For those aiming to teach/research at a university, does undergrad matter much?

<p>I aspire to earn a PhD in mathematics at some point to teach/research at a university level. Does it matter where you go to college for undergrad? I've been sent mail from top universities (Chicago, Northwestern, etc.), but I'm not sure if I would like to go to a very prestigious university, or the University of Iowa, for economic reasons. Would going to a state university harm my chances of getting into a PhD program/getting a job in the competitive field of postsecondary education?</p>

<p>What you might want to do is check with the math department and career center at the undergraduate schools you are interested in to find out what PhD programs the bachelor’s degree in math graduates went to.</p>

<p>PhD program admissions is by department, and departments often do have preferences about undergraduate schools, although such preferences are not likely to be made public. They are likely based on strength of major, rather than overall prestige, and are likely to be based on whether the school has top students in that major, rather than the school’s minimum selectivity as typical school prestige rankings are based on.</p>

<p>Look at this compilation. [REED</a> COLLEGE PHD PRODUCTIVITY](<a href=“http://www.reed.edu/ir/phd.html]REED”>Doctoral Degree Productivity - Institutional Research - Reed College) There are some surprises in the Math/Science area for PhD producers.</p>

<p>^ Under “Science and Math”, the only one I find at all surprising is Wabash. It’s a bit less selective than the others (except for NM Mining). However, it too is a fairly high-ranking school. It’s also one of the “Colleges That Change Lives”.</p>

<p>Across the whole page, most of those schools are very small. Cornell is by far the largest (~14K undergrads); Harvard is #2 (~7K undergrads). Almost all are private. </p>

<p>According to NSF data (<a href=“https://webcaspar.nsf.gov/[/url]”>https://webcaspar.nsf.gov/&lt;/a&gt;), in 2007-2011, 14 University of Iowa alumni earned PhDs in math & statistics. In the same period, 16 Carleton College graduates, 43 Caltech graduates, and 55 University of Chicago graduates earned PhDs in these fields. Carleton has ~2100 undergrads, Caltech has ~1000, Chicago has ~5400. Iowa has ~21,000. However, Iowa awards less than 2% of its undergraduate degrees in math/stat. Caltech awards about 10% of its undergraduate degrees in math/stat. Carleton awards about 5% of them in math/stat.</p>

<p>I would expect a small, private school with a strong intellectual atmosphere to produce a higher percentage of PhD alumni than a large public university with a big D1 sports and Greek scene. Whether that is due more to selection effects or to treatment effects is hard to determine. In absolute numbers, some big state schools do produce lots of alumni with PhDs in some fields.</p>

<p>While there are likely some biases in graduate school admission, by and large, the admission decision is made on the basis of the strength of your academic record (your GPA and how challenging your course selection is), your GRE scores (although it is not clear if this is a good predictor of success in a Ph.D.), your research experience (while REU programs are a good way to get such experience, doing research at your undergraduate university is important), and your letters of recommendation (this is where doing research on campus can really be a plus).</p>

<p>Students from lesser-known feeder schools can get into top programs and be successful so choose your undergraduate school based on the curriculum available and whether you have the opportunity to really get involved in significant research on campus.</p>

<p><a href=“http://www.math.uiowa.edu/undergraduates/UGH2011.pdf[/url]”>http://www.math.uiowa.edu/undergraduates/UGH2011.pdf&lt;/a&gt;&lt;/p&gt;

<p>

</p>

<p>People who want jobs in academia typically aim for the best graduate programs. The harsh reality is that the best programs in a field tend to monopolize the job market, and job prospects coming out of a weak or even average PhD program are often exceedingly poor. </p>

<p>This is why I have always said - dozens of times, at this point - that PhD production data is utterly useless without knowing where students are getting their PhDs. </p>

<p>Math PhD programs seem to be picky when it comes to graduate admissions. From someone who went through the admissions process:

</p>

<p>Universities like Chicago and Northwestern give very good financial aid that can make them surprisingly affordable despite their hefty price tags. PhD programs are typically fully funded (in fact, <em>they</em> usually pay <em>you</em> to attend).</p>

<p>

</p>

<p>Unfortunately, there doesn’t seem to be any source of unbiased, comprehensive data on where students from the top PhD producers go to get their PhDs. Some anecdotal evidence suggests that they often enter strong programs ([LIFE</a> AFTER REED](<a href=“http://www.reed.edu/ir/success.html]LIFE”>Life After Reed - Institutional Research - Reed College)). </p>

<p>“Utterly useless” is a pretty strong statement but I agree that caveats are in order, such as this one:

