<p>back to the origin of this thread: "“Out of 34 countries, the U.S. ranked 14th in reading, 17th in science and 25th in math.”</p>
<p>maybe our kids can’t add but they are tops in self-confidence:
[Overconfident</a> Students Score Lower in Math, UB Researcher Says - UB NewsCenter](<a href=“http://www.buffalo.edu/news/12126]Overconfident”>Overconfident Students Score Lower in Math, UB Researcher Says - University at Buffalo)</p>
<p>all hat and no cattle?</p>
<p>On any measure of output and achievement, US scientists are not better than those of of many other western industrial democracies, if you take into account population size. And you do depend unusually heavily on foreign immigrant scientists in your science and technology based industries.</p>
<p>^Science industries aren’t favored among college students</p>
<p>I tell you what, as much as I dislike American boosterism and chauvinism, Chinese is getting worse. I can’t wait until the Chinese property bubble bursts (shouldn’t be long now) so the “China is the future” crowd will calm down a little.</p>
<p>I looked up the data on % international students in Ph.D. programs in physics at a number of highly-ranked institutions, based on the ranking of doctoral programs just recently issued by the National Academy of Sciences. Here are the percentages of international students in the doctoral programs for a set of these schools:</p>
<p>Caltech 46.0%
MIT 50.0%
Harvard 35.8%
Northwestern 35.5%
Princeton 50.5%, but Plasma Physics as a separate field has 35.5%
Stanford 46.5%
University of California, Berkeley 21.8%
University of Illinois, Urbana-Champaign 42.4%
University of Michigan 43.4%
University of Wisconsin, Madison 25.6%
Yale 43.8%</p>
<p>These percentages are higher than I would have guessed. However, international students are in a minority in all of these Ph.D. programs, except for Physics at Princeton. When you include the numbers in Plasma Physics at Princeton, international students are probably in a minority there, also.</p>
<p>Ph.D. students are supported in three primary ways: by teaching assistantships, by research assistantships, and by fellowships. If the majority of the TA’s (or TF’s) are foreign, but the majority of the graduate students are not foreign, this suggests that the American students are favored for RA’s and fellowships.</p>
<p>Finally, I will mention that it is interesting to look at the data on the “R” rankings of graduate programs in the US. To determine the R rankings, faculty at various institutions were asked to rate the Ph.D. programs at a subset of institutions in their fields. A set of variables were identified that could be used to predict these rankings. The methodology is described (somewhat incompletely, as far as my browsing went) at the site:
[url=<a href=“http://www.nap.edu/rdp/#download]A”>http://www.nap.edu/rdp/#download]A</a> Data-Based Assessment of Research-Doctorate Programs in the United States<a href=“regression%20analysis,%20principal%20component%20transformation,%20etc.”>/url</a>
So the R rankings are not based directly on the assessments by the faculty, but rather on the implicit statistical model derived from their rankings. A range of possible R rankings were determined, and the middle 90% were used. (I didn’t find the determination of the “middle 90%” described to my satisfaction in the part of the report I skimmed, but it might be detailed elsewhere.)</p>
<p>In any event, the significant point is: At both ends of the middle 90% of all R based rankings, the % of international students in the Ph.D. program correlated negatively with the ranking of the doctoral program in the following disciplines:
Biochemistry, Ecology and Evolution, Immunology, (something I wrote down as “Int Bio” and can’t recall exactly now), Microbiology, Neuroscience, Pharmacology, Physiology, Public Health, Aerospace Engineering, Biomedical Engineering, Chemical Engineering, Mechanical Engineering, Operations, Chemistry, Computer Science, and Earth Science.</p>
<p>That is, in these fields, according to the overwhelming majority of all ranking algorithms derived from the assessment of the programs by faculty, the number of international students is a negative indicator for the quality of the program. </p>
<p>(Note that the “middle 90%” does not refer to the middle 90% of Ph.D. programs. It refers to the middle 90% of the ranking algorithms themselves.)</p>
<p>In Physics, the % of international students is correlated negatively with the quality of the program starting at the 5% point of the ranking algorithms, and positively at the 95% point. However, the absolute value of the negative correlation (at the 5% level) is much, much larger than the absolute value of the positive correlation (at the 95% level). The NRC committee refused to release the median correlation coefficients, but I’d have to guess that it is negative for international students in Physics Ph.D. programs, based on the data they did release.</p>
<p>A similar situation holds in Genetics and Astrophysics, and to a lesser extent, in Math.</p>
<p>Foreign students are clearly beneficial to the Ph.D. program R rankings in Psychology, Public Affairs, Sociology, Classics, Animal Science, and Nursing. Not what you’d expect.</p>
<p>In most fields not named above, foreign students contribute negatively to the program ranking at the 5% cut-point, and positively at the 95% cut-point. Typically the absolute value of the negative coefficient is larger than the absolute value of the positive coefficient.</p>
<p>In response to sorghum’s comment (#366) that American scientists are “not better than those of many other western industrial democracies,” I agree. Why should we be? I don’t think we are worse either, though. </p>
