Questions about Graduate School

[Panphobia]

I am currently a 3rd year at the University of Guelph and I am thinking about going to grad school. I definitely don't want to go to the University of Guelph graduate school because it's complete crap. But just to give you an idea of what I am working with, I currently have a 94% cumulative average. I took 2 graduate courses already, in which I wrote 2 papers. I am going to be taking a research credit with one of my professors in combinatorial and graph algorithms (this is the field I want to go into). I am also going to take the next course after the research course in which you write a thesis on the research you did in the first course. I am also volunteering at my local highschool after school helping the students learn algorithms and prepare for programming competitions. If I want to get into grad school in something like Waterloo or Toronto, is this enough? Or should I be doing more? Other than this, any more info or advice? Thanks!

[Dan]

If there is a professor at Guelph doing research you are interested in and you already have a relationship with him why not consider going there? Who your supervisor is, what research you are doing and your publication record are far more important than what school you are going to (assuming it is accredited). I strongly recommenced picking a potential supervisor rather than a potential university. Not all faculties will have members doing research in all areas, so it is important to find a department with a research group (or at least a professor) doing research you are interested in.

Now to address your question. The best thing you can do to increase your admission chances for grad school is to have a potential supervisor lined up before hand. You should be meeting with potential supervisors and discussing working with them, your research interests, etc. In many cases supervising faculty members have a large say in graduate admissions and having one that wants to work with you would go a long way to being accepted.

Another thing you should do is apply for funding/scholarships. You can do this before you are accepted and being funded makes you a far more attractive prospect to potential supervisors. I believe that the NSERC deadline has past for next year, but you can probably still apply for OGS (each University has a different deadline and application process for OGS now so it will depend on where you are applying).

When you say that "I took 2 graduate courses already, in which I wrote 2 papers", do you mean you published those papers? If they were published in a peer reviewed conference or journal this could help significantly. Be sure to mention this in your application and to potential supervisors.

Your cumulative average sounds great, but grades are less important for graduate applications (assuming they are over the minimum and would not make you ineligible for funding/scholarships) then for undergrad. Most departments also have their own method of calculating your grade (for example some only consider 4th year courses, your last X number of credits, courses taken in your last Y years, etc.)

Most graduate applications have a statement of purpose section that you should put some work into. There are many resources you can find on Google to help you with this but you should at a minimum have someone read it over for you.

You will also likely need at least two references. These should be professors you have worked with and trust to give you a strong recommendation. Make sure they actually complete and submit the reference form on time (you don't want your application thrown out because your reference forgot to submit their part).

Are you applying for Masters or trying to jump right into a PhD? If you are applying for a Masters I would say your odds are good if you talk to a potential supervisor before hand. PhD might be more difficult as you will be competing with applicants with Masters degrees and publication records. If you do want to skip the Masters it may be a better idea to apply for a Masters and then switch to PhD after a year or two (depending on what the departments regulations are). Personally, I don't recommend skipping the Masters though.

Finally, you should be sure this is something you want to do. Graduate degrees are not always a wise investment and in many cases you will never recoup money that could have been made if you had just gone right into work in industry (you might make a higher starting salary but will be at least 2 years behind in experience). I would only recommend it if you want a more research focused job or to go on to a PhD and then academia (risky as there are few jobs right now).

Edit: Also I should point out that sadly the theory fields in CS are not exactly the most popular right now. While this should not stop you from pursuing your interests, it does mean it will be harder to get published, get funding and later a job if it is the sole focus of your research. The big areas right now are Big Data (and Data Analytics), Cloud Computing, eHealth, Machine Learning (mostly Computer Vision) and Bioinformatics. If you can some how involve one of those into your research, it will make your life easier.

[rdrake]

Dan is wise. Listen to him.

Dan @ Wed Jan 06, 2016 6:13 am wrote:

I strongly recommenced picking a potential supervisor rather than a potential university.

If you are a strong enough candidate, your potential supervisor will fight for you. Do not depend on a school's reputation alone, your supervisor's reputation is what allows them to attract funding, not where they work.

Dan @ Wed Jan 06, 2016 6:13 am wrote:

Another thing you should do is apply for funding/scholarships.

