<p>I understand what you’re saying. But from number 3-16 to #35 is just a huge gap. Clearly, one of them must be wrong and unreliable. Princeton has also never been ranked in the top 5 for CS on USNews. One of them must be unreliable.</p>
<p>And which do you think is unreliable? :rolleyes:</p>
<p>Phantasmagoric, you raise lots of good points, and I am going to raise one which you might very well already be aware of. </p>
<p>Which is the following: if looking at ranges for rankings captures fluidity, and information can get lost upon making rankings simpler, it is also true that, while numbers do not lie, they only tell you what they are, and ultimately that is itself limited.</p>
<p>For instance, what is ‘research caliber’ - is it measured by quantity of important publications? I think the only source I would take seriously on a subject like this is the word of a professor at a well regarded school, well regarded in the subfield in question. </p>
<p>That is, if one is so interested in knowing which school should be regarded highly in research, it is exceedingly important to know what the research is that is being spoken of, what kinds of work is considered impressive, and why.</p>
<p>As an undergrad, measures like what courses, seminars, and other opportunities are offered becomes most important. And of course strong faculty to expose you to a taste of top research if so inclined. </p>
<p>As a grad student, I think rankings of any kind tend to be misleading, and the word of faculty who are true experts is the reliable source of choice.</p>
<p>mathboy98, I agree that the numbers are only so important as what they measure and, further, how you interpret those numbers. You can deem something important, like research, rank it according to those objective numbers, and still have a completely wrong ranking. But if the methodology is reasonable to you, you can go about interpreting the numbers as you see fit. So while the research rankings, for example, can’t capture what kinds of research are going on (most of the top 10 CS schools are useless to me, because they don’t do research in my area), there is some general metric, say citation impact, that can be used to compare departments. The result is only useful to the extent that you can say the top X departments tend to be the strongest, etc. (and the precise order doesn’t matter, because the metrics can’t give you that level of precision), but it speaks nothing to what subfields the department has research in and so on. Which is the whole point: simply to benchmark departments against one another. It’s almost strange to hear that the NRC rankings don’t cater to people on sites like this and instead are supposed to be useful to the departments they rank. Pfft, whatever that means. :p</p>
<p>Of course, this is all immaterial to undergrads, but I think it’s useful to use grad rankings as a proxy for undergrad quality in a department. At any rate, I don’t put as much stock in rankings as my previous posts in this thread would suggest. Rather, no ranking is right, or at best the ranking is not very wrong. But to bring my point into perspective, if I had to put stock in one, it would definitely not be the US News grad rankings. The NRC rankings often coincide well with my own knowledge, much of which I’ve gained from advisors: that Princeton is definitely top 5 in CS, etc. In my grad school search, I used the rankings broadly as a list of schools to look into (and not just top 20–I painstakingly looked at some 50 departments). Advisors’ advice was useful, definitely, but it was limited; one recommended I not apply to a certain few schools, but instead to a few others he suggested, but in the end, I didn’t take his advice, because the schools he said not to apply to appealed to me in their research. Not that the rankings were more useful on the whole, but they did point a few schools out to me that I decided I liked a lot (purely in terms of their research), even if my advisors didn’t.</p>
<p>Plus, if I listened to them, I would’ve had no “safe” options. ;)</p>
<p>Back to the topic of this thread, UCB v. Princeton ChE</p>
<p>Class of 2010 Undergraduate ChE’s
36 - Princeton
83 - UC Berkeley </p>
<p>ChE Faculty
22 - Princeton
25 - UC Berkeley</p>
<p>don’t know about UC Berkeley, but at Princeton ALL faculty are required to teach undergraduate school</p>
<p>
</p>
<p>I guess I tended to mention my interests quite a bit when I talked to them about this kind of thing. Other than that, I actually looked up the faculty webpages to see what they are working on. </p>
<p>I am sure we are on the same page and you see what I meant, so simply for clarification…</p>