<p>I also agree with the comment that we depend significantly on foreign immigrant scientists in science and tech fields. I attribute this largely to the fact that US scientists and tech-types generally make middle-class salaries. The economic lure of i-banking, the law, and medicine should not be discounted. Anyone wanting to claim that scientists in the US should be paid more has my enthusiastic support!</p>
<p>screwitlah:
</p>
<p>This is not worth my time, so removing the comment.</p>
<p>It has been pointed out to me that there are citizenship restrictions on many graduate fellowships, so the “preferences” might actually be restrictions instead. When I started as a faculty member, I was informed by my program officer at the National Science Foundation that I could support graduate students of any nationality but Japanese on my research grants–so you can guess when that was. Now there are no formal restrictions of that sort, but I wouldn’t be surprised if there is some subtle encouragement to support Americans with research assistantships from a professor’s grants. In fact, when one files the progress reports with NSF, there are questions about the citizenship of all supported on the grant.</p>
<p>That the US students are ranked low and have been so for years is all over the internet via Google, Facebook, etc while we are sitting at our PCs or Mac typing. The only funny thing is that the products I mentioned are of American origin.</p>
<p>Knowing known knowledge/technologies and creating new knowledge/technologies may not be that tightly positively correlated. As long as we allow our kids to freely imagine and experiment, we’ll be more fine.</p>
<p>Yeah yeah yeah – typing on your computer on the Web, which is of European origin.</p>
<p>Re: #369
To me, the presence of a large number of foreign graduate students in US universities is a HUGE positive thing rather than a negative problem for the US. Welcoming immigrants, especially those among the best and brightest from all corners of the world is an advantage that no other country in the world enjoys, not japan, not china, not any of the european countries. </p>
<p>I am just surprised that a physics professor at a research university exhibited such ignorance of how graduate students are admitted and funded. No department (that I am aware of) does not admit and active recruit qualified US students first before even looking at files of foreign applicants, and generally foreign students have to be clearly much stronger candidates to be admitted ahead of any domestic students. The fact the Caltech, Princeton, MIT, Harvard et al, the very top physics programs that get the 1st picks of all top US applicants, have so many foreign graduate students shows just how strong the foreign applicant pool is. </p>
<p>There are many reasons why departments/professors much prefer domestic students. First, the department does not like to have too many foreign TAs, because undergraduates (and sometime their parents) universally like to blame (justified or not) their poor class performance on foreign TAs. Second, foreign graduate students are generally much, much more expensive for the professors. Through there is no general restriction on PI’s research grants, foreign students are generally supported exclusively by professors’ grants or TAship (which take them away from the lab), while US students can also qualify for many predoctoral research fellowships from NSF, NIH, DOE, DOD etc, which are available to US students only. At public schools, the tuition cost (out of state tuition) is also significantly higher for foreign students, which are generally carried by professors’ research grants.</p>
<p>
The above statement clearly suggests (at least to me) that foreign graduate students are inferior, but it only reveals the poster’s ignorance. At many top programs, most graduate students don’t join a research lab in the 1st year but rather are offered the time and option to choose which professor to study with for their graduate career. Thus, the 1st year students are supported not by individual professors, but by the departments with training grants or TAships. These training grants (from NSF, DOE, DOD, NIH) again exclude foreign students, whose only financial source left is TAship (outside very few private funded scholarships at a few departments). </p>
<p>
This statement is just astounding, especially coming from a physics professor, with its causal relationship so clearly stated.<br>
The lower ranked schools having lower ranks is NOT CAUSED by having too many foreign students, their having mostly foreign graduate students is BECAUSE OF their lower ranks and therefore the inability to attract qualified domestic students. Not for these well qualified foreign students laboring long hours in the labs, many of these departments would have ceased to exist as research programs.</p>
<p>Re #375: Sorry, I did not mean to imply that I think that foreign students make a physics department worse. Causality of that type was not intended. What I meant was this: When you look at the coefficient for % foreign students in the R-based analyses recently compiled by the NRC, you find that it is negative at the 5% cut-point for the great majority of programs, and still negative at the 95% cut-point for many. The causality direction is this: The faculty rating departments tended to give higher ranking to those with fewer foreign students (without considering in any way the number of foreign students actually present in a department), and this resulted in negative coefficients for the presence of foreign students.</p>