If you have funding you are a much smaller liability to your supervisor. This greatly increases your odds of getting in.

Oh, and if you're already 100% funded, your supervisor may "top-up" your funding so you'll be paid more. This is unlikely if you do not have an external scholarship. This is especially important for international students as the university takes back 75-90% of what you're paid just in tuition.

Dan @ Wed Jan 06, 2016 6:13 am wrote:

Personally, I don't recommend skipping the Masters though.

I too have never heard a CS prof suggest skipping a masters. A PhD is definitely not for everyone and is a huge endeavour. A masters is a great way to determine whether or not you actually want to do a PhD, and you can always fall back to it if you don't succeed at a PhD.

Dan @ Wed Jan 06, 2016 6:13 am wrote:

Finally, you should be sure this is something you want to do. Graduate degrees are not always a wise investment and in many cases you will never recoup money that could have been made if you had just gone right into work in industry (you might make a higher starting salary but will be at least 2 years behind in experience). I would only recommend it if you want a more research focused job or to go on to a PhD and then academia (risky as there are few jobs right now).

Odds are you'll become a sessional which means you'll have no guaranteed income and be treated like dirt. Your best bet would be to get your MSc and go in to industry. This will maximize your earning potential. You have to really love your research area to even complete a PhD.

tl;dr: Picking a supervisor + having funding == almost guaranteed entry (assuming you have good enough grades, which you do).

[Panphobia]

Dan @ Wed Jan 06, 2016 6:13 am wrote:

If there is a professor at Guelph doing research you are interested in and you already have a relationship with him why not consider going there? Who your supervisor is, what research you are doing and your publication record are far more important than what school you are going to (assuming it is accredited). I strongly recommenced picking a potential supervisor rather than a potential university. Not all faculties will have members doing research in all areas, so it is important to find a department with a research group (or at least a professor) doing research you are interested in.

Now to address your question. The best thing you can do to increase your admission chances for grad school is to have a potential supervisor lined up before hand. You should be meeting with potential supervisors and discussing working with them, your research interests, etc. In many cases supervising faculty members have a large say in graduate admissions and having one that wants to work with you would go a long way to being accepted.

Another thing you should do is apply for funding/scholarships. You can do this before you are accepted and being funded makes you a far more attractive prospect to potential supervisors. I believe that the NSERC deadline has past for next year, but you can probably still apply for OGS (each University has a different deadline and application process for OGS now so it will depend on where you are applying).

When you say that "I took 2 graduate courses already, in which I wrote 2 papers", do you mean you published those papers? If they were published in a peer reviewed conference or journal this could help significantly. Be sure to mention this in your application and to potential supervisors.

Your cumulative average sounds great, but grades are less important for graduate applications (assuming they are over the minimum and would not make you ineligible for funding/scholarships) then for undergrad. Most departments also have their own method of calculating your grade (for example some only consider 4th year courses, your last X number of credits, courses taken in your last Y years, etc.)

Most graduate applications have a statement of purpose section that you should put some work into. There are many resources you can find on Google to help you with this but you should at a minimum have someone read it over for you.

You will also likely need at least two references. These should be professors you have worked with and trust to give you a strong recommendation. Make sure they actually complete and submit the reference form on time (you don't want your application thrown out because your reference forgot to submit their part).

Are you applying for Masters or trying to jump right into a PhD? If you are applying for a Masters I would say your odds are good if you talk to a potential supervisor before hand. PhD might be more difficult as you will be competing with applicants with Masters degrees and publication records. If you do want to skip the Masters it may be a better idea to apply for a Masters and then switch to PhD after a year or two (depending on what the departments regulations are). Personally, I don't recommend skipping the Masters though.

Finally, you should be sure this is something you want to do. Graduate degrees are not always a wise investment and in many cases you will never recoup money that could have been made if you had just gone right into work in industry (you might make a higher starting salary but will be at least 2 years behind in experience). I would only recommend it if you want a more research focused job or to go on to a PhD and then academia (risky as there are few jobs right now).

