What skills were most useful when beginning your PhD program?

[toad1]

Reaching out to current grad students (and beyond):

What skills were most useful when beginning your PhD program? As the admissions decisions trickle in, I can't help but think to myself: what extra stuff should I do / learn / understand prior to beginning this journey (whether it's specific math skills / programming skills / language skills / etc.)

What gave you an edge? -- or, what set you back and you wish you had known beforehand?

[buckinghamubadger]

I only have an MA, but I'd say learning statistics programs and familiarizing yourself with the core litterature are probably the two best things you can do.

[ArtVandalay]

5 hours ago, toad1 said:

Reaching out to current grad students (and beyond):

What skills were most useful when beginning your PhD program? As the admissions decisions trickle in, I can't help but think to myself: what extra stuff should I do / learn / understand prior to beginning this journey (whether it's specific math skills / programming skills / language skills / etc.)

What gave you an edge? -- or, what set you back and you wish you had known beforehand?

I'm curious about this as well, especially for those in the methods camp.

Brush up on linear algebra, real analysis, ... ? Any insights?

[guest56436]

Here's my opinion:

Sure, some people will argue that you should have language skills, programming skills (probably not all that important unless you are applying at a methods major), stats, ect. Okay, maybe these are important, maybe they are not (depends on your personal situation). I'm also not sure why you can't pick those up along the way. It takes a couple of days to learn functional latex for example; more to learn R, but it's not particularly challenging if you put in the effort either (but you'll have plenty of problem sets and term papers to practice it with anyways). There are also risks. You could spend a whole summer learning elementary Chinese and then decide you'd rather study SE Asian politics. You could also spend all year trying to learn calculus independently and figure out by the time you get to math camp that you've basically learned nothing. Hopefully the program you go to has a good quant training (hint: you shouldn't go to one that doesn't - unless you're a theorist).

You want to know, in my opinion, the bar none number one skill you should hone before and during grad school? Learn how to write academic papers. You get jobs based on publications. All your major requirements (prospectus, dissertation) are just versions of academic papers. Yet, no one teaches you how to write them. And trust me, it's a god damn process. I spent the better part of two years in my masters trying to write my first publishable political science paper and stumbling like a drunken fool through it (I eventually got there though, with a lot of hard work and guidance from my adviser). 

So how do you learn how to write an academic paper? Well..you write one. But of course, that's not that great of advice, so here's the actual advice:

- Take the best political science paper/thesis/project/dataset/whatever you have ever done and turn it into a real paper.

- Take a published paper close to your project that you pour through every word and how they frame, section, build the argument, how they support it. That means you look at how they wrote the intro and introduced the topic. You learn how they define their concepts and show the variation in the dependent variable that they are trying to explain. That means you learn how they drew from the literature - to both knock down and take pieces of it to build their own theory. Then you look at how they built their own theory. Then how they proved it - both methodologically and analytically. Then you breakdown how many words they spent on each section. Then you emulate and copy it. 

- Then revise it.

- Then send it to your advisers.

- Then revise it.

- Then you'll probably need to fix how you coded your cases/quantification/your model/your process tracing or whatever.

- Then revise it.

- Send it to anyone who will read it.

- Then revise it.

You do this (and take it seriously) and you'll achieve four things:

1) You'll learn a lot in how to write an academic paper.

2) You'll have a working paper the first day you enter a program (very nice thing to have).

3) The first one is the hardest...and it gets easier each time. I can produce a new working paper in a fraction of the time I spent on my first one. 

4) You will realize where your weaknesses lay. Was it especially challenging synthesizing the existing literature? What about building your own novel theory? What about your methods? Did it take you forever to collect and code your data? Does your writing just in general suck? Knowing your weaknesses, and accepting them, makes you a better grad student because you know what to improve and spend time on (when your time is ridiculously finite). 

Edited February 4, 2018 by Comparativist

[toad1]

@Comparativist Thank you for the thoughtful response, very helpful.

[oakeshott]

The more math, statistics, R, and LaTeX you know going in, the easier your life will be. Anyone who tells you otherwise is just lying to you.

Of course you can pick this stuff up when you get there—and for the most part I did. But that doesn't make the above statement any less true. There's no question: If I could have my last summer before grad school back, I would take some online math courses and read the Angrist and Pischke books. 

 

Edited February 4, 2018 by oakeshott

[CarefreeWritingsontheWall]

The ability to handle rejection and outright hostility without taking it personally is a skill you will need to succeed in this field (and academia more broadly, perhaps life...). As a graduate student I didn't expect everyone to love my work, but I never expected people to outright tell me it was garbage. I didn't have thick enough skin for such comments at first, and it was a struggle to find my confidence pretty early on in my program. It's important to view comments on your work as separate from comments about yourself. It's hard not to take it personally. There is something to be said about coming in with pre-existing skills in combinatorics, probability theory, and programming (be it R and/or Stata, LaTex etc), but the capacity to deal with pretty destructive (as opposed to constructive) criticism is really important. 

[toad1]

@oakeshott @CarefreeWritingsontheWall   Thank you! 

