I was an EE major in college at a good school and I can tell you that it is a long, hard grind. I never had a problem with the math/physics pre-reqs. In fact those were the fun classes.
It was the soph/jr/sr years where it was very difficult to tie what you are learning to anything real and concrete. The worst part is that you are being asked to remember facts/processes/patterns for a test, then you rarely have to go through them again. So you end up forgetting most of what you learned. Since you didn't know what it was used for in the real world, you end up forgetting about something you didn't even care for in the first place.
The courses I enjoyed, I really enjoyed. At a school on the quarters system, you end up taking about 50 total courses and I enjoyed < 10 of these. My junior year I built a 4-bit processor on a breadboard using discrete parts for everything but the control system (fpga). This was one of my best memories because I understood the real-world applications of a processor and I got to sit in a lab and build it myself.
It's a broken system. It's a long grind thats not effective at getting you up to speed as a professional or getting you interested in your career. Its also terribly inefficient at improving your problem solving and analytical thinking. It still improves your problem solving/analytical thinking through brute force and I think thats a necessary process, but there has to be a better way.
If there was a way to combine the rigors of academic courses (to improve analytic thinking/problem solving ability) with the hands-on experience of vocational learning, I would be all for that. It is more applicable to software than some of the more traditional engineering disciplines, but I think it could work for all of them.
(admittedly my situation was poor because I chose the wrong major. I was an EE who was taking all CE electives and I moved into software after I graduated)
As for not being able to graduate enough STEM majors, ffs let's drop the jingoism and bring in more foreign students.
It was the soph/jr/sr years where it was very difficult to tie what you are learning to anything real and concrete. The worst part is that you are being asked to remember facts/processes/patterns for a test, then you rarely have to go through them again. So you end up forgetting most of what you learned. Since you didn't know what it was used for in the real world, you end up forgetting about something you didn't even care for in the first place.
SRS are your friend. Remember, you will forget everything you will ever know given sufficient time. It's best to exploit study technique that take into account spaced repetition. Your skills and all the things you learn are like a muscle, either you use it, or you lose it.
Right now, I have minuscule amount of haskell knowledge loaded, ready to use when I am ready to learn or code something in Haskell. Once I learn something, I add new stuff to my Haskell deck.
The reason I advocate vocational style education is that it is inherently similar to SRS. Academic coursework is not, and students won't go out of their way to continue learning from previous quarters/semesters (although this is where CS is slightly different than other engineering disciplines, since you can program for fun in your spare time and it may correspond with what you have learned).
Flashcards Deluxe by OrangeorApple is a great one, not only because of its great use of SRS and fantastic interface, but also because it has the massive and active Quizlet library behind it: http://itunes.apple.com/us/app/flashcards-deluxe/id307840670...
I've used it to great success in a few classes in the last year. You really have to be dedicated in the first few weeks though, and not skip a day in your SRS routine. That first bit of grind really pays off down the line though, as you have to spend less and less time studying from then on. Skip though, and it'll come back to bite you.
I'm curious to hear more about your experience using that kind of system with programming knowledge. When you say "Haskell deck" are you talking about a literal "deck" of flash card type things? (whether physical cards or part of some Anki-like software)
I'm curious to hear more about your experience using that kind of system with programming knowledge. When you say "Haskell deck" are you talking about a literal "deck" of flash card type things? (whether physical cards or part of some Anki-like software)
Yes, they are merely flash cards.
I am at the point in which my programming decks are minimally useful, mostly because I have not memorize much to begin with. I think I was able to use them a few time in my day to day programming task. With a sufficiently sized deck, I would derive more uses out of them since there are more memory to draw upon.
"If there was a way to combine the rigors of academic courses (to improve analytic thinking/problem solving ability) with the hands-on experience of vocational learning, I would be all for that."
This is the goal of lab/recitation. Taking the basic concepts in class and applying them to something you physical build/simulate to get the 'feel' of that design. Rather than making these as 'cookbooks' they should be exercises similar the homework to have a love triangle between hands on learning, lecture, and problem sets.
There's a simple solution: stop subsidizing useless degrees.
If someone wants to pay money to spend four years of their life studying Gender Studies or Semiotics then good luck to 'em, but make sure they're paying the full cost of their degree, preferably up-front. Underwater basket-weaving is a fine hobby for the idle rich.
I they want any aspect of their education subsidized, they should be studying something both useful and hard. Every university degree should include serious mathematical content.
