Advice from an actual PhD (redux 2)

[The Realist]

I am a tenured associate prof in political science at a large state university. Around this time in the application season, I can't help but think about all the things that I wish that I had known before entering my PhD program. I posted this several years ago under the screen name "realist" when I first learned about this forum from a senior applying to PhD programs. Two years ago I posted it again. Now it's time for the third generation.

I've made a couple small changes from the original version but this is basically the same as what I wrote the two times before. While some of this may be hard to read, I offer it as-is, with only the thought that more knowledge is better than less knowledge.

I. Choosing Graduate School

Your graduate school choice is probably the most important choice that you'll make in your career. Do not take this lightly. There are many reasons, but they boil down to some uncomfortable truths.

The most important one is that only the best departments (say, the top 25) can reliably place students in academic jobs. And even among these "top" departments, less than half can expect to find themselves employed in a tenure-track position within 8 years of matriculation. While there are thousands of colleges in the United States, there are many many many thousand more political science PhDs. 5-7 years is a very long time to spend in a low-paying job (which is what graduate school is) only to realize that you have very little chance for promotion. Is it fair that this is the case? No. Are there very smart graduate students that are not at top departments? Absolutely, there are literally thousands of them. But this is how the world works. And you have no chance to change it from "the inside" unless you are already at a top department.

You should also be aware that advisers are fickle beings. Especially outside of the top institutions, they are busy and pressed for time, and they cannot offer you the type of guidance and support that you may believe that you are going to get. I had a very close relationship with a very influential adviser, and saw him for about 10 minutes once every two or three weeks. This is the norm. Do not assume or expect that you will have a different experience (although there is a chance that you will). Moreover, good scholars are often terrible advisers. I think that one of the worst aspects of our profession is that at middle-range departments, top scholars often will not even acknowledge graduate students.

You should also be aware that graduate school is an unequal partnership between students, who receive very little and give very much, and faculty, who have many other things to do but rely on students to do things that are in the university's best interests. Graduate students are (1) essentially powerless and (2) extremely cheap labor. Universities have an incentive to keep a lot of graduate students around to fill instructor slots and TAships. This means that they will keep on a lot of graduate students who will never have a chance at a tenure-ladder job. This is a pathological system of incentives, and I find it repugnant, but this is the reality.

So what sort of advice does this lead me to give? First off, above and beyond almost anything, you need to go to the best possible graduate school. It doesn't matter if you don't like Ann Arbor as much as Athens or Austin, graduate school matters tremendously for your future ability to get a job. It's not that hiring committees care exclusively about pedigree, although that does indeed matter to many people. Rather, the reason why you need to go to the best possible graduate program is because you need to surround yourselves with the smartest and most motivated students possible, because you will do most of your learning from them. You also need to surround yourself with the faculty who have been judged by the discipline as having the best reputations and connections. Those things are highly, highly correlated with the "rank" of the graduate program.

As a corollary, you need to think long and hard about graduate school if you do not have the opportunity to go to a top one. You should understand that even if you do, you may not have a good chance of landing a tenure track job. The ones available to you, moreover, will likely be at "directional institutions" (think Northern X State) or small, low-ranked liberal arts colleges in the middle of nowhere. Even there, you will be competing with Harvard and Berkeley PhDs for a job. It's hard. It's not as hard as English or History, but nevertheless it's really hard. You should know this and plan accordingly.

The academic job market has gotten much harder in the four years since I first wrote this. There are thousands of students right now chasing a couple hundred jobs, and every year it gets worse because most people who strike out in one year go back on the job market the next year. Do not assume that the academic job market will get easier in 5-7 years, when you are going onto the job market. First, there will still be a substantial backlog of unplaced PhDs. Second, trends in academia are leading to more adjunct and lecturer positions and fewer tenure-track positions in all but the very best schools (and it's starting to happen there too). The number of tenure track assistant professor positions in political science listed on APSA's eJobs site has dropped from around 700 per year in the early 2000s to about 450 in 2010. I would not still be in academia if I didn't have a tenure-track job.

Let's say you don't want to go be professor. Maybe you want to work in a think tank or a political consultancy. OK, fair enough: but in this case, I would recommend against getting a PhD in political science. There is little that you can gain from a PhD in political science that a think tank will find attractive that you cannot also have gotten from a good MPA/MPP/etc. program. Outside of academia, the PhD has little value-added over most professional masters degrees. Given the opportunity cost, the only people who should get PhDs in political science are people who have a passion for college teaching, or those who have a passion for academic research and who are willing to settle for college teaching if the academic research thing doesn't work out.

