There have been a number of responses to my Science Is Hard post over the last several days, and I've been trying to come up with something to say about them. This is the second of two posts responding to comments by some of my fellow ScienceBloggers.
Turning to Steinn's first post on the subject, I actually hadn't intended to link the "Science Is Hard" post to the "Why They're Leaving" post. Those two subjects just happened to catch my eye on the same day. Their juxtaposition was not meant to imply that students leave science because science is hard-- in fact, the particular difficulties I was talking about are not things that most students encounter in science classes at all.
Steinn has a pretty good explanation in his post, where he writes of science majors that:
The competition is professional degrees: medicine, law, business and others - there the work is hard and technically challenging, but the probability of someone getting through the hoops actually getting a job directly in their field is much higher. Perception is that promotion is rapid and assured and that there is a realistic possibility of high pay and choice of location.
A career in research science is not a path to riches, or even stable employment. Anyone who thinks so is sadly deluded, and if sure promotion and a fat paycheck are your primary goal (and you're good at math), you should become an actuary or an accountant or something in that vein. A career in research science can be very rewarding, but the rewards are not necessarily financial (though I hasten to add, I'm not making a bad living, either).
(Disagreement below the fold.)
Of course, at the level that they were talking about in the Inside Higher Ed piece I was citing in the "Why They're Leaving" post, I don't think that career prospects really enter in, either. While there are some exceptional students who come in knowing exactly what they want to do with their lives, and having a clear and accurate picture of what they can expect from that career track, there aren't many of them (and, frankly, they kind of creep me out). Most students, in my experience (both as a student and as a professor), sort of wander into a college major based on classes they happen to like or dislike, and get to their junior or senior years with only the haziest idea of their career options, let alone career prospects.
The students the Inside Higher Ed piece was discussing leave science well before that point-- a factoid I picked up at an AAPT meeting a few years back is that only something like 3% of students who take introductory physics take another course in the discipline. They're not getting out based on a rational assessment of career possibilities, they're getting out because they don't like the first class or two that they take. By the time they find out about the lousy career possibilities, they're too far in to change majors...
(This is not to say that we don't need to work on informing students about their options, and improving the range of those options. We should do much more than we do, on both of those fronts.)
I also want to object to a apssing remark near the end of Steinn's post:
Addendum: I Am Not A Lawyer, but... my cousin is, and three of my close relatives are MDs.
My perception of work in the professions, is that they do indeed work very hard, and most likely worked harder through university than I did. BUT, the work was mostly rote - it was learning stuff that was already out there - precedents and principles, binary tree decision making and information dumps on which to make decisions, experiential stuff.
I'm not a lawyer either, but I'm married to one, and I think this is a little unfair to lawyers. Saying that law is all about precedents and principles is a little like saying that physics is all about multiplication and addition-- while it's strictly true that most physics problems come down to fairly simple math, the hard work comes in getting the problem to the point where you can just plug in numbers. This is the biggest sticking point for most of my intro students-- I've heard dozens of them say "I can do the math just fine, but..."
In the same way, most of the work in lawyering appears to be in framing the issues in such a way as to allow the application of precedents and principles. If it were all rote, we wouldn't need people to do the job-- a look-up table would suffice. The hard part is getting a complicated situation boiled down to where you can apply the rote learning.
- Log in to post comments
RE medical education...I can't speak for law school, but when I was in med school, most of the learning WAS in fact rote. Lots of memorization of muscles, bones, pathways, diseases, but relatively little that actually required critical thinking skills. It was tedious but compared to PChem in undergraduate school, a piece of cake. The challenge occurs after you graduate, and you find people expect you to be able to evaluate clinical evidence and make diagnoses. Med school is about background, residency is about reality.
3% of students who take introductory physics take another course in the discipline.
Let us not forget that some form of introductory physics is required for pre-meds. Additionally, it's required for engineers. Generally, two semesters are required, but those two semesters make up a single "course" (at least to my thinking).
If those folks are counted in these statistics, then it's no surprise to me. The vast majority of individual classroom Physics students at Vandy are engineers and pre-meds; those two populations overwhelm by a factor of 4 or 5 the population of "science distribution course" physics students. (If you throw in astronomy, then they only win by a factor of 3.) The "science distribution course" physics students at Vandy, in turn, are a factor of 4 or 5 more numerous than physics majors.