What about the students who aren't good at mathematics? They're probably not benefiting from a university education anyway. Let 'em go get the low-level white-collar drudge jobs that they'd wind up in after graduation anyway, and let 'em do it at the age of eighteen and start building up some wealth rather than forcing 'em to wait 'til they're 22 with a net worth in the negative six figures.
This, of course, will never happen. But it's nice to think about.
I want to agree with your point, but I strongly suspect that universities are much more capable of price discrimination than the government is, and that close to 100% of subsidies are captured by tuition increases.
Put it this way: if Obama waved a magic wand and said "Twelve months from now, anyone majored in engineering gets $100k from the Treasury. No questions asked.", I think you'd see approximately every college in the country debut a tiered pricing structure by next semester.
Incidentally, I'm not just being cynical here: most selective universities have official policies where a) they bid for desirable students with grant aid (academically inclined students, desirable minorities, etc) but b) external scholarships decrease grant aid, typically at a 50% ratio. If someone external to the university subsidizes your education, the university acts to capture the subsidy, justified entirely by "Because we can."
I paid for a good portion of my college education with external scholarships. Each time I scrounged up another $2,000, the university financial aid department said "Great! We'll decrease your loans by $1,000 and decrease your grant aid by $1,000." (My mother aggressively negotiated with them on this, every time, and got the clawback rescinded "as a one-time exception" something like eight times. I thought this was embarrassing at the time. I plead youth and stupidity on that.)
As a proud owner of a physics masters degree, I say that this is a load of elitist, prejudiced, confused nonsense.
> This, of course, will never happen. But it's nice to think about.
I strongly disagree. It's horrible.
Mathematics isn't the only 'hard' subject. What is hard to one person may be easy to another. I'm absolutely sure that you haven't even tried Semiotics or Gender Studies, and I'm also sure that the skills needed to study in those fields wouldn't come as naturally to you as mathematics does.
Also, just because you don't see the relevance in those fields, it doesn't mean that there is no relevance. Gender Studies, for example, has a place in sociology, psychology, politics.
And of course, wealth is not the ultimate objective of everyone. My peers who wanted to persue a scientific career certainly aren't loaded. In fact, those with art degrees made more contacts at university and many have managed to land well paid jobs.
Gender Studies (along with Critical Theory, Intersectionality, etc) are some of my non-programming educational pursuits, and I agree. That stuff is hard. And not just because of the density of the language, a lot of it is conceptually very difficult to grasp, especially for those whom it does not viscerally effect.
Maybe engineers should be required to take an intense, no-hands-held Gender Studies class, if the other departments are required to take difficult math classes.
I walked out of university with a degree in engineering (CS), conversational fluency in Japanese, and a degree in cultural studies. When I think of moments of difficulty during my education, "Women can be black and blacks can be women, too, which complicates things, because the experience of black women is not the same as the experience of black men or white women. This is ignored by a lot of writing about the experience of being 'black' or being 'women.' That is bad." is not one of them. (This is the thirty-second version of Intersectionality.)
I think Classics or Comp Lit might be more daunting than Gender Studies.
That is, a field of study with a nigh-bottomless reading list. The more you can read, and the deeper and more varied your cultural vocabulary, the better your essays will be. A field with clear advantages for ability to read or speak multiple languages. That is, if you want to excel.
Some of the commenters here have clearly never taken a humanities class with a challenging reading component.
One data point: I was the sole major in Comparative Literature at my undergrad college. At the time, the department had a requirement that you do four higher-level literature courses in two languages that were not English. That meant that you had to reach fluency in two (non-English) languages by the start of your third year. Nobody did it, and eventually (years after I left) the department began allowing people to get Comp Lit degrees by reading works solely in translation (something my professors found hideous).
One nice thing about being the sole graduate in your subject? You win all the university and/or departmental awards in that subject by default. Looks impressive (to outsiders) but feels awkward.
For what it's worth, my two languages were Ancient Greek and Latin (I went on to study Classics in graduate school), so apparently, you wrote your comment just for me.
I knew a German lit student and looked at his summer reading list (to prepare for grad school). It contained about 100 books, and he aimed to grind through a substantial fraction of them that summer. I read a lot (back then), but not at that level.
That was one of the experiences, besides the time I took French and had my ass handed to me by people who were also simultaneously studying other languages, that brought home to me the linguistic workload that can be part of a humanities degree.
Gender Studies (along with Critical Theory, Intersectionality, etc) are some of my non-programming educational pursuits, and I agree. That stuff is hard. And not just because of the density of the language, a lot of it is conceptually very difficult to grasp, especially for those whom it does not viscerally effect.