Do not choose graduate school based on one individual who you "want to work with." Instead, you should choose the best program (by subfield) that you can. Why? Let's say that you identify one faculty member whose research interests match yours perfectly. For this to be the person upon whom you rely for your entire PhD course of study, it must be the case that (1) your research interests don't change (which is rare), (2) that your potential adviser is a nice and approachable person (which is about a 50-50 shot to be honest), (3) that your own research is interesting to that potential adviser (which you should not assume, regardless of what is said on recruitment weekend), and (4) that that adviser doesn't leave (which is common, especially for productive faculty at top-50-ish departments). If you chose a program based on that individual and any of these don't work out, you're in trouble. If you've chosen the best program, you'll be OK because there are other options; if you've banked on one faculty member, you're out of luck.

You should be flattered by faculty who are nice and approachable during recruitment weekend. But recruitment weekend is not like the other 51 weekends a year. Remember, faculty are approachable during recruitment because you provide them with an unlimited supply of discount labor. They have their own worries and incentives, and these rarely align with yours.

Likewise, funding matters. You should not go to graduate school unless you have full fellowships (teaching or otherwise) for five years and a stipend large enough to live on. Without these, graduate school is a long and expensive process with little reward. There is a constant demand for doctors, so doctors can pay for medical school and still come out ahead. $200,000 in debt and only qualified for a very low paying job is a terrible situation that many PhDs find themselves in.

It is tempting to think that a potential adviser's kind words mean that you are special. You are special, but so are many many others. Wherever you are, you will likely not even be the smartest or most successful member of your cohort. Do not fool yourself into thinking that you are the one who will buck the trends that I have described. It's just not likely.

Finally, I have made a big point about top 25 schools. We all know that Stanford is and Purdue isn't, but what's the definitive list? Simply put, if you have to ask, your school is not in the top 25. And of course subfield matters more than overall ranking. Emory is not a top-25 theory department so think long and hard about going there for theory. JHU is not a top-25 American politics department but it's a different story altogether for political theory. If you need to convince yourself that your program is a top-25 program, it's almost certainly not.

II. Your Career

If you decide to go to graduate school, congratulations. I mean this sincerely. You are embarking on the most intellectually rewarding period of your life. Of course, intellectually and financially rewarding are not the same, but given this choice, here are some tipcs. The best political scientists are the following five things: smart, creative, diligent, honest, and nice. Smart is obvious. The rest are not.

The best political scientists are creative. They look at old problems in new ways, or they find new problems to look at. A good way to land a middling job (or no job) is to find a marginal improvement on an existing estimator, or take lessons from Paraguay and apply them to Uruguay. The best political scientists ask new questions, and they find new things to estimate.

The best political scientists are diligent. They think about problems for years and years, they rewrite their draft papers repeatedly, they collect giant datasets from scratch, and they go into the field, learn the language, and stay there until they have learned something. There are no quick research trips, there are no obvious philosophical points, and there are no downloadable datasets left to analyze for easy and quick results.

The best political scientists are honest. There are many points at which you might fudge your work: creating a new dataset fromscratch, during fieldwork, in writing up your results. You will be astounded at how frequent this is in our profession. Don't do it, for it always hurts you in the end. Being wrong and honest about it is OK. Being wrong and hiding it never works.

Finally, the best political scientists are nice. It is tempting to be prickly to make yourself seem smart or to protect your ego. But the same person you criticize today might be in a position to give you a job tomorrow. As they say, make your words soft and sweet, for you never know when you may have to eat them.

************

I hope this helps you all. I wish you the very best of luck with your careers.

[orangeMan]

First off, above and beyond almost anything, you need to go to the best possible graduate school. It doesn't matter if you don't like Ann Arbor as much as Athens or Austin, graduate school matters tremendously for your future ability to get a job. It's not that hiring committees care exclusively about pedigree, although that does indeed matter to many people. Rather, the reason why you need to go to the best possible graduate program is because you need to surround yourselves with the smartest and most motivated students possible, because you will do most of your learning from them. You also need to surround yourself with the faculty who have been judged by the discipline as having the best reputations and connections. Those things are highly, highly correlated with the "rank" of the graduate program.

I have been educated in the UK. This is so true in the UK as well. 5 universities, Cambridge, Imperial, London School of Economics, Oxford and University College London dominate the UK academia.