Two factors of 5 give you 4%, so my off-the-head numbers are consistent with the 3% you cite. I'm not sure how alarming it is, because engineers know they want to be engineers, and as such have to take physics. Similar, although less rigid, for pre-meds. (We get the very occasional Physics major pre-med, but they're not common enough to be statistically important.)
-Rob
factoid I picked up at an AAPT meeting a few years back is that only something like 3% of students who take introductory physics take another course in the discipline.
What are the numbers for other disciplines? I only took one physics course, but I also only took one botany course, one sociology course, and one classics course.
Let us not forget that some form of introductory physics is required for pre-meds. Additionally, it's required for engineers. Generally, two semesters are required, but those two semesters make up a single "course" (at least to my thinking).
This was at a conference on teaching introductory calculus-based physics, so I believe they were only counting the calculus classes in that figure. Most schools offer an algebra-based class for the pre-meds.
Engineers are still counted in that figure, though.
And yes, it's more or less consistent with our historical enrollments-- we teach something like 100-130 students in our intro mechanics course, and have averaged something in the neighborhood of 5 majors over the last N years. More recently, we've had a some very large classes (8 this year, 13 next year, 10 the year after), but it remains to be seen if that's a blip or a trend.
(Personally, I like to attribute it to the dynamic young faculty, but it might just be the statistics of small numbers...)
Well, I am a scientist, and I'm seriously thinking of bailing out altogether. Forget riches, the lack of a dependable income is killing me, mentally. Living from grant to grant, never knowing if you're going to get a paycheck in six months' time, was fine when I was in my twenties, in grad school and single. Now I'm older, no longer single and no longer able or willing to change countries for the next project or temporary job.
There used to be a time when a post-doc was two-three years or so, between your degree and finding yourself a permanent position. No longer. Today ten years or longer seems to be quite normal - and that is at a time when people generally want to settle down. And it seems more and more people never find a permanent position in their field at all. Most of my former grad student colleagues have a steady position and are working within their field of research - but few of them are actually doing research anymore. Doing usability testing on mobile phone designs is good, steady employment and very interesting, but it's not research, not even if you have a PhD in the field.
Is it that strange to know what you want to do? I don't remember when I decided I wanted to be a physicist because I never rember wanting to be anything else. Granted I started out wanting to do astro theory and now seem to be going heavily towards small scale experiment but....
I also have the opposite problem described above, the physical principles make sense...I just don't get along with the math (hence the change of heart)
On the other hand I'm still an idealistic undergrad (senior this year)
Chad: " A career in research science can be very rewarding, but the rewards are not necessarily financial (though I hasten to add, I'm not making a bad living, either)."
As I understand it, the normal fate, at the top universities, is for 1/3 of new Ph.D.'s to get a tenure track position, 1/3 to get a post-doc, and 1/3 to well,....
The end result is that, if you've persevered to a Ph.D. in one of the top programs in the world, is that you might be an apprentice professor, with a 50% chance of becoming one, or you might be a perpetual one-year lab temp, always working in somebody else's lab, on *their* research, or you might be SOL, and envying the post-docs.
And that's for excellent programs. If you've gotten a Ph.D. from a merely good program, then you're SOL to start with.
In any other field than academia, nobody would be surprised to see a lot of people deciding that that's a sucker's deal. It's a mark of modern academia that such a bad deal is not immediately considered to be the prima facie (sp?) explanation for the later drop-off. I'd credit poor math skills and stadium-sized intro courses for the first two years of undergrad drop-off.
Barry, do note that with a PhD in any decent science program you do have lots of very good career opportunities, just not so many where you get paid doing basic research. Also note that a fair amount of PhDs are going into this with open eyes, not aiming for (or wanting) a research position afterwards.
Doing a PhD is not a suckers game. The simple truth is that if you love doing research, those years as a PhD student are probably among the best you'll have no matter what career you end up having afterwards. If you love your subject it's well worth it just for those years, no matter what comes afterward.
Barry,
As I understand it, the normal fate, at the top universities, is for 1/3 of new Ph.D.'s to get a tenure track position, 1/3 to get a post-doc, and 1/3 to well,....
This isn't correct. In most of the sciences, *everyone* does a postdoc if they're staying in the business. In physics, for example, the numbers might be:
2/3: one or two postdocs, and then either tenure-track faculty (1/3), permanent position doing science at a big lab or in private industry (1/3), or leaving science to do finance, management consulting, or something else (1/3).
1/3: leave academia, either to do permanent position doing science at a big lab, work in private industry, finance, management consulting, or something else...
Realistic percentages of getting faculty job are quite different from 1/3.
