Advice from an actual PhD (redux 2)

[Helix]

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.

Do you think the NRC survey-based measure captures the same thing as the "perception from within" for the USNWR rankings? I've been struck by what look like meaningfully sizable differences between the regression-based and survey-based NRC rankings. And I get that rankings are meaningless on the margin, and I get that the NRC rankings are "outdated" because the data is several years old, but in general I wonder if the perception of expertise > whatever "expertise" the regression-based measure supposedly captures.

[Jwnich1]

Do you think the NRC survey-based measure captures the same thing as the "perception from within" for the USNWR rankings? I've been struck by what look like meaningfully sizable differences between the regression-based and survey-based NRC rankings. And I get that rankings are meaningless on the margin, and I get that the NRC rankings are "outdated" because the data is several years old, but in general I wonder if the perception of expertise > whatever "expertise" the regression-based measure supposedly captures.

In the end it would seem that S is more important than R for rankings for the purposes you mentioned. In the end does actual quality "matter" as much for placement purposes as quality as perceived by a search committee? (assuming that the skills imparted by any ~top 30 program will be roughly the same by subfield / interest)

[kaykaykay]

R rankings probably capture better how people percieve the programs .

The S method calculates weights on program characteristics based on what people reported as important in a survey.

R calculates the weights based on how these people actually ranked the programs - what is actually important for them.

See more here:http://themonkeycage.org/blog/2010/09/28/the_nrc_rankings_of_political/

Note that the R ranlings and USNWR rankings are more closely correlated than S rankings and USNWR rankings.

Overall, however, NRC rankings have been attacked for many good reasons.

[Jwnich1]

Imagine for instance a hypothetical candidate trying to decide between Rochester (top 10 - 20 in S rankings, 20 - 30 in R rankings) vs UW - Madison which is almost exactly vice versa.

Aside from all of the other factors, which one is "more highly ranked". I'd say it's a toss up. thoughts?

Edit: I picked these schools because they are illustrative of a situation where R and S diverge significantly. (and USNWR throws it all off by ranking them almost exactly the same). I'm not trolling or trying to start a "this school is better" war.

Edited February 18, 2012 by Jwnich1

[RWBG]

R rankings probably capture better how people percieve the programs .

The S method calculates weights on program characteristics based on what people reported as important in a survey.

R calculates the weights based on how these people actually ranked the programs - what is actually important for them.

See more here:http://themonkeycage.org/blog/2010/09/28/the_nrc_rankings_of_political/

Note that the R ranlings and USNWR rankings are more closely correlated than S rankings and USNWR rankings.

Overall, however, NRC rankings have been attacked for many good reasons.

Everything said here is right. Just use the USNWR rankings to get a sense of the broad distinctions, and go from there by looking at placement data, etc. Moreover, Rochester is probably one of the least useful comparisons given that they are so specialized in formal and statistical training; their overall ranking is lower as a result of their specialization, but they place better than many schools ranked higher (or equivalent) to them. However, Rochester would probably be a bad choice for many people not interested in that approach to political science.

[Jwnich1]

Everything said here is right. Just use the USNWR rankings to get a sense of the broad distinctions, and go from there by looking at placement data, etc. Moreover, Rochester is probably one of the least useful comparisons given that they are so specialized in formal and statistical training; their overall ranking is lower as a result of their specialization, but they place better than many schools ranked higher (or equivalent) to them. However, Rochester would probably be a bad choice for many people not interested in that approach to political science.

RWBG - this is precisely my point (admittedly I made it badly). None of the ranking are really accurate when there is a specialization or focus beyond broad subfield. In the end is placement not the most important thing for us in the end? (not the only of course!)

[brent09]

RWBG - this is precisely my point (admittedly I made it badly). None of the ranking are really accurate when there is a specialization or focus beyond broad subfield. In the end is placement not the most important thing for us in the end? (not the only of course!)

This is why having a close relationship with advisors in your field is important. Certain schools have reputations that exceed their overall ranking -- like Rochester, NYU and WUSTL for methods/formal or UVA, BC and ND in theory. Similarly, a fairly high ranked program like MIT would likely struggle to place a Straussian theorist (assuming that they admitted one). No single ranking methodology really captures those dynamics. The best bet, then, is to:

1. Discuss your interests with your advisors and figure out which programs have the best reputations in that area.

2. Look at program placement records, find the students they've placed and see where graduates in your field end up.

That's the advice I've gotten, and it seems solid. It takes a lot of work tracking down people from all the programs you're considering, looking at their CVs and all that. But hopefully it's more fruitful than relying on simple USNWR, NRC or Hix rankings.