[RWBG]

Thanks, The Realist, your advice has consistently been amongst the most insightful on this forum. I hate to be the guy that pesters you with more questions, but given the quality of your advice in the past, I feel I would be remiss if I didn't. Do you have any insights into what schools are considered strongest on the job market for IPE? I have been fortunate enough to be admitted to Wisconsin, and there is some chance I may need to decide between Wisconsin and some others. Placement data is only so helpful in getting a sense of how schools are perceived by subfield, especially for a department like Wisconsin which has had a lot of faculty move around recently, given that placement is a lagging indicator. Anyways, even if you don't respond to this, thanks again for all the advice you've given now and in the past; I have taken it strongly into account when making grad school related decisions.

[blaspheming]

Withdrawing all of my applications in the morning. Thank you.

[wuerzburg]

Withdrawing all of my applications in the morning. Thank you.

I assume this is a joke??

[puddle]

Wuerzburg! One of my favourite German cities. Those are some fine wine cellars you guys hide in the basement of the palace.

[blaspheming]

I assume this is a joke??

Yes I was joking still a terrifying post, though.

[deleteduser0333]

Yes I was joking still a terrifying post, though.

I was seriously thinking about giving up...somehow...

[BrazilianGuy]

@The Realist:

Excellent post!

[whirlibird]

This is all really wonderful advice. Thanks, Realist, for not only emphasizing being honest and being nice, but for actually walking the talk with this post. That we should be fortunate enough to be in your position one day, and that we should follow your example and shoot straight with the next generation of hopefuls.

[leecrimson]

Good Post. I do take some issue with the notion that people aiming for jobs outside of academia would be better off without a Phd. I have my MPP and the truth is that you will spend many years doing "doctoral" level work without the salary bump that comes with a Phd. I believe this is particularly true if you intend to work primarily in research and not on policy implementation (i.e. working in a compliance office of a government agency).

[catchermiscount]

I first read this the second time I applied (three years ago), and it gets truer and truer every year.

[blackcoffee64]

This is all really wonderful advice. Thanks, Realist, for not only emphasizing being honest and being nice, but for actually walking the talk with this post. That we should be fortunate enough to be in your position one day, and that we should follow your example and shoot straight with the next generation of hopefuls.

Well said

[The Realist]

Thanks, The Realist, your advice has consistently been amongst the most insightful on this forum. I hate to be the guy that pesters you with more questions, but given the quality of your advice in the past, I feel I would be remiss if I didn't. Do you have any insights into what schools are considered strongest on the job market for IPE? I have been fortunate enough to be admitted to Wisconsin, and there is some chance I may need to decide between Wisconsin and some others. Placement data is only so helpful in getting a sense of how schools are perceived by subfield, especially for a department like Wisconsin which has had a lot of faculty move around recently, given that placement is a lagging indicator. Anyways, even if you don't respond to this, thanks again for all the advice you've given now and in the past; I have taken it strongly into account when making grad school related decisions.

It's hard to give specific advice on this question. Lots of schools are good at IPE, but placement varies pretty widely among even the best programs. I would still recommend choosing a program based on a subfield (like IR or PE) rather than sub-subfield like IPE, but that's splitting some very fine hairs.

A choice between UW-M and a school that's excellent but not notably strong in IPE, like Chicago, would be tough. A choice between UW-M and, say, Princeton should be easier to make.

[The Realist]

I first read this the second time I applied (three years ago), and it gets truer and truer every year.

In the spirit of helping today's applicants, it might be worth contributing a post of your own on what you've learned since starting the PhD program...from a current grad's perspective, what do you wish that you'd known when getting into the process? Of course, you should only do this if you are making solid progress on the diss.

[yinche]

Hello Realist,

first of all thank you so much for wonderful advice. I was given admission to Penn State with full funding for 5 years. My interests are influence of domestic politics on foreign policy and international conflict behavior. Do you think Penn State is a worthwhile school to attend?

[catchermiscount]

Hello Realist,

first of all thank you so much for wonderful advice. I was given admission to Penn State with full funding for 5 years. My interests are influence of domestic politics on foreign policy and international conflict behavior. Do you think Penn State is a worthwhile school to attend?

Let's not get into the practice of bugging The Realist too much. But, I will attempt to follow his/her advice on this one (that is, helping today's applicants) by putting my two cents in (I am interested in conflict broadly, but take this for what it's worth). If you're interested in domestic-IR, Penn State is a very good place. Glenn Palmer is a nice, smart guy that appears to be a good advisor (for example, Phil Arena got placed at SUNY-Buffalo, has been productive, and has co-authored with Palmer). There are some other examples of good IR placements. Generally, good technical training. Penn State is certainly worth considering hard.

Edited February 2, 2012 by coachrjc

[yinche]

Thank you coachrjc. Once I become a grad student too, I'll help out tomorrow's applicants.

