Could use some advice on how to improve profile for applying to a PhD later this year

[correlatesoftheory]

Greetings. As stated in the post, I just want some advice on where I can improve my profile. Also, if someone is willing, address some of the (potential) weaknesses of my profile, and how I might address them. My goal, like many people here at Grad Café, is to attend a top-tier PhD program. I'll just go ahead and post some of my stats below...

 

I'm a transfer student, so I'll breakdown my GPA like so: 

• Currently a senior attending a top-10 public research university in California
• UC GPA: 3.88
• Community college GPA: 3.4

Quantitative coursework:

• Elementary statistics
• Short Calculus A (currently enrolled)
• Short Calculus B (taking next quarter)
• Quantitative Analysis of Political Data
• Scientific Study of Politics
• Also, I've produced an honors thesis which contains multivariate statistical analysis/time series regressions.

Research Experience:

• Currently a research assistant for two "Distinguished" professors. One of which I've been working for since last October, and will continue working for through June. The other I've been working for since January of this year, and will most likely be finished by the end of this month
• Honors thesis is currently under review for publication in the undergraduate research journal 

Letters:

• Both aforementioned professors have agreed to write me good letters of recommendation. It should also be mentioned I took one class with each of them, and excelled.
• One letter of recommendation from respected Associate Professor who supervised my honors thesis 

Other information:

• Studied abroad at well-known French university in Paris and achieved high marks plus became proficient in French
• Knowledge of STATA, Python 

 

Concerns/Potential weaknesses:

• Low GPA and went part time my first three semesters at community college. After that, I enrolled full-time and achieved a high GPA
• My last three quarters at my university I've been going part-time. The reason is because I finished all of my coursework for my degree last quarter, then was offered a research position and also wanted to take two math classes in order to bolster my quantitative skills. My fear is that an admissions committee will see this as slacking off ("Why didn't he take more classes in his final year?") 
• Have not taken the GRE yet, but am still worried about getting mediocre scores--especially the quantitative section (advice on how to avoid this would be nice) 

 

Questions:

• I have one quarter left, and then I'm done. I'm currently enrolled in Short Calculus B and one other required writing class. What other coursework could I take that might improve my profile without bogging me down too much? I'd like to start studying for the GRE by next month and start working on my statements of purpose.

Schools where I think are a good fit for me, given my research interests:

• Stanford
• NYU
• UCSD
• University of Michigan, Ann Arbor
• Duke
• Rochester
• Princeton 

 

Let me know what you think. I appreciate anyone who responds. Thanks! 

[cooperstreet]

I don't see any weaknesses really. I think it shows character that someone attended a CC, did well, transferred to a good school, then did really well.

 

My suggestion for you: do a lot of reading of political science research. Beef up your contextual knowledge, rather than just your technical skills. This will enable you to generate interesting research topics/questions/ideas in your SOP. Your math background is more advanced than much of my cohort.

[correlatesoftheory]

I don't see any weaknesses really. I think it shows character that someone attended a CC, did well, transferred to a good school, then did really well.

 

My suggestion for you: do a lot of reading of political science research. Beef up your contextual knowledge, rather than just your technical skills. This will enable you to generate interesting research topics/questions/ideas in your SOP. Your math background is more advanced than much of my cohort.

Good call. I read plenty of journal articles and a couple books for my honors thesis, but I should be reading more as I prepare my statements of purpose later this year. Thanks! 

[law2phd]

Good call. I read plenty of journal articles and a couple books for my honors thesis, but I should be reading more as I prepare myee statements of purpose later this year. Thanks! 

 

I think the SoP may be the most important aspect of your file.  The rest proves that you are minimally competent to be at the department; the SoP shows that you know what you are getting into, that your interests are in areas of strength for the department, and that you are likely to do the sort of scholarship in your career that will bring prestige to your degree-granting institution.  It's good that you have the better part of a year to make it as strong as possible.

 

I would particularly focus on reading articles from recent editions of top publications.  Pick up a few recent issues of the American Political Science Review, the American Journal of Political Science, etc.  Read all of the abstracts.  This will give you a lens on which subfields and sets of research questions are of most interest to you.  Read some of the articles that seem most interesting to get a feel for how political scientists talk (how do you say what you want to say in the SoP as a political scientist?) and argue.  Be able to convince admission committees that you know even more about the discipline than your credentials indicate (although your credentials are strong as it is).