Edit: Also I should point out that sadly the theory fields in CS are not exactly the most popular right now. While this should not stop you from pursuing your interests, it does mean it will be harder to get published, get funding and later a job if it is the sole focus of your research. The big areas right now are Big Data (and Data Analytics), Cloud Computing, eHealth, Machine Learning (mostly Computer Vision) and Bioinformatics. If you can some how involve one of those into your research, it will make your life easier.

Well in recent years the PhD program at Guelph got totally cancelled because the students they were graduating knew next to nothing. They only recently brought it back and it isn't even a computer science PhD you'd be getting, you would have to do an interdisciplinary PhD. On the topic of good fields to be in, is Computer Security a good field? Personally I really like the theoretical computer science, and find a lot of the fields you listed as kind of boring, with the exception of machine learning. I was thinking of finishing a PhD and going into academia in the long run. Another question is, what are some of the things people look for when hiring you in academia? My professor also suggested that I finish a Masters before PhD because a PhD isn't for everyone and it's better to waste 2 years of your time than 3-4. The field that my professor is in is really combinatorial algorithms and he studies de Bruijn sequences (if you've heard of them, apparently they have an application in biology). Anyway, since I know this is anecdotal so it shouldn't be taken as fact, but all of the professors I know at my University that are in a theoretical field have no problems getting published, actually they publish the most papers. Also, the fields you specified seem over saturated to me, in my school the professors that take students under these fields have something like 10-15 students. One of the arguments my professor made for going to graduate school is that, as undergraduate students we are already used to being poor, so being poor for a few more years shouldn't be a problem (hahaha).

[Dan]

Panphobia @ 6th January 2016, 10:35 am wrote:

Well in recent years the PhD program at Guelph got totally cancelled because the students they were graduating knew next to nothing. They only recently brought it back and it isn't even a computer science PhD you'd be getting, you would have to do an interdisciplinary PhD.

If that is the case, that would be good reason to not get a PhD there but you could still do a Masters with the professor you mentioned. It is common (even encouraged) to do your Masters and PhD at different universities. Having all of your degrees from the same institution is sometimes looked down upon (especially if you were to go on to work at the same institution).

Panphobia @ 6th January 2016, 10:35 am wrote:

On the topic of good fields to be in, is Computer Security a good field? Personally I really like the theoretical computer science, and find a lot of the fields you listed as kind of boring, with the exception of machine learning.

Computer Security has a number of sub fields/branches (e.g. cryptography, applied cryptography, access control, software engineering relating to security, etc.) with some being more popular than others. My research is on the access control side of things and I can tell you that it is not as popular as I would like. What you can do is combine the field you are interested in with a currently popular topic. For example with access control you could do research on access control for cloud computing or big data (both more popular topics right now).

Panphobia @ 6th January 2016, 10:35 am wrote:

I was thinking of finishing a PhD and going into academia in the long run. Another question is, what are some of the things people look for when hiring you in academia?

I am still just a PhD student but from what I have heard the big thing will be your publication record. There are a number of metrics for evaluating someones publication record (e.g. h-index, i10-index, etc.) but almost all are based on how many citations your papers have received. While doing your Masters and PhD you should also be trying to publish your work in peer-reviewed journals and conferences. Personally, I think a good goal is to have a published paper for each chapter of your thesis (not including the intro and conclusion chapters) or at a minimum one paper a year. I have also heard people say you should have an h-index equal to or greater the numbers of years since you have completed your PhD.

Another big thing will be proof that you have some teaching experience. For example working as a lecturer, teaching awards, teaching certificates (most universities have teaching courses for grad students) and having been a teaching assistant while in grad school.

Showing that you have successfully applied for and received funding form external sources is also a big deal (e.g. NSERC, OGS, etc.).

Smaller things would be experience supervising students (e.g. supervising an undergrad thesis student while being a PhD student), committee experience, having helped run/organize conferences, being a reviewer for a conference/journal, etc.

Being a postdoctoral researcher after completing your PhD and before trying to get a tenure track job would help greatly with all of the above.

Panphobia @ 6th January 2016, 10:35 am wrote:

My professor also suggested that I finish a Masters before PhD because a PhD isn't for everyone and it's better to waste 2 years of your time than 3-4.

