<p>It is not impossible to do physics research at an LAC as a professor, but it would be quite difficult to do really important research. There was a logician named Post, who did important work in mathematical logic while teaching in a high school in New York, and there are other examples of people without much support accomplishing significant research, but it is truly very hard.</p>
<p>The “teaching load” at an LAC is 2 to 3 times the “load” at an R1 university. (It’s probably not called a “load” at an LAC.) The time not spent on undergraduate teaching is available to guide graduate students and personal work on research, for a professor at an R1 institution. This makes a difference in how much can be accomplished. Also, a professor at an R1 institution will be working with a group of graduate students and post-doctoral fellows–as well as undergraduates–who help to push the work forward.</p>
<p>There is one advantage that I can think of, for a future physicist to go to an LAC: namely, the odds of being hired later at an LAC are higher. Physics is a tough field, for someone who wants to stay in it and conduct research. The last I saw information in Physics Today, there were faculty positions at research institutions for only about 10% of the post-doctoral researchers in physics. This may ease up in the future, but right now, the hiring situation is still pretty difficult.</p>
<p>This is an illuminating thread for someone whose kid has just decided to attend an LAC. The information from ground level is informative. A completely different perspective is offered by an NSF study of the undergraduate origins of science and engineering Ph.D. recipients. Here is the link: </p>
<p>Page 6 gives the percentages of graduates who subsequently earned a Ph.D. Here are the top 20 schools: </p>
<ol>
<li>California Institute of Technology </li>
<li>Harvey Mudd College</li>
<li>Massachusetts Institute of Technology</li>
<li>Reed College</li>
<li>Swarthmore College</li>
<li>Carleton College</li>
<li>University of Chicago</li>
<li>Grinnell College</li>
<li>Rice University</li>
<li>Princeton University</li>
<li>Harvard University</li>
<li>Bryn Mawr College</li>
<li>Haverford College</li>
<li>Pomona College</li>
<li>New Mexico Institute of Mining and Technology</li>
<li>Williams College</li>
<li>Yale University</li>
<li>Oberlin College</li>
<li>Stanford University</li>
<li>Johns Hopkins University</li>
</ol>
<p>LACs are well represented. There are several caveats. First, the raw numbers do not tell us about the quality of the graduate institution or the students’ subsequent career paths. Second, other studies have shown that faculty members’ kids attend LACs disproportionately, so the performance of LACs here could be partly a function of student preferences. Third, schools that have many professional schools at the undergraduate level suffer in these comparisons because the professional schools add to the denominator but do not add much to the numerator. (For example, a nursing school adds to the number of students but does not add much to the number of Ph.D. recipients.) Fourth, the NSF definitions include social sciences.</p>
<p>It’s worthwhile to look at the table on p. 7 also, which lists the undergraduate institutions of Ph.D. recipients in science and engineering, in order of the total number of doctorates. Penn State is number 6 on this list. The University of California, Berkeley, the University of Michigan, the University of Illinois at Urbana-Champaign, the University of Wisconsin, the University of Texas at Austin, and UCLA are all in the top 10 on this list, along with Cornell, MIT, and Harvard. The other “usual suspects” are there, too, a little lower down. </p>
<p>coase mentioned that a nursing school at an undergrad institution will add to the denominator in number of bachelor’s graduates, but not so much to the numerator, in terms of Ph.D.'s and so can suppress the fraction of undergraduates who go on to receive doctorates. The fraction who go on to a Ph.D. does affect the overall climate of the school, and it is a worthwhile number to know. On the other hand, at a large R1 university, students who are interested in eventually pursuing a Ph.D. will tend to know and interact with each other, and there are large numbers of them. The academic and even social environment of a physics major will not be much affected by the students majoring in business, communications, elementary education, etc., who are not going on for Ph.D. degrees.</p>
<p>It might be worth noting that Harvey Mudd, #2 on the list for the highest percentage of students going on to earn Ph.D. degrees (p. 6 of the report coase cited), still had fewer than half the number of Ph.D. recipients (in the time frame studied) of the #50 school on the list of the baccalaureate institutions generating the greatest total number of students who received Ph.D.'s eventually (p. 7 of the report).</p>
<p>It would appear that any “research - very high” activity school listed in the table on page 6 would be the best choice (but they are pretty tough to get in to). They have the resources AND a high proportion of their graduates inspired to seek a full-time science/engineering career.</p>
<p>This is an interesting set of data. Off topic, but I find it notable that few Southern schools appear in the table on page 7 (U Florida = 13, Georgia Tech = 44) or in the table on Page 6 (Duke = 30, but is this “Southern”?, Hendrix = 42). Where are Vanderbilt and Wash U? We see a lot of hype about U Alabama, but it is also absent. U Georgia? Emory? Rhodes?</p>
<p>A quick glance at Wikipedia tells me the undergraduate population at each school is:
Notre Dame (#50 on the list) = 8, 371
Harvey Mudd = 738</p>
<p>Kids who wind up matriculating to Harvey Mudd College usually know exactly what they’re looking for. They get top-flight instruction, loads of attention from professors who are there to teach, and large quantities of hard work.</p>
<p>Percentages are so misleading. Large U’s not only have students of the small LAC caliber but also have a mission to teach the average college student. It is so limiting to only interact with so few students. The same numbers of high caliber students can be found at the large flagship U’s and can do as much. Top flight instruction, …attention, …hard work equivalent or better than at a LAC- especially in the sciences.</p>
<p>It is a valid point that a larger school may have more students with similar interests. At the same time, the classes with 400 students and the large “lower tail” may detract from the educational experience. Having taught for several years at a large university you may know (the one located on an isthmus where everyone wears red on Saturdays), I would not discount these factors. </p>
<p>I mentioned in my earlier post that the children of faculty members disproportionately attend LACs. (They also attend private research universities disproportionately, but not so markedly.) Here is the relevant study: </p>
<p>Reasonable people can disagree about the correct interpretation of the findings in this study. All I can say is that D1 attends a small research university, D2 will attend an LAC, and we did not encourage either to attend a large school despite their interests in science and engineering.</p>