No, because it's bullshit. Bullshit can be made arbitrarily hard to understand, and it's generally productive to do so if you're an academic trying to justify your own existence.
That's, unfortunately, a common reply by people who simply dislike some areas, and refuse to believe that not everything they don't understand the purpose of immediately is crap. I even used to be one of those holier-than-thou, "this is pomo obscurantism!" insufferable science undergrads myself. Of course, like most, I hadn't spent any real effort researching the subject before reaching such a "conclusion" (conclusion in quotes because I didn't really investigate the matter, just picking up a few quotes from Chomsky and Sokal or whatever)...
I actually still dislike lots of it, especially the secondary literature (which is, I think, full of garbage), but it doesn't take that much reading in the history of philosophy to gain a basic understanding of what problems 20th-century continental philosophy was responding to, or to realize that they weren't all insincere idiots who somehow pulled the wool over everyone's eyes for 80 years.
I'm not talking about dense language. Actually, there are some pretty brilliant orators and writers who take these ideas and put them in pretty accessible language. But the ideas are still difficult. They lead in increasingly complex directions in terms of how we got here, where we are, and what to do.
I recognize that unnecessarily obscure language is a hallmark of academia and socio-political thought. I think tomes like Negri's "Empire" wallow in it. But what I'm referring to is what happens when you get past that, either as the reader, or as the theorist.
You should spend some time with some serious, high-level astrology. I am not kidding. You will find that there have been brilliant defenses of it over the centuries, and it is full of various tidbits of ancient wisdom from the wisdom literatures that you may find quite profound.
It is, nevertheless, bullshit, because it can not be meaningfully mapped back to reality. Its predictions are bunk, the framework it offers you doesn't work when pulled into the real world, if taken truly seriously it decreases your ability to make accurate predictions about the world and the people in it. People had centuries to polish the turd, even to find painfully clever ways of including little tidbits of non-turd in the turd, but it's still a turd at its core.
The presence of a few good tidbits in a discipline proves nothing about it. Indeed, one would be hard-pressed to define something totally false in every detail. The question to be asked is how much truth the discipline has, and how well it can not merely assert the truth, but demonstrate it in a non-self-referential manner. Pretty much every social science in the modern era fails this test miserably.
Ideas that sound profound are a commodity, truth is harder to come by.
That's only looking at it through one particular lens. The cultural value and historical context of these writings are useful to explore. Also, as a forerunner of modern astronomy, astrologers' observations were incredibly accurate and are still relevant to the field.
Liberal arts degrees subsidize the hard science and engineering degrees since the per-credit-hour tuition is the same but the guy doing gender studies is way cheaper than the girl using the microbiology lab.
This is generally not true; the overhead charges from grants and research money just about pay for the added cost of the (much) higher faculty salaries, travel, research equipment, and the increasingly onerous grant application and accounting/auditing/reporting processes, but not for undergraduate education. In fact it's usually prohibited to cross-subsidize education with research grants, and funding agencies are surprisingly strict about this, doing things like auditing whether computers bought on a grant are used in teaching a course (not allowed).
In many cases grant overhead doesn't even pay for the cost of research; only the top-tier research universities are particularly successful at pulling in enough, and big enough, grants to do that (see this from Georgia Tech's former Computer Science dean: http://innovate-wwc.com/2011/05/18/if-you-have-to-ask-ten-su...).
The liberal arts, meanwhile, are so cheap to teach that in many cases they are actually making money, i.e. their expenses are lower than tuition fees, and the remaining tuition is used to cross-subsidize science education: http://www.today.ucla.edu/portal/ut/bottom-line-shows-humani...
STEM students subsidize liberal arts by creating a world in which somebody whose only skill is art history can live without starving to death or dying of a childhood disease
On a related note: a good alternative to student loans would be letting students sell a small percentage of their income over the next (say) 20-30 years in exchange for all or part of their college tuition. This would be lower-risk to students than loans, and give investors an incentive to help students get a lucrative education and a decently-paying job.
Sounds like Australia's system [1]: "This funding is in the form of loans that are not normal debts. They are repaid over time via a supplementary tax, using a sliding scale based on taxable income. As a consequence, loan repayments are only made when the former student has income to support the repayments."
There's a huge difference: the Australian system doesn't allow student loan investors (just the government?) to get more than the inflation-adjusted value of the interest-free loan. A percentage-of-earnings scheme, by not placing such a limit on the amount of money investors can make, would have two interesting effects:
1. Investors would have more incentive to help students succeed financially, since they get a cut of the profits.
2. Students who strike it rich would, in a competitive market, end up subsidizing everybody else. Think startups: the returns on VC's money come disproportionately from a few successes, and this lets them invest in a bunch of companies that probably won't be big hits, to get the few that will get big.