Yes, a large fraction of folks staying in science does postdoc. In fact MOST do it - even industrial research labs require postdoc experience. A solid fraction of students aiming for "research" academic positions might do two postdocs.
But chances of getting that tenure-track position are quite a bit slimmer than 1/3. Only about one in three universities is hiring in any given year (see CM-AMO rumor mill for more statistics). Some may hire two people, some will end up not hiring at all or hiring a senior person away from another school, but most will end up with a single candidate.
So if you are aiming at top 50 school (a much further drop-off translates into some exponentially increased struggles to get reasonable start-up package, grad student/postdoc helpers and any substantial research funding) - there may be only 15-20 positions available any given year. Not sure how many enrolling grad students are in the same top 50 schools a year, but I am guessing at least an order of magnitude more. Probably about 50 times more.
I would estimate that only about 2-5% of grad students end up as faculty. Another way to think about it - how many students go through an average faculty's group in his lifetime? How many postdocs? Since most departments aren't expanding at significant rates, the advisor/advisee lifetime ratio (about 1:50 - 1:100 in my experience) is also approximately how many people become faculty.
I am sure Chad will point out that one can be a faculty at schools not known for their research reputation and work with undergraduates, instead of grad students and still do top-notch cutting-edge science with rather limited resources. Perhaps. But if research is your passion, finding a permanent position as staff member in one of government laboratories (or whatever is left of Lucent/IBM/Hitachi/HP etc.) may be a far better choice.
As I understand it, the normal fate, at the top universities, is for 1/3 of new Ph.D.'s to get a tenure track position, 1/3 to get a post-doc, and 1/3 to well,....
I would say that even at the very top universities, the fraction of people going directly from the Ph.D. to a tenure-track job is well below 1/3. It does happen-- I know of something like two cases-- but most people do a post-doc, or go into a visiting position first.
The AIP has employment data to back this up-- in 2003, 69% of newly minted physics Ph.D.'s took post-docs, and something like 25% took "potentially permanent" jobs. They also have data on academic hires showing that 40% of new hires at Ph.D. granting institutions were previously post-docs, and 34% of new hires at undergrad-only colleges.
I would estimate that only about 2-5% of grad students end up as faculty. Another way to think about it - how many students go through an average faculty's group in his lifetime? How many postdocs? Since most departments aren't expanding at significant rates, the advisor/advisee lifetime ratio (about 1:50 - 1:100 in my experience) is also approximately how many people become faculty.
That ratio seems a bit high to me.
A faculty member who takes ten grad students at a time, each taking five years to graduate, would need a 50 year career to reach a 100:1 ratio. If you shorten that to a 30 year career, you would need 2-3 post-docs at a time, doing two-year stints, to make up the difference.
That seems like a lot. I know some groups that have that many grad students and post-docs, but that's on the high side, and they haven't been doing it for 30 years.
I am sure Chad will point out that one can be a faculty at schools not known for their research reputation and work with undergraduates, instead of grad students and still do top-notch cutting-edge science with rather limited resources. Perhaps. But if research is your passion, finding a permanent position as staff member in one of government laboratories (or whatever is left of Lucent/IBM/Hitachi/HP etc.) may be a far better choice.
Oh, sure. Those are fantastic jobs.
But I'd bet that there are fewer openings for staff scientists at national labs than there are tenure-track faculty jobs available in any given year.
Chad, I am a little afraid you will be offended at my characterization of liberal arts colleges as a job where job responsibilities are tilted heavily towards teaching, rather than research, but maybe you could weigh on this with actual experience instead of listening to hacks like me who have no idea what they are talking about. :)
At least this is a common pre-(mis-?)conception that I happen to share.
As to the ratio - some people happen to have large group and are easily in 50-100 range. In fact a university I am associated with just celebrated 100th grad student of one of the faculty (he is quite old, naturally). But even if you throw out my number and adopt something on the order of 20-30 students, that would fall not too far from my estimate of 2-5% or whatever I said.
Well, actually maybe I have to correct those numbers after all. I just looked at NRC statistics from 1995 and it appears that the average faculty/student ratio at top 50 schools is 3.55 (actual number, not made up or anything). Assuming an average faculty career lasts about 5-6 times the average time it takes to get PhD, each advisor will go through ~17-21 students on average. So I guess I may have been weighing active group more than "lonely" research professors. Perhaps theorists or other fields have smaller groups, I don't know.