<p>I think what I wanted to bring out is that I find people use rankings a lot more than not to think about things, and that really a lot of the well known schools are way up there to the point where in my experience, individual taste and where one actually gets admission become most important. </p>
<p>My own experience with mathematics programs is that people seem to regard a certain class of schools (Berkeley, Stanford, Harvard, MIT, Princeton) in the top 5, and no matter how one slices it, I dislike the effect - there is such an incredible collection of faculty,offerings, etc in various areas that the rankings of US News miss, and looking up NRC still leads me to believe it is tough to make an actual conclusion based on their info, even if I agree it is objective, informative, and interesting in and of itself.</p>
<p>I hate to put it this way, but for grad, I think one really may have no tools aside from checking the websites of the actual faculty and visiting the school, at least for something like mathematics. One can use rankings to isolate some sizable double digit number of schools to consider, but I find there little to be said at the level of top 5 - I guess what I mean is top 5 is too small, unless multiple schools occupy each of the 5 positions on whatever constructed rankings. Even an individually constructed top 5 is only useful for those very committed to a certain ultra specific subclass of research.</p>
<p>Now maybe undergrad is a slightly different story, since people are more on the same page in undergrad.</p>
<p>mathboy98,
</p>
<p>I did too. My advisors were aware of my interests (after all, we worked together on research), but they didn’t think that some of the schools I’d chosen were worth it when I could be applying to a few other schools they recommended. This actually stemmed more from a difference in philosophy over research. They tended to emphasize machine learning faculty more than I did, whereas I emphasized pure research in my area (though ML is useful and was also a consideration). They didn’t think that was as important, so long as there was at least one professor who had very similar interests as me, but I felt that having a few professors with very similar interests was more important (rather than a bunch of additional professors who mainly did work in ML but would also be able to advise me). This is a specific case, but the takeaway point is that the grad rankings helped me to find these; had I relied solely on their advice, I might not have applied to a few of the schools I did.</p>
<p>On a case-by-case basis, I think you’re right that students have no real tools to help them other than looking at professors’ websites. But in terms of saying, “What schools are ‘the best’?” in general–for whatever reason or motive you might have–it’s possible to identify a few schools. Top 5 is an arbitrary number; some fields it’ll be 4, or 6, or 3, but 5 just happened to be the number for CS. These are the schools that seem to do really really well, and the next set of schools might have a noticeable difference in quality. Sometimes, there isn’t such a difference so identifying a set of very top schools is too difficult/arbitrary (it’s always arbitrary, but in some cases it’s more so). But like you said, in the end it’s useless to list these, because quality is very personal and specific to a student’s interests, so it only makes sense to figure out which professors are a match. Of course, the rankings are helpful in identifying a bunch of schools to explore.</p>
<p>Out of curiosity, are you a grad student at Berkeley now?</p>
<p>meanwhile, it appears that NYU has made a move in the Mathematics Rankings</p>
<p>2010 NRC Mathematics Graduate Department Rankings
- Princeton
- Harvard
- NYU
- UC Berkeley
- Stanford
- MIT
- Yale</p>
<p>i would never be able to turn down princeton, especially considering that it’s supposedly undergrad focused and well berkeley is less so</p>
<p>
</p>
<p>Right, when it comes to making decisions, this is what I think. I think what I figured out was that there are so many superstar professors scattered around in the Ivy Leagues, Caltech, some top state schools, that a well defined top few is really hard to come up with. Added to this is that international students tend to be very strong, and end up all over the place, so while Berkeley for instance is placed ahead of many Ivy Leagues for math, in many general senses, the students at schools in something like 15 or so schools can be really strong. And the faculty can be insane in even more schools.</p>
<p>As for being a grad student at Berkeley - well maybe in the future! </p>