Who decides what's "useless," and on what timescale? Even within the STEM fields, should studying or researching string theory be subsidized? What about atomic physics? Astronomy? Pure math?
That is true at the graduate level. If you pursue a PhD in life science, physical science or computer science your tuition is waived and you (usually) get a stipend. If you want a PhD in liberal arts you have to pay tuition.
Are you just saying/guessing this? I have a PhD in Classics, and although I did my graduate work 20 years ago, I'm still in contact with people in that field. The standard is that for a five year program, nearly everyone can expect (1) to be a TA for 1-3 years (and receive pay that way), (2) to receive a writing fellowship (often from your university) for 1-2 years (and thus be able to receive pay while focusing on a dissertation) or (3) to hit the jackpot and receive an external grant for 1-5 years which spares you from some of the teaching duties in (1). In addition to that, everyone (except non-US citizens) has his or her formal tuition waived by some bizarre sleight-of-hand that the university does with itself.
Let me stress that since Rariel brings it up below: nobody pays tuition (again, except non-US citizens).
People take out loans for extra money (the grants and TA stipends ranged in the 9,000-12,000 dollar range per year in my time, 1992-1997), but nobody pays their tuition. One standard joke was that we were so much better off than the professional students (Business, Med School, Law School) because they did pay tution. I never paid a dime in tuition and finished graduate school with nothing more than mild (< $3,000) credit card debt.
I have no reason to believe that this system has changed. Do you have any sources for your claim?
Somewhat, but most liberal arts PhDs only get a small stipend from being a TA plus in-state tuition if it's a public University. Maybe what you're thinking about is being funded, that allows you to get your PhD essentially for free. I'm sure that is available for STEM as well?
There's an odd cultural aspect to it that I'm not sure is entirely the same as difficulty per se. Among some programs, at least, there's an almost masochistic love of all-nighters, long-as-hell problem sets, etc., etc., and if you're not into that culture, you'll probably feel like a foreigner.
Fortunately, where I went, that was common in engineering but not in CS, which had more of a "work smarter, not harder" attitude--- still quite a bit of work, and sometimes people pulled all-nighters, but there was a cultural difference in that people didn't see pulling all-nighters as a good in itself, some sort of hazing-esque badge of pride, but just something that, unfortunately, sometimes happened due to too much work, poor planning, or procrastination.
That's a bit different from the actual difficulty of the material; you can study difficult material without that kind of culture, and in the other direction, it's quite possible to grind someone down with piles of work even if they find the subject matter itself easy, depending on how you design courses and assignments.
I'm completely speculating, but I wonder if the bottleneck of math being hard is tied directly to how notoriously bad the standard of math in K-12 education is? I've always wondered how in the US the public school system is very weak (or at least people constantly complain about it being so) and yet they have some of the best technical colleges in the world (MIT, Stanford, Caltech, etc.) I've always wondered how students cope up with this big jump from high school to college.
I think that there is a very high variance in math education (and people) in public K-12 schools. My high school math consisted of advanced algebra, proof-based geometry, trig and a smattering of calculus. This seems to be typical for college-bound student across the country. I went to a second-tier technical university (CalPoly, San Luis Obispo) and I was well prepared for the typical first year physics and math classes. From what I can tell, the first-tier schools you cite have substantially similar first year math and science curriculum.
That said, I know people who went to the same school middle and high schools I went to, but who were never able to get past the introductory algebra class I took in 8'th grade. Sure, some people probably got better or worse teachers, but I believe some peoples brains are just better at math than others.
Then why does the US look so bad at math? I don't know. I suspect that both of these statements are true:
1. Many countries have better alternate education paths for people who aren't
that interested in college. People on these alternate paths are excluded from
foreign statistics, while people who ought to be on them dilute US statistics.
2. Doing poorly in school is more acceptable to Amercan parents than it is to
parents in many other developed countries. (Ironically, this may be because a
lack of alternate education paths led parents to becmoe disillusioned with the
value of school)
We home educate our kids and I can tell you the differences in math ability between each child.
One of my daughters can hit anything thrown at her. My other daughter is more advanced than most kids her age at math, but I know that she has already learned more math than she will ever use in her life (she is in eight grade but is doing very advanced algebra). Her interests will never lead her to a career that requires more than she already knows.