[RWBG]

RWBG - this is precisely my point (admittedly I made it badly). None of the ranking are really accurate when there is a specialization or focus beyond broad subfield. In the end is placement not the most important thing for us in the end? (not the only of course!)

Sure, but very few departments are as specialized as Rochester. You can still get the broad contours from the USNWR, maybe crossreferencing with this:

http://polisci.fsu.e...nt_Rankings.pdf

Ultimately, I think the main points are (1) consider carefully the costs and benefits if you're going to a school outside the top 25 (2) If you have offers from schools from different tiers (i.e. top 20 v. top 10) then you should seriously consider whether any perceived advantages of the lower ranked school will outweigh the reduced competitiveness on the job market. Beyond that, the rankings aren't going to tell you anything informative about the fine differences (7th versus 9th), and schools like Rochester are always going to be difficult to evaluate on the basis of those rankings. And, of course, you have to consider your school's strength in your subfield. I don't think the Realist's point is to have the rankings make the decision for you, but merely to give a bit of a better guide as to the kinds of considerations you should be making when deciding between schools, with the important note that "pedigree" (an awful word) matters.

Edit: Re Brent: Sure, which is great if you have a bunch of advisors very familiar with US schools in your subfield. Not everyone has those, especially students from overseas, students from tiny schools with a small faculty, etc. Also, you should always consider that each advisor is only one person; in my experience talking with individual people, I've gotten really good advice about things I didn't know about, but I also think (mind you, this is based on my own evaluation) that some of the advice I've received in the past was skewed by personal biases and casual perceptions that can be idiosyncratic. My advice: do everything. Consult with advisors, read professors' CVs, use rankings, read placement records, go to campus visits, talk to graduate students, talk to students at different schools who won't be biased, read the CVs of the students of potential advisors, check to see which faculty are new and which have moved and what effects that might have on placement, check who's publishing in the journals you like, check who's participating in specialized conferences in your subfield, and try to find recent students who have been placed in your subfield to get a sense of how they did. Even then, there's no certainty your choice will be the best one, but relying on any individual measure is likely to be foolhardy.

Edit 2: Or maybe something slightly less obsessive but still comprehensive...

Edited February 18, 2012 by RWBG

[Jwnich1]

Sure, but very few departments are as specialized as Rochester. You can still get the broad contours from the USNWR, maybe crossreferencing with this:

http://polisci.fsu.edu/news/Placement_Rankings.pdf

Ultimately, I think the main points are (1) consider carefully the costs and benefits if you're going to a school outside the top 25 (2) If you have offers from schools from different tiers (i.e. top 20 v. top 10) then you should seriously consider whether any perceived advantages of the lower ranked school will outweigh the reduced competitiveness on the job market. Beyond that, the rankings aren't going to tell you anything informative about the fine differences (7th versus 9th), and schools like Rochester are always going to be difficult to evaluate on the basis of those rankings. And, of course, you have to consider your school's strength in your subfield. I don't think the Realist's point is to have the rankings make the decision for you, but merely to give a bit of a better guide as to the kinds of considerations you should be making when deciding between schools, with the important note that "pedigree" (an awful word) matters.

Thanks for the useful advice! I have such a weird mix of admits and rejects that I'm having more and more trouble deciding as I keep doing research. What seemed like a clear cut choice no longer is. Unfortunately I've been out of school for a few years, so I no longer have an advisor, just the interwebs.

Best of luck to all!

[RWBG]

Also, it should be emphasized that your school isn't everything! I think the school you go to matters, but it's not like it fully determines what level of success you'll have. Even if your choice of school isn't precisely optimal, if you're at a strong school that's a decent fit for you I suspect what you do in grad school is going to matter a lot more to your eventual success on the market.

[Jwnich1]

Also, it should be emphasized that your school isn't everything! I think the school you go to matters, but it's not like it fully determines what level of success you'll have. Even if your choice of school isn't precisely optimal, if you're at a strong school that's a decent fit for you I suspect what you do in grad school is going to matter a lot more to your eventual success on the market.

Agreed. I have a feeling that the people who gravitate to gradcafe will do great things. I'm just amazed at the shift in rankings prestige. (and associated confusion). I've even heard "The Fall of Chicago" referenced.

Anyway my musings aside. Thank you for the comforting words.