 

This is the stuff I really wish I had known when I was applying.  I focused on research topics and subfields as they were discussed in the poli sci classes I had already taken.  Unfortunately, those classes are often largely restricted to landmark articles and ideas that are a decade or three out of date.  The methodology I'm suggesting is what I've been using to find advisors post-admittance.  I think I would have been accepted to more programs had I been doing this from the beginning.

Edited March 14, 2015 by law2phd

[correlatesoftheory]

I think the SoP may be the most important aspect of your file.  The rest proves that you are minimally competent to be at the department; the SoP shows that you know what you are getting into, that your interests are in areas of strength for the department, and that you are likely to do the sort of scholarship in your career that will bring prestige to your degree-granting institution.  It's good that you have the better part of a year to make it as strong as possible.

 

I would particularly focus on reading articles from recent editions of top publications.  Pick up a few recent issues of the American Political Science Review, the American Journal of Political Science, etc.  Read all of the abstracts.  This will give you a lens on which subfields and sets of research questions are of most interest to you.  Read some of the articles that seem most interesting to get a feel for how political scientists talk (how do you say what you want to say in the SoP as a political scientist?) and argue.  Be able to convince admission committees that you know even more about the discipline than your credentials indicate (although your credentials are strong as it is).

 

This is the stuff I really wish I had known when I was applying.  I focused on research topics and subfields as they were discussed in the poli sci classes I had already taken.  Unfortunately, those classes are often largely restricted to landmark articles and ideas that are a decade or three out of date.  The methodology I'm suggesting is what I've been using to find advisors post-admittance.  I think I would have been accepted to more programs had I been doing this from the beginning.

Understood. Any advice on what classes I should be taking for my last quarter without bogging me down too much? Right now (as mentioned above) I'm enrolled in Calculus B (differential, integral calculus) and one other required writing class. My math professor gave me a hard time saying I should be doing more, but I've finished the coursework for my degree and can't think of what else I could do that would be worthwhile and not too time consuming.

[law2phd]

Understood. Any advice on what classes I should be taking for my last quarter without bogging me down too much? Right now (as mentioned above) I'm enrolled in Calculus B (differential, integral calculus) and one other required writing class. My math professor gave me a hard time saying I should be doing more, but I've finished the coursework for my degree and can't think of what else I could do that would be worthwhile and not too time consuming.

 

I honestly don't think course selection matters that much--including how much math training you have.  Even top departments will teach all of the math you need, provided that your GRE score indicates that you have the aptitude to learn.  To be competitive for all of the schools you selected (I'm particularly looking at Stanford and possibly NYU), I would focus on a minimum 165 quantitative, which is doable with a lot of practice at high school level math.

Edited March 14, 2015 by law2phd

[AmericanQuant]

I honestly don't think course selection matters that much--including how much math training you have.  Even top departments will teach all of the math you need, provided that your GRE score indicates that you have the aptitude to learn.  To be competitive for all of the schools you selected (I'm particularly looking at Stanford and possibly NYU), I would focus on a minimum 165 quantitative, which is doable with a lot of practice at high school level math.

 

Most of the programs you're applying to have some version of a math camp, where they'll go over all of the minimal math you need in grad school. It won't do it in much depth, and learning a bunch of math in 2 weeks isn't the best way to retain it.

 

In particular, if you have any interest in quantitative research, you'll get a lot out of a linear algebra course.  Also, some statistics courses will help you understand the literature better.  If you want to be a methodologist, you basically can't take too many stats courses.  If you're doing formal theory (i.e. game theory), you would benefit from real analysis.

 

Have you taken any econ? In many respects, economics is the language of political science.  Taking intermediate would be valuable, especially at NYU and Stanford, which have many people trained as economists on the faculty.  

 

Being a part-time student won't be a problem if you can show that you used your time well.  But if you don't have much to show for it, then it could be an issue, particularly if  other applicants will have more to show from there time in college.

 

Have you asked your "distinguished" professors about these things?  What do they say?

Edited March 15, 2015 by AmericanQuant

[correlatesoftheory]

Most of the programs you're applying to have some version of a math camp, where they'll go over all of the minimal math you need in grad school. It won't do it in much depth, and learning a bunch of math in 2 weeks isn't the best way to retain it.