The problem is pushing everyone to one standard, when they do not need it. It makes them feel stupid, and is a waste of time that could be used developing other skills.
Current undergrad taking math courses at one of the colleges mentioned. I would say that what you suggested is definitely part of the problem. The other part is that even the good* programs in high school are very different in structure from university math courses. There are very few study skills developed because most of the focus is on remembering the testing algorithm and does not necessarily include the rigor required to fully understand mathematics.
*good in this context means AP level. This is not because AP is the correct measurement, it is because AP and the like is the measurement used to judge high schools
In high school, the track I was on ended with high school seniors taking calculus at the local engineering college. No AP classes were offered where I lived, but I'm convinced that one college level class did more for me than a half-dozen AP classes would have. Especially because it forced my high-school math teacher to think more about preparing students for college-level work.
> Especially because it forced my high-school math teacher to think more about preparing students for college-level work.
I think the intended preparation is actually really important. With AP the flow is Learn ch2 algorithm > learn ch3 algorithm>...> do well on test > not have to take calc in college. If you actually prepare a course for rigorous understanding, I suspect you might end up sacrificing your AP score somewhat for much greater gains later on. However, the system only cares about that short term result.
> Especially because it forced my high-school math teacher to think more about preparing students for college-level work.
I believe high school students who take college courses can be a great asset for their high school for this very reason.
As a junior in high school I took college algebra at a community college in the evenings. As a result of the conversations with my math teacher, she made many changes to the subjects she covered in her pre-calc and algebra 2 classes.
It's easy, the American universities are good because they attract plenty of foreign students coming from places where the universities may not be as good but where high schools are arguably much better.
Most of the students who make it to Caltech in a math program didn't depend on the K-12 "system" to teach them math. It's practically a requirement to do a lot of self-teaching outside of school.
I don't know about MIT and Stanford, but I went to Caltech, and while there are first-year courses with names like 'Calculus', they are categorically nothing like the first-year courses at UCs or CSUs.
To put things into perspective, there is zero AP credit offered at Caltech. The only way to skip classes is to pass Caltech's own placement exams. I took the first placement exam (after finishing the entire AP math curriculum, as well as the entire math curriculum at the local community college—through linear algebra, diff eqns, and multivariable calc), got a 96% on Caltech's placement exam, and was recommended to take an extra, remedial mathematics course in addition to the regular freshman load to help bring me up to speed.
As another (recent) Caltech grad, I'll take a stab this. The Caltech intro calc class is very heavily proof-based. We started with easy problems that had nothing to do with calculus like proving that 1/sqrt(3) is irrational, went into epsilon-delta proofs for limits, and continued on to showing that the Dirichlet function is discontinuous everywhere. I don't think we evaluated a single integral in that class.
A strange feature of the class was that the weeks where we studied derivatives were a lot harder than the weeks when we studied integrals. Although derivatives are a lot easier to evaluate than integrals, it's a lot harder to prove that they exist. Most of our problems were proving the existence of derivatives and integrals, not evaluating them, so it was a weird inversion in what is normally considered difficult in calculus.
If you want a sample of the problems, here's one of the problem sets from this year's class:
Oh, that's just elementary analysis. That's still "calculus" of course, but in most colleges you might not be expected to take that unless you were a math major. I can respect that.
(Note: I studied math as well as chemistry and CS at a state college so my background is a little unusual.)
There's still a reasonable number of public schools in middle class areas in the US that offer good educations. The interesting population is the eng/science/math majors at good colleges who didn't go to such great public or private high schools, how did they get leapfrogged? I remember some of my friends going to community college classes while in high school, math camps, programming comps, stuff like that.
Those program offerings are really important, they're like a parallel education track to public K-12. Now there's Khan Academy, the MIT and Stanford courses online, all kinds of resources, but you have to get motivated/gifted together with like minded ones.
The article specifically says this isn't the main issue--many of the students dropping out are the people with the highest SAT math scores, and a bevy of AP calculus/science classes under their belt. In theory, these should be our strongest students.
It's really simple. Smart people are not going to work 1000 times harder for less payoff.
Smart students with everything going for them get to make this decision:
(A) Dedicate their entire life at college to doing useless problem sets. No free time. No social life. No girlfriends. Spend all your time with stinky nerds drilling on useless sets. Then get out of school and make $80k doing the engineering slog. Max out your salary at $140k, 20 years down the road.
