next »

[ucprof]

I am in a field where the course requirements for graduation with a bachelors typically are well below what we like to see for admission to our PhD program.  Many seemingly smart students apply who do not have the background to pass our entrance qualifiers so we are mainly admitting from Ivies, R1 publics, and overseas.  I'd like to hear from people who train undergrads your perception on how to get better trained US students for Ph D programs in math and science.  We have a small program at my R1 that steers undergrads towards the PhD track but even those find it very hard compared to the standard undergrad courses.  Biggest issues are students taking too many non-major electives vs. advanced major and graduate courses, and students from colleges where the advanced major courses tend to be too watered down to prepare them well for PhD programs at our R1. The latter applications we filter out with GRE subject exam and other metrics that make sense for us.  Would like to see more well-trained US students applying to our program.  My sense is biggest problem is lack of training rather than smarts.

[namazu]

Do you offer any kind of post-baccalaureate or master's program that could serve as preparation for your PhD program?

[scampster]

I'm in engineering, but I'm just a grad student so I haven't been on the admissions side of things. Is there no national standard for what is acceptable for a math or science major, regardless of school?

This is my third engineering school I have been in and the meat of the curriculum between the three schools has been generally the same - but they have all been R1s and I think the accreditation program keeps a certain amount of uniformity. It may be that because of this uniformity that admissions committees have lower standards when it comes to advanced coursework in the field.

Do you think it is that the students don't have good faculty advisors for choosing courses? Or is it that they may have academic advisors who aren't faculty members and don't realize they are hamstringing their bright students by not telling them they need to take more rigorous course work?

I don't really know how you get around the watered down advanced major courses. If the program in general isn't strong then they probably have a hard time catering to those bright students who can handle the challenge.

Are entrance qualifiers standard in your field? By entrance qualifiers, I am guessing that means they have to take the exam the first semester they are there and thus have no time to fill in some minor gaps in their background? our qualifiers typically take place a year or two into the program, so there is time to catch up.

[namazu]

Are entrance qualifiers standard in your field? By entrance qualifiers, I am guessing that means they have to take the exam the first semester they are there and thus have no time to fill in some minor gaps in their background? our qualifiers typically take place a year or two into the program, so there is time to catch up.

Yes, this was my other thought.  Could you institute some kind of conditional/probationary acceptance, requiring students to either pass an entrance qualifier or, if unable to pass the entrance qualifier, to take some "remedial" advanced courses at your institution to get up to speed? 

Could you offer a "summer bridge"-type program for incoming grad students?

Do you publicize your entrance prerequisites (including details of topics that should be covered) prominently on your admissions website?

Also, I'm a little confused -- are you saying that even your own institution's students are unprepared for graduate school in your field?  Or just that very few students from other schools are qualified for your graduate program?  It would probably be easier to effect change within your own institution -- adding some courses to the major, making the curriculum more rigorous, offering an "advanced"/pre-PhD track, etc. -- than it would be to raise the preparation of the entire nationwide or international student pool, which is why I suspect that your institution may need to offer advanced training opportunities if it desires to recruit and retain students from a wider range of undergraduate backgrounds.

Finally, as Scampster suggested, could you work through your professional organization to draw up a standard curriculum for undergrads who seek to pursue doctoral degrees?  Do others agree that inadequate preparation is a problem, or is your graduate program simply more rigorous (or unforgiving, if early qualifiers are an issue, e.g.) than most?

[ucprof]

Do you offer any kind of post-baccalaureate or master's program that could serve as preparation for your PhD program?

We do offer a masters, but it has been greatly scaled back and we are discussing completely revising this.  The problem with the masters is that the students will have to pay to enroll and with many students in science/math going straight to PhD rather than masters, we're not convinced we will get the best students into this program.  Nor can we guarantee them a slot in our PhD program which has become pretty competitive.  The other issue with a masters program is that it means enlarging our curriculum which is going to be hard to do with the current budget-cutting, course cutting situation in the state of CA.  We are staffing our graduate program with a small crew of senior faculty and using alot of adjuncts/postdocs to teach our upper division major courses because of lack of senior faculty.  Still it bothers me that there are lots of smart kids out there who essentially have no shot at our PhD program (but they do apply) because they are not prepared. 

[namazu]

Do you offer any kind of post-baccalaureate or master's program that could serve as preparation for your PhD program?