 

In particular, if you have any interest in quantitative research, you'll get a lot out of a linear algebra course.  Also, some statistics courses will help you understand the literature better.  If you want to be a methodologist, you basically can't take too many stats courses.  If you're doing formal theory (i.e. game theory), you would benefit from real analysis.

 

Have you taken any econ? In many respects, economics is the language of political science.  Taking intermediate would be valuable, especially at NYU and Stanford, which have many people trained as economists on the faculty.  

 

Being a part-time student won't be a problem if you can show that you used your time well.  But if you don't have much to show for it, then it could be an issue, particularly if  other applicants will have more to show from there time in college.

 

Have you asked your "distinguished" professors about these things?  What do they say?

The Quantitative Analysis of Political Data course I took last quarter was basically intermediate regression analysis. One of my "distinguished" professors told me take 16B over linear algebra because the calculus is much more difficult to learn than the linear algebra. I've taken basic micro and macro economics, and my honors thesis is based partially around international trade. 

 

I guess it's a matter of showing how I've used my time. I'm currently enrolled in the short calculus class B and a required writing class. I could sign up for linear algebra, but I'm not sure how much this would burden me. I'll also have my research assistantship through June, and maybe through September, and the second one I'm currently involved in should last until the end of the month.  

 

One 'distinguished' professor has told me the most important part of the application is the GRE, the GPA, and the letters of rec. Statement of purpose is also important, so too is the writing sample, but he's emphasized the former three. Also, I was advised not to take anymore statistics courses because the statistics taught in grad courses will be taught in the context of the discipline. Rather, I should focus on calculus because it tends to be the type of math that students struggle most with. 

 

I met with one Stanford professor last year and I was told they look for whether 1) I know what real political science is and I know what I'm getting to and 2) I can handle the math (this can be demonstrated through taking calc courses, statistics within the writing sample, and having a high quant score on the GRE). 

 

But I take your point, and I'll be sure to ask them next time I get a chance to see them. 

 

I should also mention that I'm trying to save money on tuition, since I'm finished with my degree and am only taking math courses to bolster my quantitative skills. 

Edited March 15, 2015 by correlatesoftheory

[cooperstreet]

calculus is more difficult, imo. But the typical 3rd semester of calculus isn't that useful for what you want to do. You'll rarely be asked to integrate something by hand. The concepts are useful, yes, and doing things by hand is a great way to learn them, but its a lot of material you wont use (like trig integration, etc.).

 

I agree with whoever said that linear algebra would be useful. it will be useful for one key reason: so much shit is written in matrix notation and uses, just conceptually, not computationally, linear algebra that if you know the concepts intuitively that will really help you. 

[correlatesoftheory]

calculus is more difficult, imo. But the typical 3rd semester of calculus isn't that useful for what you want to do. You'll rarely be asked to integrate something by hand. The concepts are useful, yes, and doing things by hand is a great way to learn them, but its a lot of material you wont use (like trig integration, etc.).

 

I agree with whoever said that linear algebra would be useful. it will be useful for one key reason: so much shit is written in matrix notation and uses, just conceptually, not computationally, linear algebra that if you know the concepts intuitively that will really help you. 

So you're saying take linear algebra over calculus, considering how useful it will be in the future? 

[AmericanQuant]

Hard to say what to take because what's taught in the first semester or second semester of calculus varies from university to university.  Here are some college-level math topics, in order of importance for a social scientist:

 

1) Differentiation 

2) Integration

3) Basic Linear Algebra (Matrix Ranks, matrix inversion, bases)

4) Eigenvalues and Eigenvectors, this is usually in the first linear algebra course, taught after 3.

5) Taylor Series, infinite series, convergence (usually a second semester calc topic)

6) Computational Linear Algebra Topics (SVD, QR Decomposition, etc.)

7) Multivariate Calc (div, grad, curl, surface integrals, etc.)

8) Real Analysis

 

I'd strongly recommend learning 1-3 in college. They'll all come up in your first semester or two of methods classes in grad school, and they're sort of essential to reading the literature and understanding what's going on when you run a regression. Very possible that you can't do your problem sets without knowing them.  After that, everything is relevant, but not essential.  

 

In a math department, partial differentiation is something that is taught in a multivariate calculus class, but you can teach it to yourself pretty easily. It's not any harder than regular differentiation and you'll want to learn how to do it at some point.