(B) Take easy classes while being a professional partier. At parties, have loads of sex and take lots of drugs while meeting many many people. Build up 1000+ Facebook friends and forge lifelong bonds with important well-connected people. Get your degree and use your connections to get hired for $80k. Have your employer then pay for your MBA. Max out your salary at $400k 20 years down the road, not including bonuses and side deal opportunities.
Phillip Greenspun was right. Unless you're a masochist or a social reject, engineering is for foreign students. A top American student from a good family has no incentive to do engineering: not lifestyle, not financial.
When it comes to software this is even more true because you don't even need an engineering degree to write software. Paul Graham has a philosophy degree and most of his wealth came from people skills: convincing Blackwell and Morris to hack for him, and later convincing young engineers to hack for him.
Steve Jobs vs Steve Wozniak is another perfect example. Woz did the work, Jobs got the glory. Woz stayed home nights hacking while Jobs was out fucking the prom queen and taking credit. Our society does not reward engineers so anyone with the ability and foresight to be a doctor, dentist, MBA, or clinical psychologist is not going to torture themselves.
Hacker News entrepreneurs are just a bunch of masochists who never learned that partying, taking drugs and having sex is more fun and pays better than writing Ruby apps.
FYI, medicine has a very bad expected value for time input. MBA and engineering are decent -- not great, but OK. The probability of making decent money is pretty good, while the probability of making a lot of money is small.
If you're smart (and I mean really smart), finance is generally a good bet. Once again, adjusted for time input it's not great but it does have a higher expected value.
You're still missing the real value, which is not the "degree" but rather the connections.
Just make friends with rich people and get them to hire you at their friend's giant corporation where their job is to undermine democracy and fool people. It's easy and pays well.
That’s a bit of an over-generalization. While there are individuals that can use the party-sex-drugs track as a networking opportunity into future lucrative careers, such individual are an exception, not the rule. Most students on the party-sex–drugs track are just enjoying their youth (and racking up student loans). I’ll agree that if you’re a young Machiavelli attending an elite school, then rubbing shoulders with your wealthy classmates can work out quite well.
> Paul Graham has a philosophy degree and most of his wealth came from people skills: convincing Blackwell and Morris to hack for him, and later convincing young engineers to hack for him.
id think having a deep enough knowledge on lisp to fill two books probably had something to do with it as well.
IMHO, the tl;dr here is "math is hard." I do feel like math is the sticking point. At my school, it wasn't intro programming courses that weeded out so many CS students, it was sophomore level discrete math. That class had a 50+% failure rate.
Edited to add: I'd like to pose a question. I've always felt that math divided people into two classes -- those for whom it was mostly easy (or at least approachable) and fun and those for whom it was inherently unpleasant. As someone from the first group, it's hard for me to relate to some of the ideas in the article, such as making classes feel more socially relevant. To me, that seems superfluous. My question is, has anyone experienced these sorts of approaches directly, and how did they work for you?
The tl;dr I got was different. What I got was a.) that our society has focused too much on the stimulation of our dopamine pathways. b.) Many professors forget that the enthusiasm they posess for a subject is not self evident.
The hardness of STEM was orthogonal to this piece; the focus was on how pointless the seemingly unapplicable dry list of equations and facts were to many students. This made it hard to continue the perceived self flagellation to little gain. They gave an example of a smart kid with an 800 SAT score in math who quit to psychology because his mechanics class was dull in comparison to the more active psych class. They also mention that classes like english give better grades than chemistry and math so people switch cause they feel they are doing better.
Their prescription is more interesting and interactive classes. That is a good idea as long as the essence is not lost to fluff and education does not degenerate to an exercise in marketing I guess.
But if the more important problem is diminution of student capacity / capability for the work, better teaching helps only so much. Anecdata like "helicopter parenting" suggest more kids lack emotional preparation hard work and differentiating curricula. I certainly don't see any sign that K-12 has gotten _better_ at math preparation.
Then there is the premium attached to integration into social networks. Wall Street bankers are scarfing down large chunks of the national income, more from government guarantee of leverage than productivity. Prices are signalling to focus on placement in the social network, rather than on productive work; and once there, to get good at predicting / managing regulatory intervention. The interest in studies orienting people to the political and social milieu shouldn't be surprising. Especially when these offer less risk of marked out as a loser by poor grades.
There is a premium for government work, too. It's less remunerative, but more secure, and has very high status, in many circles. The effect carries over to nearby careers like law, and journalism. Median income, housing prices, and unemployment rates in greater Washington D.C. all suggest that the worldly payoff for public service isn't bad.