We do offer a masters, but it has been greatly scaled back and we are discussing completely revising this.  The problem with the masters is that the students will have to pay to enroll and with many students in science/math going straight to PhD rather than masters, we're not convinced we will get the best students into this program.  Nor can we guarantee them a slot in our PhD program which has become pretty competitive.  The other issue with a masters program is that it means enlarging our curriculum which is going to be hard to do with the current budget-cutting, course cutting situation in the state of CA.  We are staffing our graduate program with a small crew of senior faculty and using alot of adjuncts/postdocs to teach our upper division major courses because of lack of senior faculty.  Still it bothers me that there are lots of smart kids out there who essentially have no shot at our PhD program (but they do apply) because they are not prepared. 

Yes, that is tricky, especially given the budget and curricular constraints you mention. 

Is your program doing any outreach or targeted recruiting yet?  While it may be near-impossible to dictate the content of other schools' courses, you might be able to at least broadcast to wannabe-grad-students the courses that they should include in their curricula to be well-prepared for entry and success in the program.
- Could you sponsor some REUs (assuming you're in an NSF field) or some equivalent, and "groom" students from less-prestigious schools (with the idea that they'll get some good research experience and internalize ideas about preparation, and even if they don't matriculate in your grad program, they'll be successful elsewhere)? 
- Do you have colleagues at other institutions (within the UC or Cal State systems, or elsewhere) whose help you could enlist?  If they are advising grad-student-hopefuls, they might be able to pass along your advice re: curriculum choices and note that competitive schools like yours will want to see those courses.

Also, if you're not in a position to offer a master's degree, but you do teach the advanced pre-req courses at the upper-division undergrad level, could you admit some of the underprepared students who seem to have great potential, and require them to complete some of these courses during their first couple of years in the PhD program?  (I don't know if that would be feasible in terms of funding or curricular requirements.)

[daniel_von_flanagan]

All departments in math, physics, and chemistry (among others) are facing the same questions, especially since there is increasing pressure on departments to justify their existence using the number of majors.

I'm torn on this - I think every major in these and related fields should be prepared for graduate work in some science discipline, on the other hand why shouldn't a student who enjoys (for example) math and has reasonable facility in it be allowed to get a non-grad-school-prep degree in it?  I would like to see more relatively light courses in these fields available for seniors, not fewer.  Moreover, college is a convenient time for students to get a broad education, or be exposed to more field options, so forcing more credits on your best students might not be doing them a favor.

These fields do have national standards for majors, but they don't guarantee success in grad school: like the OP, we regularly get majors from top programs who have had a very good selection of courses and therefore a very credible undergrad degree, but for whom it is soon apparent that they are not ready (or possibly good enough) for graduate work

In my department, and many others, it is common for admitted students to take some number of senior-level courses their first year to patch the gaps in their background.  There is nothing new about this, I'd say that 30% of the students in my grad school cohort decades ago did the same.  (On the other hand, few if any of these students made it through to their PhD!)

Really, I think the problem is viewing the situation as a problem. All we can do is admit students with some flexibility for deficiencies, give the students their best/fairest opportunity to prove themselves when they arrive, and find as pain-free as possible a path out of the program for those who don't succeed.

Now, if only I can convince our administrators that our small proportion of math/physics/chemistry majors is due to the difficulty of the major and the superiority of job opportunities in engineering, business, and computer science, rather than because our programs are unreasonable... - DvF

[ucprof]

Lots of great questions and comments.  Rather than quoting I will try to answer here:

1.  Engineering - different situation due to accreditation and the professional masters.  I work with some engineering students incl. masters and the situation there is completely different from science and math. No shortage of good masters students but most of them are in it for the masters and will go work in industry afterwards.

2.  Do we post requirements?  Yes in great detail including prior qualifying exams posted on the web for people to download.  Do people read our web page in great detail?  Not sure about that since I often end up answering these very same questions on the phone with prospective students.

3.  Qualifier: In reality it's not an "entrance" qualifier but the first qualifier for the PhD that we like students to try when they arrive - they have to pass it fairly quickly anyway and it has been a weed-out in the past.  We also do offer an intensive summer course for incoming PhD students to better prepare them for the first qual.  We also fund them to attend this course.  Despite all of this we still find that students without adequate preparation can not master the qualifier with just the summer prep program.  And they will flail in our first year PhD courses.