So perhaps revised number should be in 5%-6% vicinity, but it's still far fall from 1/3. Number of people getting faculty offers at research schools straight out of grad schools bypassing a postdoc is a rather small fraction, especially in experimental fields. Maybe one out of 20-30 offers.
Possibilities in government research labs are not as limited as some may think. Obviously strongly depends on your field. But it's substantially less competitive than faculty positions, and offers nearly parallel tenure-track-like career option. The negatives is lack of grad student slave labor, less prestige and a lot of red tape. The positives is lack of teaching/committee responsibilities (some may consider these as negatives), more secure long-term funding/access to facilities, more collaborative research effort, more reasonable hours.
The industrial labs (like Lucent) used to offer great opportunities too. Sort of like government labs with more freedom to do top research, a lot more money. No wonder faculty positions of top schools are full of former Bell Labs/Lucent scientists, especially in condensed matter/lasters/materials/electronics research. Not just experiment but theory too.
I have a bit of recent perspective on leaving physics after one introductory course because I just did this last year. More than that, I thought I was going to be a physics major when I first got to college. Even more than that, I quite enjoyed my first course. I left (to math and philosophy) for 3 main reasons:
1) I'm a thorough-going theory guy, and looking at the current hep world of research I was thoroughly uninspired. The string theory domination just wasn't that appealing- not enough experimental evidence, lots of complication with not a whole lot of apparent guiding principles, not a lot of progress (at least visible to an undergrad) in the last 30 years, etc. LHC could change a lot of that, but from my uneducated perspective it just looked/looks like a horrible mess.
2) Physics discussion sections were horrible. This was supposed to be a high level course at an elite institution, but the sections were taught by uninspired and untrained grad students and covered the most boring things imaginable for an excrutiating 90 minutes.
3) The next course on the physics track was E&M with a lab- not my cup of tea.
I still love physics and hope to continue learning it in some manner, but the (hep theory) picture just looks very bleak right now.
1) I'm a thorough-going theory guy, and looking at the current hep world of research I was thoroughly uninspired. The string theory domination just wasn't that appealing- not enough experimental evidence, lots of complication with not a whole lot of apparent guiding principles, not a lot of progress (at least visible to an undergrad) in the last 30 years, etc. LHC could change a lot of that, but from my uneducated perspective it just looked/looks like a horrible mess.
I would probably say that leaving after one year because you don't like the look of the research landscape might be premature. Bear in mind, you've got at least four years before you'd really be actively working in research (as a grad student)-- a lot can change in four years. (And, really, if you include grad school, you're looking at eight to ten years before you'd need to make a career of any of this.)
2) Physics discussion sections were horrible. This was supposed to be a high level course at an elite institution, but the sections were taught by uninspired and untrained grad students and covered the most boring things imaginable for an excrutiating 90 minutes.
This is a major problem in physics, and probably the main factor in driving bright people out of the field. A lot of intro classes are really badly run, and I think that hurts the field.
(Special Bonus Fall Preview: I'll be teaching our Integrated Math and Physics intro class this fall, which is a different spin on classical mechanics, team-taught with someone from the Math department. Expect this to be the subejct of several future blog posts...)
3) The next course on the physics track was E&M with a lab- not my cup of tea.
If you're into theory, E&M is where you start getting the real cool stuff-- fields and potentials and symmetry and all that really start to be important in E&M, in a way that you just don't get with Newtonian physics.
Of course, if you don't like lab, that's a big stumbling block. For me, the lab was the highlight of that class...
Of course, I have no chance of talking you out of leaving physics, and, really, you should do whatever you want. College shouldn't be about slogging through unpleasant classes that you don't like for careerist ends. Unless you're pre-med, of course.
You all talk about the postdoc world, but wasn't the question as to why people don't major in UNDERGRADUATE science.
Jobs jobs jobs. Not just riches but any job.
Students these days have a wealth of information available to them that just wasn't there 10 or 15 years ago. The fact that most students will graduate with a huge amount of debt from college they need to have the feeling that there is some sort of job they will be able to get with their degree.
Lets face it, it is a simple question of economics at work. If there were jobs available for science majors more people would major in science. But if you look at the extended period of time that even PHd s spend looking for jobs stuck in low paying post doc positions you can see there is a SURPLUS of scientists not a shortage.
I don't believe they even have to be particullarly good jobs. Hotel management seems to be very popular and the average starting wage there is something like $36,000.
Truth be told if I could go back in time I would have taken your advice at the top and gone into accounting instead of my useless Molecular Biology degree.