<p>I am generally in agreement with NCL, actually. I agree that the top programs have their pick of the American students, and then admit some truly outstanding foreign students to fill out the classes. We saw a noticeable change in the quality of our foreign graduate students (downward) when the University of California schools began to admit significant numbers of foreign students in my field.</p>
<p>Typically our foreign students are stronger than our domestic students in theory and in mathematical background. They are usually less strong in computer programming, use of symbolic mathematics programs such as Mathematica or Maple (though this might be changing), and construction of new electronics. Don’t mean to start a firestorm here, just stating my observations, which could well reflect the stratum of my university (not the top).</p>
<p>When on sabbatical at a top school, however, I did not notice that the foreign students were better at research than the (presumably top) American students.</p>
<p>It is true generally that foreign students must have much higher GRE scores than American students, to be admitted to an American university for graduate work. This certainly holds in the stratum of my university, and I suspect that it holds higher up, also. Ironically, I have fought for many years for my department to admit more foreign students, because it seems to me to be a question of justice in admissions decisions. My colleagues have argued that the foreign students prepare intensively for the GRE, including subject tests, while the American students take it cold, and this skews the scores. I do see some rationality in this argument–and observationally, the statement appears to be true of American students I know. On the other hand, I cannot imagine what level of preparation it would take to compensate for 21 years of living as a native English speaker, when it comes to the verbal portion! I had a superb Chinese graduate student who scored at the 80th percentile on the verbal part of the GRE, well ahead of many of our American admits. I think there is heavy prejudice against foreign students on the part of some of my colleagues.</p>
<p>I should retract the use of “favored” with regard to the selection of American students for fellowships (for which they may be the only ones eligible) and for research assistantships, and just say that more of the American students are given these opportunities, apparently.</p>
<p>Mainly, my post was made in response to screwitlah, who claimed that there were few if any American graduate students in the top programs. Just over 50% doesn’t seem to fit that bill.</p>
<p>Finally, I note that the tuition charges for foreign students vary by university. At mine (public) the costs of tuition and stipends are identical for all Ph.D. students, regardless of citizenship or residency.</p>
<p>A quick addendum: </p>
<p>I think that the timing of grad students joining groups depends on the university and field. In my university, it is common for incoming students to know already which group they wish to join, and they usually sign up as soon as allowed.</p>
<p>And another quick addendum:
When I say the costs of tuition and stipends, I mean the costs charged to grants of the professor who acts as the research adviser of the student(s) in question.</p>
<p>And yet another quick clarification:<br>
When I say it is “ironic” that I have been fighting for years to get my colleagues to admit more foreign students as a matter of justice, I mean that it is ironic because I apparently seem like a xenophobe to some on this thread–which I am reasonably sure that I am not.
I do not mean that it is ironic because it tends to depress our R ranking (although it does).</p>
<p>
</p>
<p>That might be said about the international assessment and the SAT/ACT tests, showing limitations of any single test when comparing some prepare for it intensively while others take it cold.</p>
<p>@post #374: C’mon, everyone knows that Al Gore invented the internet.</p>
<p>Seriously, though, the internet is the descendant of ARPANet, followed by NSFnet, followed by the internet as we know it now. It runs on the TCPIP protocol. The web is an application that sits on top of the internet, analogously to the way that a word processing program sits on top of the op system.</p>
<p>The web is indeed a European creation, from CERN. The web added the visuals and other elements to the earlier internet-based information retrieval system, gopher, which was developed at the University of Minnesota.</p>
<p>^ Of course, but gopher doesn’t support in-document hypertext linking does it? It was that that really allowed the web to take off.</p>
<p>One other measure of success of foreign-born scientists and engineers can be found at the Silicon Valley. Just go and looked at the composition of employee in those high tech companies. In fact, in Applied Materials, over half of the engineers, engineering managers, directors are from India, China, Europe, Japan, Korea.
The phenomenon is primarily due to the fact that brightest student in our high school today do not choose science and technology as their career. Most of them (my S included) are leaning towards Medical, IB and other profession that makes better return on their education investment.</p>
<p>@keepittoyourself, you are right that gopher did not have that feature. However, gopher represented a conceptual change in the use of the internet, which at that time was heavily driven by scientists and engineers running heavy-duty computations (well, heavy-duty for the era, anyway). It was a significant step on the way to the web.</p>
<p>^ But it was beaten to the punch by minitel by about 10 years in that respect.</p>