4.  Recruiting: Ten years ago we had significant attrition from the PhD program so we tightened up admission standards, beefed up recruiting, and had a number of dept successes so we are now ranked top 5-10 in almost all subfields. 
Getting good PhD students is not the problem - it's just that they are all now coming from MIT, Stanford, Harvard, Princeton, and similar level places, plus a smattering of top R1 publics. And these good, prepared students mostly pass the first qual in the first year.  I could name an extremely well-known SLAC where students are highly unlikely to get in because they just can not handle our graduate courses straight out of undergrad. That said, there are some smaller colleges from which we are happy to recruit - Harvey Mudd is one and their students are typically well prepared. 
5.  Our own undergrads - we have a couple of honors programs, one of which gives them a masters in 5 years and involves them passing the first PhD qualifier.  So we do recruit from that program, but it is just a couple of students a year who are interested to stay and whom we want to admit.  It's a very small program, and we graduate close to 300 majors every year.  I will write a separate post with what I see is the issue with the SLACs and and other places.

[ucprof]

Anecdotally, I have some colleagues who teach at SLACs and I hear from them that they are under alot of pressure to hold the hands of the students and make them comfortable in courses.  The problem is that when you are doing hard sciences and math the students just may not be challenged enough.  Faculty `in the know' at some of these schools are frustrated that they can not teach at the PhD prep level without going way out of line from how others are teaching at their institution.  However I think they are doing the students a disservice because they are not getting the same education as those at the Ivies but are led to believe it is better due to the attention.  The one exception I can point to is Mudd (and being in California I know that one fairly well) where they do not coddle the students so much, although they still get a lot of attention and that alone requires some adjustment when the students go to grad school.

A completely different situation is where you have an advisor at a SLAC who does not know what the prep should be and encourages the student to take courses in topics outside of the major.  One applicant that drove me nuts was an obviously smart student from a SLAC on the east coast who had very high grades in the few courses in the major that she took, and lots of courses in - I forget - but I think it was dance.  She took the bare bones needed to graduate in the science major and then applied to our PhD program with great enthusiasm.  I don't know what she or the advisor was thinking but there was no way she was ready for our PhD program with that preparation. 

Interestingly enough, we've found that students who get a masters at one of the Cal States are sometimes better prepared than students coming straight out of undergrad at what should be TOP SLAC.  So I think something is amiss at those schools.  I mean the students at the SLACs are paying through the nose in tuition and should be trained to handle a PhD program at any of the top places in the country, you'd think?  I've also talked to people at the National Science Foundation who have collected hard data on this and the numbers are consistent with my anecdotal experience. 

In contrast we now get a lot of excellent foreign students applying who come with masters degrees.  Some can pass more than one or all of our PhD quals walking in the door.   They are much more expensive to fund due to the NRT in the state of CA.  Also they can have trouble with the language and when it comes time to writing the first (or second) research paper you have to spend a lot more time holding their hands on the writing than you do with undergrads trained in the US, on average.  Over time, I see the US students catch up to the foreign ones, especially when they can much more quickly get papers out because they know how to write and don't need to rely on the advisor to do heavy editing of the manuscript.

[ucprof]

Regarding hosting REU students from other institutions -- it's just not enough time to train them in what they need.  My school has a famous REU program in my field - and it's competitive enough that we are getting the same level students in it that we recruit for our PhD program.  Something like one in ten applicants gets into that REU program.  Also it's a research program not a classroom program so it won't cover basic curriculum needed.  Nor could you cover it in eight weeks.  We also have an internal REU program for our own undergrads where we run a GRE prep course that seems to help some of them go over the advanced undergrad material.  But again it's just a few weeks in the summer.  The real issue is that the students need a couple of years of training at a certain level of rigor and challenge.  And if they don't have it we can not bring them up to speed in a short period of time.  Also our PHD program is not structured to comfortably allow students to do a whole year of undergrad courses starting out, nor should it be I think.  It is, after all, a PhD program.

[daniel_von_flanagan]

A completely different situation is where you have an advisor at a SLAC who does not know what the prep should be and encourages the student to take courses in topics outside of the major.  One applicant that drove me nuts was an obviously smart student from a SLAC on the east coast who had very high grades in the few courses in the major that she took, and lots of courses in - I forget - but I think it was dance.  She took the bare bones needed to graduate in the science major and then applied to our PhD program with great enthusiasm.

I'm with the advisor on this one.  SLACs encourage broad knowledge, and I think that that is a strength, not a weakness.  If a math student (for example) has enough undergrad training that they can prove theorems, but perhaps only has classwork in algebra and topology but not in senior-level analysis or probability, they can make up those deficiencies later (or are not graduate-level material in the first place).  If your placement exam penalizes students for lack of breadth, then it is an idiotic exam, and if your program doesn't have a place for such students then it should be overhauled.

I've also talked to people at the National Science Foundation who have collected hard data on this and the numbers are consistent with my anecdotal experience. 

