go out of their way to create such opportunities for themselves. in other words, a student from a less well-regarded school who has done a lot of research and has good letters from labs she's worked in is more likely to get into a grad program than a student from a highly-ranked school who has just taken classes for four years. trust me, I've seen those people get rejected, despite good grades.

just on the anecdotal level, though, I do know of at least three U of T undergraduates who went on to do cognitive science in the US. two went to MIT. we had one in my program last year, but he left halfway through the semester - he wasn't happy living in the States.
posted by dropkick queen at 4:22 PM on August 29, 2007

oh, I just saw the comment about master's vs. Ph.D., so I'll comment on that, too - logicpunk is right that you should apply to Ph.D. programs if at all possible. those are generally fully-funded and you'll get a master's degree of some kind on the way. Master's programs are less prestigious and usually don't come with funding. they can be useful as stepping stones to better programs, but you should only go that route if you get rejected from all the Ph.D. programs you apply to and don't feel like waiting to reapply for the next year.
posted by dropkick queen at 4:26 PM on August 29, 2007

Alex404, Ph.D. students are usually guaranteed funding, but masters students are usually not. In CS at least, people entering a program with a masters degree and attempting a Ph.D. are kind of rare, because you'll end up duplicating a lot of work. So usually one applies to the Ph.D. track straight away (without a masters yet), or when you finish your masters you immediately apply for the Ph.D. program.

If you apply as a Ph.D. student, you will then be getting paid enough to live at a subsistence level without taking on large debt, and be treated as more of an equal and colleague. When you have performed masters equivalent work, you can leave at any time with that degree. Some professors might view intentionally taking this route as underhanded, because they devote more time and resources to Ph.D. students than to masters students. So when you talk to professors about grad school, do not under any circumstances mention something like this, because it would be a huge red flag and bring up bad blood. I feel kind of bad even mentioning it, but it happens fairly frequently, and is very much in students' interest.

But most of all, talk to your current professors about their research, find out more, read their papers. If you can, try to be able to read the current literature in general in the field a year or 18 months from now (this is *big*, being able to read a current research paper and discuss it with a potential advisor will put you ahead of most grad school applicants, but it will also require a huge time investment and a good attention span). See if you can go to a couple research group's lab meetings. Keep an eye out for seminars, and sit through them despite not understanding much of anything, but take notes and learn what little you can. You want to be able to formulate interesting questions in the field as well as understand what other people are doing.

This will all help you determine whether you should even go to grad school. By the way, you shouldn't assume that grad school is where you want to go, or that grad school is where all the smart people go (it isn't). There's an opportunity cost to spending that year or two to get a masters, and you should figure first if it would appeal to you, and what opportunities that it would open up that you would otherwise not have. One should only go to grad school for specific purposes, with specific goals. If the most specific and long term goal you can formulate is "get a masters degree" then its probably not in your best interest.

(Sorry if I'm ranty, I'm a disgruntled grad student who's glad he's doing what he's doing, but was totally caught unaware by what he was getting into.)
posted by Llama-Lime at 4:50 PM on August 29, 2007

Response by poster: I just wanted to thank everyone again who's posted. This has been really eye opening.

I've been very keen to do research, but I was afraid to pursue it because I assumed it was a luxery only available to the top performing students.

It will be priority number one for me now though. Even if I have to build a talking robot head and carry it around with me on a tour to every intelligence researcher at U of T, it will be done.
posted by Alex404 at 4:56 PM on August 29, 2007

Response by poster: Oh and Llama-Lime, I appreciate the advice and grad school warning. I am certain as I can be though that this is what I want to do. I would be embarassed to describe my true feelings towards cognitive science, but sufficed to say, they are strong.

You've definately let me know though that I need to explore a lot more what grad school is really about.
posted by Alex404 at 5:01 PM on August 29, 2007

I would be embarassed to describe my true feelings towards cognitive science, but sufficed to say, they are strong.

Nothing like grad school to cure you of those feelings, one way or the other...

Anyway, I don't have too much to add. If you are really kicking ass now, and can get good recommendations, I would not worry too much. In-major GPA is important. Research is great, yes, but not at all required anywhere but really top schools. I would not worry about your GREs, frankly. If you are good at CS you should be absolutely fine at the quantitative section (a perfect score is far from the 99th percentile), and the verbal section is not too bad (or, really, too important, unless english is your second language). As everyone else has said, apply to PhD programs. Drop out with a Masters when you realize that CS academia is a total joke. Make $$$.

Unless things have really changed since I sat on the admissions committee for my particular grad program (a good school, less than 5 years ago), I do NOT think most graduate schools expect or require publications from undergraduates to get in. Yes, they will make you more attractive to the really top programs, but so many undergraduate publications are a joke I wonder that CS programs doesn't try to curtail them a bit. Frankly, I think it is a little antithetical to the undergraduate experience to expect people still learning the breadth of their subject to have the time and knowledge to contribute significant research.
posted by ch1x0r at 6:37 PM on August 29, 2007

Some of the advice here may be more about grad schools in the US. I know in philosophy, anyway, it's much more common in Canada than in the US to get a master's before going to a PhD program. You should speak to an advisor - or several! - in your department to get advice tailored to your situation.

Seconding that grad school admissions does recognize an improving record of grades. Also they generally care a lot about what grades you get in the field you're applying for, and not at all about unrelated grades. Eg if your hard science grades suck, that's a problem until you can get some marks on your transcript showing that you can handle hard science. But if you have bad marks in English lit, that's not as much of a problem.

If you are an older and determined student, the thing to do is speak to the people in your department who can help you get real experience, and who would be the ones writing your letters. (Yes, the letters are probably the most important part of the application.) They will be glad you're excited, and they will help you figure out what steps to take. And you'll be laying the foundation of a better relationship with them, which they can then talk about in the letters.
posted by LobsterMitten at 8:40 PM on August 29, 2007

As I know in most US Universities they consider the GPA of last two years of your Bachelors degree. I think this might be to your advantage.
posted by WizKid at 7:59 AM on August 30, 2007