So I worry that we're training less capable people, and encouraging them to think more about where they work than how they work. It can't go on forever, but a lot of folks will get hurt as it ends.
It does seem like our society is too focused on making learning "fun." While I wholeheartedly support design projects, especially if they're based on real-world problems and can give students extra experience beyond typical class assignments, I get the impression that sometimes they're offered so that students can get the whiz-bang effect from otherwise dry material. We too often expect to be entertained.
It sounds a bit like it's also the shock of encountering such hard math.
I'm jealous of other countries that work their high school students hard. I remember meeting international students that had learned all of my engineering college math in high school. Back then, neither me nor my parents knew that I could and that I _should_ absorb that kind of math, let alone how to go about doing it. It would be nice if that notion was more institutionalized in the USA.
As someone doing a second master's degree in a mathematical field, I can't say I've found it 'mostly easy', or even mostly approachable. It can be fun once you've developed strong intuitions for the material, but often a hard slog to get there.
Those who give the impression of finding it easy, I reckon tend to be either:
(a) freakish geniuses of the rare-even-in-academia variety
(b) modestly hiding the fact they've actually spent hours and hours of hard study to push past a series of brick walls in order to get here
(c) both the above
(d) overconfident and heading for a slip-up
(e) not challenging themselves enough.
"Mostly easy" was probably a bad way of putting it. Perhaps a better way would be that, at the outset, some students can handle mathematical challenges, even when they require hard work and perseverance; after a while, a positive feedback loop ensues, where being good at math makes it more enjoyable and finding it enjoyable makes you better at it. (That doesn't imply you'll never hit roadblocks.)
If, early on, math is painful and frustrating (perhaps through no fault of your own), then that cycle never develops and you form an aversion to the subject.
I hated hated hated math through to high school Pre-calculus. Algebra was a terror. I could get the concepts easily enough but I'd always make some mistake in the calculation.
Calc I and II in university were difficult but fun. There was a lot of problem solving and very little tedious calculation.
I switched from a chemistry major not because it was difficult but because it was tedious and uninteresting.
Not the classes. The lab.
And as an 18-year-old I thought the lab was simulating what my future career in chemistry would be like. And so I switched to a liberal arts major.
I don't think it's simply that math is hard. Many of these courses are taught in ways that make it hard for students to see how the tools they're using are going to be useful.
Students would be better-motivated to take the challenge if they were applying it to problems that seem relevant to them. In CS, for instance, which is more fun: slogging through an algorithms book front to back, or researching the best way to solve a problem you face in one of your own code projects?
To answer your question about relevance: I made it through my undergraduate physics major not because lectures and homeworks were fun, but because I got involved in research early on. I found it much more fun to learn in the context of a project I felt a sense of ownership of.
At my school, it wasn't intro programming courses that weeded out so many CS students, it was sophomore level discrete math. That class had a 50+% failure rate.
In classes that have brand new or complex topics the teacher matters a lot. Intro programming and discrete math 1 were the weed out classes at my school. Many people failed both. I managed to get an A in the intro programming only because it was all old material to me. The now retired teacher was horrible.
I go an F in the discrete math course. On the last day of class the teacher admitted to being horrible and then failed everyone. The weirdest thing was I actually felt like I understood the material, but was lost on every test. I had alway done well in math until that class with that teacher. When I retook the class with a different teacher I got an A. The same goes for discrete 2 and every other math course I took in college.
Now, to your question. I don't think the classes need to be more socially relevant, but teachers need to figure out ways to connect the material with students. The problem is you need a teacher who deeply understands the material and cares enough to figure out various ways of teaching it. A good teacher is akin to a good artist. They can take the same base concepts and materials that every other teacher receives, but can weave them into a coherent lesson.
In the end, I don't think absolutely everybody will be able to do the heavy math degrees, but the horrifying state of math education does bear a significant share of the responsibility as well. We have sucked all the life out of mathematics in the name of pedagogy and it's a bit much to expect that to pass by without effect. We can't accommodate all the student's desires for easier degrees, but the students complaining about looking at years of abstract theory without grounding are fundamentally correct, too.
I would argue the opposite problem: those who are interested in math at the college level are forced to sit through two years of applications-based mathematics before ever touching the theory behind it. Nearly all of higher-level undergrad mathematics is hidden behind courses designed to teach engineers problem solving techniques using calculus. As a result, the intro calc courses don't do a particularly good job of imparting either theory or applications, unless you have an unusually talented professor. I came out of Calc II being able to solve some nasty integrals only to realize that you really only need a couple simple techniques, and with no idea how to set up problems until I went through a few physics courses. What was the point of all that - we could have learned integration by parts and u-substitution in a two week session and moved on, or actually spent a semester of analytical work preparing for higher level math.