What exactly do these "hard data" support?  That students with broad degrees from good SLACS do worse in STEM grad programs than those with a narrowly-focused undergraduate background in the field?  I do not believe this. - DvF

[namazu]

[I wrote most of this before I saw DvF's follow-up, so if some of it is repetitive, I apologize.]

Yes, I understand that the REUs are research experiences -- I just meant that while you had the students' captive attention, you could emphasize the slate of courses you felt were necessary for them to be well-prepared for grad school, not that you would be providing those courses during the REU.  But if, as you suggest, the REU students are coming from the places that are already over-represented in your grad program (the MITs and CalTechs and Ivies and Harvey Mudd), then this sort of outreach would be "preaching to the choir", I suppose.

It sounds like you're torn, to a large extent, between wanting to encourage applicants from a greater diversity of backgrounds, but at the same time wanting to be sure that all students have a very rigorous (but not-very-flexible) and comprehensive curriculum.  I'm not sure you can really have it both ways.  Especially as far as recruiting SLAC students is concerned, I have to guess that there's a large amount of self-selection going on.  Students who want a broad liberal arts curriculum, the freedom to explore, and perhaps some amount of handholding (or at least "small class sizes" and a "supportive environment") are the students most likely to choose a SLAC, and these same students may be predisposed (and encouraged!) to sample from a wider variety of electives than their counterparts at science-focused schools or at large, sciency R1s.  I'm not convinced this makes them, a priori, less likely to succeed in graduate work in science, though it does put them (temporarily) behind in terms of field-specific content knowledge compared to their more "focused" peers.  And smaller, non-science-focused schools and public or private schools with smaller departments may also be unable to offer the breadth of advanced courses that would benefit students headed to your program.  So it really is a tough thing to reconcile.

It seems as though some of your better options, if these pre-reqs are really dealbreakers, may be to step up your broadcasting of prerequisites -- talking to advisors in science departments at some of the top SLACs and other schools to make it clear that advanced courses, beyond the bare minimum for a major, are necessary for admission/success in your competitive program -- and perhaps to do as DvF suggests and allow for a little more flexibility in your entrance requirements.  Perhaps it's unreasonable to admit students who would need a whole year of remedial work, but maybe a course or two?  Perhaps even a full quarter or two (or a semester, if you're at Berkeley)?  Maybe you could make it work, if encouraging students from a variety of undergrad backgrounds is important to your program.

(I should also note that I'm coming from an interdisciplinary science field that is graduate-studies heavy, i.e. not routinely explicitly offered at the undergrad level.  Thus there is an expectation that most incoming doctoral students will require some shoring-up of their background in component fields in the early years of their doctoral study, and this sort of background-building is actively encouraged, along with applications from students with master's degrees in related fields.)

[ucprof]

These are all really good suggestions.  If our revised masters program ever takes off it may address many of these issues.  The old way of doing things had too much attrition from the graduate program and hence wasted resources. 
Yes if the student only needs one or two courses we have been known to make exceptions.  We've also had some conditional offers in the past.  Since we are a public institution we do try to err on inclusiveness when we can.
And it's not to say that none of our students come from SLACs - we have some excellent ones with good training. 
Also your suggestion about talking to undergrads about what is expected is something we are already doing. 
I gave a talk to about 50 REU students this summer about the graduate program and many of them stayed for an extra hour asking questions.  Hopefully when the budget situation in CA improves we can work more on the masters program so that it will better fill in the gap between undergrad and grad.  And hopefully fees at the UC's will not be so high by then as to preclude many students from participating in the program.  They are already talking about 30% fee increases over the next two years (fee = "in state tution" at the UC).

[cgfunmathguy]

See, I never understood the trend of getting a BS and going straight to a PhD program, but this seems to be the norm in math and science. When the trend was BS to MS (with a thesis required) to PhD, you got better (if fewer) students in PhD programs. Maybe it's time to push back against the current trend.

[pollinate]

See, I never understood the trend of getting a BS and going straight to a PhD program, but this seems to be the norm in math and science. When the trend was BS to MS (with a thesis required) to PhD, you got better (if fewer) students in PhD programs. Maybe it's time to push back against the current trend.

Chime!  In my experience, going straight for the PhD is OK if the student has been intensely focussed on a narrow area since early undergrad, has exceptional ability, and has no other strong interests.  For everyone else, especially PhD's who will end up collaborating or teaching outside of their PhD research area (especially if this will include anything like survey courses in their field) broader training and experiences seem to produce better results.  Both in grad school and as faculty I've seen numerous examples of people with such narrow knowledge bases that they were completely at sea when having to deal with any other aspect of their supposed field. This was funny in a pathetic way in some of them and downright dangerous in others.