In addition to that, doing a Masters first lets you get more experience in teaching and research. It also gives you an opportunity to publish more and build up a better record than you might have if you just jumped right into a PhD. If your university has an appropriate PhD program, there is normally a process to transfer from Masters to PhD if you chose to do so after a year or so of the Masters. Personally I would recommend starting in a Masters program as that gives you the option to exit sooner with a degree if things go horribly wrong.

I was in a similar situation as you a few years ago. The school I did my undergrad at did not have the best reputation for their graduate program (due to their small faculty and few students doing thesis based programs) and only had a Computer Science Masters (the only PhD option for CS was interdisciplinary). I did my Masters at that school but moved to a larger more reputable one for my PhD.

Panphobia @ 6th January 2016, 10:35 am wrote:

Anyway, since I know this is anecdotal so it shouldn't be taken as fact, but all of the professors I know at my University that are in a theoretical field have no problems getting published, actually they publish the most papers.

The problem is more about funding and getting a job after your PhD (also if a field is more popular it is easier to get your work noticed and there are more places to publish it). If you are doing good research getting published should be possible so long as there is a journal/conference that covers the topic.

Panphobia @ 6th January 2016, 10:35 am wrote:

Also, the fields you specified seem over saturated to me, in my school the professors that take students under these fields have something like 10-15 students.

More funding often means more students. More funding = more students = more papers = more citations/reputation = even more funding. It does mean the prof has to spend less time with each individual student but it also means those students have more opportunities to collaborate with the other students and publish more.

Panphobia @ 6th January 2016, 10:35 am wrote:

One of the arguments my professor made for going to graduate school is that, as undergraduate students we are already used to being poor, so being poor for a few more years shouldn't be a problem (hahaha).

It's not that bad if you get funding. I was able to buy a house while being a PhD student but that was largely do to my wife also having an income and being eligible for some assistance programs. It did end up burning down, but that's another story.

Edit: One more PHD comic about picking a field:

[Panphobia]

I have actually been a TA for nearly 2 years. I have TA'd mostly first year computer science, but I did TA a 3rd year course last semester because the University was short on grad students. About the NSERC scholarship, my professor thinks I am a good candidate so he is trying to prepare me as much as he can for when you need to apply. Also another thing about graduate school in general, does your degree have more to do about what school you went to or what advisor you had? I know I don't want to be at Guelph for a PhD but then what should I aim for, Waterloo or Toronto, or somewhere else with a "better" advisor? A comment about your "profzi" comic, one of the reasons the Guelph graduate program is so bad is because it has very poor funding, we only have something like 30-40 grad students and a little under half of them are being advised by one professor.

[Dan]

Panphobia @ 8th January 2016, 4:32 pm wrote:

About the NSERC scholarship, my professor thinks I am a good candidate so he is trying to prepare me as much as he can for when you need to apply.

I think the NSERC deadline for PhD has past for next year. You could probably still apply for Masters version of NSERC for next year or OGS if you hurry.

Panphobia @ 8th January 2016, 4:32 pm wrote:

Also another thing about graduate school in general, does your degree have more to do about what school you went to or what advisor you had? I know I don't want to be at Guelph for a PhD but then what should I aim for, Waterloo or Toronto, or somewhere else with a "better" advisor?

Far more to do with your supervisor in my opinion, especially since you want to go into a largely theory based field where you don't necessarily need access to any special resources, equipment or labs. Normally you will research a topic closely related to that of your supervisor so if the "better" school has no one in that area you would have to change what you are studying or go somewhere else. You should look into potential supervisors publication records and look up information about their research (e.g. are they part of a research group, how many grad students do they have, are they a research chair, what kind of funding do they have (it can be seen as impolite to as this to them directly), etc.) to evaluate who the "better" advisor is and not base it on the school.

There some advantages to going to a bigger and "better" school. You might get paid more for TAing and more guaranteed funding (this information should be available on the schools website). You will also have more graduate courses to pick from (most CS graduate programs require you to take a few CS grad courses). There also may be better research resources and equipment. However there are also downsides to going to a big school. You will be a small fish in a big pond and your supervisor may have a large number of students. You will have more competition for internal awards and resources and have to fight harder to be in the top of your class.