[victorydance]

Honestly, you have more than a sufficient math background already, not sure why the discussion has focused mainly on this. I would focus on four things you really can control at this point:

 

1) Doing well on the GRE.

 

2) Preparing a strong master SOP and then tailoring them for each program (particularly really nailing down your research interests and potential questions).

 

3) Doing the proper research on each department and their faculties/fit with your interests. 

 

4) Perfecting your thesis into a good writing sample.

[law2phd]

Honestly, you have more than a sufficient math background already, not sure why the discussion has focused mainly on this. 

 

I realize that I've already expressed my opinion here, but I strongly agree.  I think the responses here are needlessly stressful for OP, especially if further math classes might hurt his GPA (and therefore his application file).  If he has to pay extra for these classes--as I thought I understood from one of his posts--they make no sense at all.  If the question is what OP needs to do to get into a top program, the answer is that his quant training is fine provided that it translates into a 90th percentile or better GRE score.  I was accepted to multiple quant-heavy programs, including from OP's list, and I haven't had formal math training beyond high school level calculus; a strong GRE score and a SoP that displays a working knowledge of contemporary mathematical methods as relevant to his subfield is sufficient.

 

I realize that OP may not think I'm contributing anything here, and that's fine, but the people who find this via a google search a year or two down the road are otherwise likely to make some very questionable decisions which could foreseeably cost them admission to Top 10 departments and/or cost thousands of dollars and months of time for no ascertainable benefit.  Admission committees do not need to see these courses on transcripts, and anyone who can't learn game theory or linear algebra on their own with a couple of textbooks lacks the aptitude for a quant-heavy field in the first place.

Edited March 16, 2015 by law2phd

[PrincipalAgent]

Piggy backing off of the previous two posters, I cannot stress how important the SOP and the writing sample are. These are the two things that visibly demonstrate whether you can actually do political science research. 

[correlatesoftheory]

Piggy backing off of the previous two posters, I cannot stress how important the SOP and the writing sample are. These are the two things that visibly demonstrate whether you can actually do political science research. 

This is the advice I've received from nearly every professor. As of now, I have to concentrate on writing a solid statement of purpose(s). I'm confident that my writing sample (honors thesis)--even in its current state, reads like a political science article. Though, I do need to go back and do some revising. 

 

Hard to say what to take because what's taught in the first semester or second semester of calculus varies from university to university.  Here are some college-level math topics, in order of importance for a social scientist:

 

1) Differentiation 

2) Integration

3) Basic Linear Algebra (Matrix Ranks, matrix inversion, bases)

4) Eigenvalues and Eigenvectors, this is usually in the first linear algebra course, taught after 3.

5) Taylor Series, infinite series, convergence (usually a second semester calc topic)

6) Computational Linear Algebra Topics (SVD, QR Decomposition, etc.)

7) Multivariate Calc (div, grad, curl, surface integrals, etc.)

8) Real Analysis

 

I'd strongly recommend learning 1-3 in college. They'll all come up in your first semester or two of methods classes in grad school, and they're sort of essential to reading the literature and understanding what's going on when you run a regression. Very possible that you can't do your problem sets without knowing them.  After that, everything is relevant, but not essential.  

 

In a math department, partial differentiation is something that is taught in a multivariate calculus class, but you can teach it to yourself pretty easily. It's not any harder than regular differentiation and you'll want to learn how to do it at some point.

I'm just getting through differential calculus and my next quarter of calculus will be integral. In one of your posts you mentioned that going part time won't hurt me as long as I have something to show for it. I should mention that the reason why I'm part-time now is because 1) I finished the coursework for my degree last quarter, 2) I wanted the research position, and part of the obligation was to be a continuing student at least through June and 3) I wanted to take calculus and the regression analysis course. I am taking three courses next quarter, one of them being math, the other a required writing class, and the other being a gardening class. Might anyone have suggestions what else I can do to be "productive" while not being overwhelmed/breaking the bank with tuition? 

[AmericanQuant]

As others have said, you're in pretty good shape. Having recommendation letters from well-known people seems to count for a lot, and you have them. Try to get the most out of your classes and your RA gig.  I think you'd be well-served learning as much math and statistics as you can, not for admissions reasons, but because it'll help you once you're in.

[Cactus Ed]

C'mon. This is just a humble-brag disguised as a naive question about the adequacy of qualifications. You're fine.