[polisciphd]

I echo Penn State as a good place to do quant IR/Conflict, they have come a long way in a few short years. I can also give advice for anyone going through the process, I'm a fourth year IR at a Big 10 school, working on dissertation (posted on this forum years ago under a different screen name tidefan)

[studentized]

Also consider UNC, Florida State, and TAMU for quantitative IR/Conflict.

[yinche]

Thanks, studentized. I also applied to UNC and FSU, but not TAMU. What is your main area of interest?

Edited February 8, 2012 by yinche

[dienekes]

Finally, I have made a big point about top 25 schools. We all know that Stanford is and Purdue isn't, but what's the definitive list? Simply put, if you have to ask, your school is not in the top 25. And of course subfield matters more than overall ranking. Emory is not a top-25 theory department so think long and hard about going there for theory. JHU is not a top-25 American politics department but it's a different story altogether for political theory. If you need to convince yourself that your program is a top-25 program, it's almost certainly not.

Sorry for bumping an oldish topic, but (and maybe this is just my ignorance) I'm genuinely confused as to what this means from a practical advice standpoint. There's no definitive list, but it's imperative that I make precise decisions about program quality anyway? All of the different ranking systems I've looked at place different departments all over the map--I've seen disparities from some schools being listed around 15 in one ranking and 45 in another. And that's a "global" ranking, not subfield specific. Granted, the top five or so schools stay fairly static across rankings, but--particularly if you're applying to public state universities--it seems like the rankings vary wildly. A professor at my undergrad institution who wrote me a LoR has been willing to give his best guess at program quality, but, yunno, that's one professor's subjective take on an already subjective ranking system. And it doesn't account for a fair amount of fluidity in program quality--new faculty, faculty retirements, faculty who take jobs elsewhere, researchers who produce important work, etc etc. I can understand this advice if you're making relative decisions and there's a clear (and large) disparity in quality (say, between a school that's in the 15-30 v. a 35-50 school), but it gets into weird hairsplitting territory between schools in the 10-30 range.

I guess I'm just saying that trying to make a measured, rational decision based on imperfect information is a tall order, but, then again, the whole endeavor of pursuing a career in academia seems to require a rather lot of those decisions.

Anyone want to weigh in on that?

[kaykaykay]

Frankly I think the Realist was very clear.These rankings are subjective but if a school jumps in and out of the 25, maybe it is not considered to be in the top 25. But of course you and you can only decide whether you want to take his/ her advice.

[dienekes]

Frankly I think the Realist was very clear.These rankings are subjective but if a school jumps in and out of the 25, maybe it is not considered to be in the top 25. But of course you and you can only decide whether you want to take his/ her advice.

I wasn't questioning his/her clarity, I was questioning whether it's actually possible to act upon it, given what I see to be a high level of variability between rankings, and their general all around dubious dubiousness. Ohio State: 4th Best Political Science Program in the World / 17th Best in the Country!

Then again, maybe I just think the whole idea of a hard cutoff point is basically ridiculous in a subjective ranking system. If I go to the 26th best school in the country I'll be eating ramen for the rest of my life, but the 24th and I will be revered and respected throughout the field? And how am I supposed to tell the 26th from the 24th anyway?

Edited February 17, 2012 by dienekes

[catchermiscount]

It's possible to act on it.

First, don't use the Hix rankings at all. Those are about faculty productivity, not graduate training. All they'll do is induce even more variability when it's not needed.

Any given ranking is an attempt to tap into some unobserved, underlying "PhD program quality" dimension. They're all imperfect, and they're not even all that correlated anymore. I think TheRealist's point (and I hate speaking for him/her) is that "you know it when you see it."

And let's be honest: while top 25 is arbitrary (24th versus 26th isn't all that much of a difference, indeed), departments (and their deans) worry a helluva lot about top 25 status, or top 20 status, or whatever. As an example, see Emory's site, which proudly trumpets their status as a top 20 political science department based on one placement measure. Emory is an excellent department that has undergone a lot of improvement lately, but I don't think many people would agree that it's a top 20 PhD program just yet.

If you must use a ranking, the US News one stinks but is likely the best we have for these purposes in that it offers a sense of perception from within. Sure, it's cronyistic and all, and sure, taking the mean of an ordinal variable as the only statistic is dumb, but you get the idea. And I would probably guess that you could safely call the top 20 schools on that ranking "top 25 departments" for most individualized purposes (though again political philosophy brings its own problems due to departments like Johns Hopkins, Virginia, Notre Dame, and so on).

So yeah, 25 might be an arbitrary cutoff point, but the general logic is just fine.