[quote]
[Baccalaureate origins] studies are best viewed as descriptive data that provide a narrative report of student activity, rather than evaluative indicators that measure educational quality in any absolute terms.<a href=“%5Burl%5Dhttps://wfs.pomona.edu/jlr04747/research%20reports/Baccalaureate%20Origins%20Methodology_v2.pdf?uniq=kfc8lc%5B/url%5D”>/quote</a></p>

<p>Suppose you’re looking for a less selective alternative to a school like Carleton or Swarthmore. High PhD production at Wabash or Earlham, in fields that interest you, may be one indication that these colleges at least have the atmosphere (or “student activity”) you’re looking for.</p>

<p>Since it seems that the “LAC vs. RU” argument is in the background here, we really do not know much about where the OP is in terms of math courses.</p>

<p>But if the OP, like a lot of prospective math majors on these forums, is two years advanced in math (calculus BC as a junior and college sophomore level math as a senior in high school), note that the usual advantages of LACs (small faculty led lower division course) won’t be much of an advantage for the OP in math, since the OP will be going directly to upper division courses (which in math tend to be small faculty led courses even at big RUs). Plus, such an advanced student is more likely to exhaust the math course offerings of a LAC and want to take graduate level math courses.</p>

<p>Well, it doesn’t matter in the sense that when you get hired no one is going to care where you went to undergrad. The only thing is that some LACs have a tendency to prefer professors who went to an LAC themselves for undergrad, because they understand the mission and environment. But there are certainly people who went to large prestigious universities and now teach at LACs.</p>

<p>It matters more in an indirect sense, in that you need good undergraduate preparation to succeed in graduate school. You need to go to a department that fosters deep thought and theoretical understanding of mathematics; a place that has good library and research resources; and a place where professors are willing to supervise you in independent study and research. A place where the departmental resources are at least partially geared towards helping undergrads get into good grad schools.</p>

<p>The University of Iowa - being a large state university with lots of research and respect - is a place like that. You can definitely get into graduate school from there, and no, it won’t harm your chances of getting into a PhD program. It certainly won’t hurt you on the job market for teaching positions, because nobody will care where you went to undergrad by then. It’s also a great point that ucbalumnus makes - if you go to a small LAC with no graduate-level classes, and you are an advanced math student, you are eventually going to hit a wall. If you go to a place like Iowa (or Northwestern or Chicago, honestly) you will be able to take higher-level and even graduate-level math classes if you advance that far.</p>

<p>I would imagine only 14 University of Iowa alumni earned PhDs in those 4 years because only 14 University of Iowa alumni had the interest to pursue it. Large state universities tend to be populated by local kids and people of wide ranges of ability, whereas top colleges like Carleton, Caltech and Chicago attract wealthy and upper-middle-class kids who know they want to get PhDs off the bat. But again, that doesn’t mean anything for the OP’s chances. If he or she gets solid grades in math, does some mathematics research in undergrad and makes good relationships with professors, he or she can still get in.</p>

<p>You might want to read this thread about another math professor aspirant:
<a href=“http://talk.collegeconfidential.com/parents-forum/1510699-day-life-professor.html[/url]”>http://talk.collegeconfidential.com/parents-forum/1510699-day-life-professor.html&lt;/a&gt;&lt;/p&gt;

<p>

</p>

<p>That begs the question.<br>
If Iowa fosters deep thought and theoretical understanding of mathematics, if professors are highly engaged with students (supervising them in independent study and research, helping them get into good grad schools), then why aren’t more Iowa math majors motivated to pursue the highest degree? </p>

<p>(I don’t know the answer to that question; it may not be due to any deficiency in the program.)</p>

<p>

</p>

<p>So compare Iowa’s 14 to the numbers from other public flagships.</p>

<p>School … Alumni PhDs earned in Math/Stat, 2006-2011
Berkeley … 89
Illinois … 24
Maryland … 19
Michigan … 50
Ohio State … 20
Texas … 40
UC Boulder… 27
UCLA … 39
UNC-CH … 16
UVA … 18
Wisconsin …31</p>

<p>Tiny St. Olaf College is only slightly more selective than Iowa. Iowa gets far more freshmen with high SAT-M scores than St. Olaf does (Iowa has 7x as many undergrads and 25th percentile SAT-M scores nearly as high, 690 v. 710). 52% of St. Olaf freshmen came from Minnesota in 2010 (compared to 50% of new Iowa freshman who came from Iowa). Yet St. Olaf still generates slightly more Math/Stat PhDs than Iowa does.</p>