<p>As for your experience with professors’ advice, I do find that it depends on the professor. Unfortunately, some professors seem to be under the impression that you should just go to a strong school with a general idea of what you want to do, with some specific plans, and then work hard - which leads to advice less suitable for those of us with subtler research tastes. But I luckily encountered some especially understanding ones who say ‘ok, if you really have to study exactly so and so aspect of number theory, then forget so and so top school which is otherwise exceptional and strong in some related fields.’ That is advice I think many professors would be hesitant to make and would instead say ‘eh, apply to top programs that have stuff that looks cool, and that’s important, since you will change your mind anyway in grad school!’</p>
<p>
</p>
<p>I bet people have made this claim about Harvard too (not undergrad focused). I don’t buy it without specifics. What do people actually mean by undergrad focus? I think in a small school like Caltech, people naturally know each other more, because they take a common core, for example.</p>
<p>
</p>
<p>One perhaps good measure in comparing like schools is grad/undergrad ratio. At Caltech (and Columbia with a core) it is over 1, i.e., there are more grad students on campus than undergrads. At Princeton and Dartmouth it is below 1.</p>
<p>
Very good point.</p>
<p>yeah what they said with regard to undergrad focused. also, princeton, at 7k students in total, is pretty small for a university anyway</p>
<p>*One perhaps good measure in comparing like schools is grad/undergrad ratio
[/quote]
</p>
<p>What does this ratio mean? As in, if I were hypothetically a student trying to go to a certain school, how would I best use such data to actually help decide where to go?</p>
<p>
</p>
<p>mathboy, there have been significant studies done on the grad/undergrad ratio. The results all lead to conclusions that a ratio of 0.494636 is the optimum for universities.</p>
<p>A high ratio means a major RESEARCH Uni. And research has little/nothing to do with undergrad teaching. </p>
<p>A high ratio may mean that undergrads are an after-thought. (At one time, not to long ago, there was some serious discussion at UChicago about dropping its undergrad program so it could focus on research, research and more research.) A high ratio may mean lotsa grad students are vying for faculty time and resources, and the lowly undergrads are last in line. A high ratio may mean that undergraduate teaching is the lowest of low priorities.</p>
<p>Just things to query when you visit campuses.</p>
<p>
Haha…to that level of significant digits too…:)</p>
<p>You could also survey university presidents and administrators…USNWR asks to name universities with a strong commitment to undergrad teaching. Princeton and Berkeley were two universities mentioned. ;)</p>
<p>[Best</a> Undergraduate Teaching | Rankings | Top National Universities | US News](<a href=“http://colleges.usnews.rankingsandreviews.com/best-colleges/rankings/national-universities/undergraduate-teaching]Best”>http://colleges.usnews.rankingsandreviews.com/best-colleges/rankings/national-universities/undergraduate-teaching)</p>
<p>
</p>
<p>How would they be last in line? A grad student tends to get an hour or so with a single faculty he is working with weekly. Some professors reserve a day a week to meet with PhD students.</p>
<p>Undergraduates tend to benefit from office hours - that is the primary time to chat with faculty. Even if not enrolled in a class, undergrads can usually just walk in and talk to the friendlier faculty. Yet when it comes to less friendly faculty, no amount of coaxing gets them to reach out.</p>
<p>I think faculty to student ratio in a university may be a more useful measure (and sometimes one must restrict to the given major, or even subfield in the major to get the really useful data) - the roles of undergrads and grads tend to be separate, and as long as a course is not crowded, the professor usually has plenty of time to chat during office
hours.</p>
<p>
</p>
<p>A comment on this - there is a high correlation between major research university, and interesting departmental activity in my experience researching schools. Caltech is a small school with not many students in the math dept, yet it offers surprisingly lots of stuff to go to and learn / benefit from.</p>
<p>Now the quality of teaching is another story, but an undergraduate can benefit from availability of resources a lot, since these are the years to survey stuff and when it is important to be exposed to lots. In fact, in the graduate years it can be fine to attend a small school that does very well in a narrow area of interest and has not as much going on outside of it.</p>