[oakeshott]

13 minutes ago, CarefreeWritingsontheWall said:

The ability to handle rejection and outright hostility without taking it personally is a skill you will need to succeed in this field (and academia more broadly, perhaps life...). As a graduate student I didn't expect everyone to love my work, but I never expected people to outright tell me it was garbage. I didn't have thick enough skin for such comments at first, and it was a struggle to find my confidence pretty early on in my program. It's important to view comments on your work as separate from comments about yourself. It's hard not to take it personally. There is something to be said about coming in with pre-existing skills in combinatorics, probability theory, and programming (be it R and/or Stata, LaTex etc), but the capacity to deal with pretty destructive (as opposed to constructive) criticism is really important. 

I couldn't agree with this more. One of the best pieces of advice I've ever read on here.

That being said, the ability to handle rejection is a difficult skill to acquire without actually going through the grind. 

[SerenityNow!]

For folks thinking of learning statistical packages prior to entering PhD programs I would highly recommend this course for learning R: https://www.coursera.org/learn/r-programming

While you can learn R on your own if this is your statistical package I think it helps to be guided through! However, not all schools like R so before you devote a lot of time to a specific language you may want to see what programs your PI or the department you plan to join generally favor. E.g. where I earned my masters everyone used SAS and R was rarely used and essentially only for graphics. 

 

 

I only have a masters and while this isn’t a skill exactly, I would recommend getting your personal life in order. I did this before my Masters and it made a huge difference in my quality of life. If you always wanted to workout regularly begin NOW, if you wanted to start stretching start TODAY, stop eating junk food, or even something as simple as beginning a new skincare routine! Once your program begins it will be really hard to develop those good habits with the pressures of coursework, research, TA’ing etc... 

Again, huge grain of salt because I only have a masters! But in my experience having those good routines and habits really improved my quality of life during my program.

[encyclopediabrown]

My two cents from the (junior) faculty side of things! I'd spend the months before you start a Ph.D. program getting out from behind your computer. Go hiking, paint, dance, read novels, read plays, and read quality non-fiction about politics --- newspapers and books about the areas of politics you're interested in. There are two reasons to do this. First, relaxing is good for you. Second, doing these things helps you be a nimble and creative thinker, and someone who knows a lot about the substance of politics. The hardest part of being a researcher is finding questions that are interesting, and being a creative person who is broadly knowledgable about the world helps.

[StrengthandHonor]

All of the above have good advice. 

If you're either in a quantitative field or have quantitative classes to make it through your first year, knowledge of statistical programming and linear + matrix algebra will make your life much easier. 

If you are weak on academic writing, it helps to brush up. Nota bene, there is a big difference between "good writing" for an undergrad, and the particular styles and conventions of academic writing for political science journals. 

Above all, take some time to go outside. I spent much of my last summer travelling with my spouse, camping, and backpacking. I also spent plenty of time indulging in intellectually stimulating reading that was only somewhat related to my work. Once you start the term, your time will soon become entirely eaten up with the various tasks of graduate school. Enjoy the break while you have it. It's also a good time to make sure that you have entrenched healthy lifestyle habits regarding sleep, nutrition, and exercise.  

[dagnabbit]

Lots of excellent advice here - I don't really have an original contribution to make, but I'll add my experience to the pile.

1. Basic knowledge of LaTeX (particularly math and BibTeX) is something you can pick up over the summer, and it proved to be really helpful for me when I started my program. Of course you can pick it up as you go, and many (most?) do, but a lot of my cohort-mates found it frustrating to have to struggle with LaTeX under the pressure of problem set deadlines.

2. If you have little to no familiarity with any statistical software/programming language, it would be beneficial to gain basic proficiency. The Coursera tutorial recommended above would be a great option for R, and I am also a big advocate of DataCamp's short courses. I had some prior knowledge of R before I started, and what was helpful wasn't so much my limited knowledge of how to conduct statistical analysis (the department wants to teach you this anyway) as much as simply being comfortable with programming terms and concepts.

3. If you don't already use a reference/citation manager (Mendeley, Zotero, etc), I would strongly recommend that you pick one now and learn how it works. Not only is it essential for keeping all of your readings organized, they also auto-update BibTeX files for each of your class/project folders. I wish I had started using Mendeley from the beginning of the first semester instead of picking it up halfway through and having to work backwards.

Edited February 6, 2018 by dagnabbit

[CandyCanes]

Is it weird to not be working from now until the start of the fall semester, and to just be taking time off and learning things independently?

[megabee]

Does anyone have a resource for online stats courses? Even better if the course is from a polisci perspective, or specifically for first years. I've been looking around but so far the only one I've found is this one:

https://ocw.mit.edu/courses/political-science/17-872-quantitative-research-in-political-science-and-public-policy-spring-2004/

[dagnabbit]

5 hours ago, megabee said:

Does anyone have a resource for online stats courses? Even better if the course is from a polisci perspective, or specifically for first years. I've been looking around but so far the only one I've found is this one:

https://ocw.mit.edu/courses/political-science/17-872-quantitative-research-in-political-science-and-public-policy-spring-2004/

A great way to find these is to look at  methods ABDs' websites; many of these people have taught the "math camp/refresher/etc" and TA-ed for the stats sequence, and often have links to the course materials. For example, here's Harvard's most recent math "prefresher."

David Seigel at Duke also has some really great videos on math for social scientists.