<p>As I said above, we really don’t know if the different PhD production numbers reflect “selection effects” (cherry-picking strong/motivated students at admission) or “treatment effects” (better undergraduate instruction). I suppose it’s possible that many students at the most productive colleges (Caltech, Mudd, Carleton, Chicago, Grinnell) are gravitating to weak PhD programs … but I doubt that. I would say that low numbers (especially if they are low compared to similar schools) is a signal to at least ask a few probing questions, if a career in academia is what you’re after. If you simply want decent employment prospects (not necessarily in academia), I bet you can get that from most any Math degree at a wide range of schools.</p>

<p>^ Interesting that UVA and UNC, both considered among the best public Us, are right near Iowa’s numbers and lagging most others.</p>

<p>I suspect that the best math students in North Carolina, Virginia, and Iowa go to the state land grant universities. NC State, Virginia Tech, and Iowa State might be percieved to have stronger programs than those at the flagships.</p>

<p>^^ Well UNC, UVA, and Iowa are smaller than the other universities I listed. However, it doesn’t appear that size alone would fully account for the differences. Berkeley for example is bigger than Iowa, but not 6x bigger.</p>

<p>

Number of graduating math majors:</p>

<p>67 St. Olaf
50 Iowa</p>

<p>

All right. Let’s take econ as an example. Courtesy of the Journal of Economic Education, calculating PhDs per capita (# PhDs produced / # majors in the subject) for economics results in the following ranking:</p>

<ol>
<li>Illinois Wesleyan</li>
<li>Swarthmore</li>
<li>Bemidji State</li>
<li>Earlham </li>
<li>Reed</li>
<li>Grinnell</li>
<li>Wheaton (IL)</li>
<li>U Memphis</li>
<li>Agnes Scott</li>
<li>U Alabama</li>
<li>Concordia</li>
<li>Oberlin</li>
<li>MIT</li>
<li>Carleton</li>
<li>Trinity U</li>
<li>Hood</li>
<li>Loyola-Chicago</li>
<li>Miami U</li>
<li>Ithaca</li>
<li>Ohio Wesleyan</li>
<li>U Miami</li>
<li>Macalester</li>
<li>Western Kentucky</li>
<li>RIT</li>
<li>Frostburg State</li>
<li>La Salle</li>
<li>College of Charleston</li>
</ol>

<p>Interesting list, eh? No Harvard. No Chicago. No Princeton. No Columbia. MIT is the only top university on the list, and it’s only #13. </p>

<p>Perhaps I’m looking at this wrong. Perhaps a university like Bemidji State (23% four year graduation rate, 8% in top 10%, 21/22 average ACT) or U Memphis (12% four year graduation rate, 15% in top 10%, 22/23 average ACT, econ too weak to be ranked by either USNWR or NRC) is indeed a much better place to prepare for econ PhD programs than Chicago or Harvard. Or perhaps not… :rolleyes:</p>

<p>

</p>

<p>Iowa has 3 state universities. California, with 12 times the population, has 32 state universities. So we can expect state universities in California to each take a much narrower slice of the college-attending population than the state universities in Iowa. Berkeley ends up with a narrow slice at the top, while UI takes a very broad slice. So it is not surprising that Berkeley has a much larger number of top students. On the other hand 6 times might be seen as having fewer than the expected 12 times if each school gets all of the top students in the state, though one has to account for top students being distributed among some other state universities (Berkeley may “share” some of the top students with UCLA and a few other UCs, while UI probably only “shares” with ISU; the tendency of top students in each state to go to private schools also matters).</p>

<p>

</p>

<p>There are too many confounding factors (like the ones ucb mentions) to draw sweeping conclusions from PhD production data alone about one specific school & major v. another specific school & major. </p>

<p>Only <em>one</em> U Memphis alumnus earned a PhD in economics between 2007-20011. <em>No</em> Bemidji State alumni earned a PhD in economics during the same period, according to NSF data. (<a href=“https://webcaspar.nsf.gov%5B/url%5D”>https://webcaspar.nsf.gov</a>). So the Journal of Economic Education may be looking at an atypical time window in which those schools generated a very small number of PhDs in economics from a very small number of economics majors. In that case, they’d be calculating per capita rates from very sparse data. Not good.</p>

<p>More significant, perhaps, is the fact that so many schools on that list are LACs. This is consistent with other PhD production data. So it’s reasonable to ask if there is something about small LACs that makes them better places to prepare for certain PhD programs than other kinds of schools. It may be that small classes and good mentoring trump famous faculty (all else being equal).</p>