As a senior in CS and math, I completely agree. Calc II, III, elementary differential equations, and matrix theory classes are about 85% engineers at my university. As such, theory was not the focus. All I had to do to get an A was learn an algorithm and apply it. Most of my friends in CS seem to think that higher level math courses are the same and if that is the general assumption I don't blame people for not pursuing math degrees.
I wasn't sure about flatline's post, but I'm sure about yours: You should read the PDF I posted. You certainly aren't disagreeing with me (or Lockheart).
My pet theory is that math is not taught well because teachers for the 6-12 group just aren't good at math. They may be good at pedagogy but may lack the deep understanding that may be needed to teach more effectively.
Now can society can divert a bigger proportion of the few people that somehow got it to teach kids?
Some teachers don't even try, my Grade 11 Principles of Math teacher just took us on field trips to the park every day instead of teaching because she didn't have enough balls to say no to the kids who asked if we could go to the park. It made my Grade 12 year and first college math class a bit difficult at some points.
I completely agree with this. I went to one of the top 10 private high schools in Illinois and I was stuck with a math teacher, for two years, who's job it was to make sure no one got less than a C average. Understanding the material was not a primary concern, making sure no one had an abysmal average was. We had a 99% college matriculation rate to preserve after all.
If this is the attitude in an elite school, I can only imagine what it is like in the public school system.
"At my school, it wasn't intro programming courses that weeded out so many CS students, it was sophomore level discrete math. That class had a 50+% failure rate."
At my school, the CS program was designed so that the Intro to Programming class and the Calc I class were filter functions: students had to get at least XX% in one and YY% in the other in order to be admitted to the CS program (and thus be permitted to enroll in the upper level courses required for the degree). As at your school, more students bombed the Calc class than the CS class.
There was no effort made at disguising the filtering function of these courses - in fact, when program enrollment was high, XX and YY increased, and when program enrollment fell around 2004-2005 they decreased (at one point YY was 50%, effectively meaning you simply had to pass the class).
The tl;dr I got was that high school didn't adequately prepare students for college. Just to have h.s. kids enthusiastic (as someone else mentions in the comments here, high on dopamine) they make things entertaining/easy. While some may argue that college level should have more real world applications, I would argue that high school level should have more hard challenges that fail half the class. Maybe have a class that doesn't count towards your GPA that is super-challenging.
There's "culture shock" of being comfortable within the high school world where there's a lot of busy work but it's easy to get an A in any class as long as you put in the hours and can memorize a few things, even if you don't honestly understand the material. And then getting tossed in the deep end of hard-science in college where you have to put in enough hours to actually understand the material, not just do the busy work. And you have to be able to learn in class and learn on your own as well, because there's quite a lot of material to go through. For some people this is like going from George Jetson's job to digging ditches, and it's too much to adjust to.
Relatedly, high schools have been doing a piss-poor job of actually instilling mastery of basic skills like math, reading, and writing so a lot of students will come into college at a disadvantage. When you have, realistically, about one or two years (if not more) of remedial education that you need to get through before you're able to grapple with the actual college level material then college because a lot more of a slog (and potentially a very much more expensive slog) than it might be otherwise. If you don't have calculus, trig, and technical critical reading and writing skills nailed by the time you start tackling chemistry, physics, biology, math, etc. in college you will be at a severe disadvantage.
And finally, as others have pointed out there's the grade inflation differential. When your class work tends to be things like homework problems, lab assignments, and in-class quizzes where answers are either right or wrong then your scores tend to be directly numeric. Grade inflation is a lot easier in "softer" classes where most of the scores are subjective. Though that can, of course, vary depending on the school, but the average certainly seems to be that way.
Followup: additionally I think there's the general problem that most undergraduate degree programs just aren't that good. It's a lot easier to concentrate on rote memorization and "covering" a lot of material than actually instilling a substantive knowledge of a complex technical subject. The difference is that college can easily afford to suck at providing a quality education or experience. In K-12 if the students aren't graduating the school stops receiving funding. At the college level the difference in revenue between a student who graduates and one who drops out after 2 years is less than 50% (because 100 and 200 level classes are typically cheaper to teach). And if the student merely switches majors, the school still makes money.
There's a catch-22 of sorts. If you got a bad education in HS and you get into a good technical college program you'll be in way over your head. If you get into a bad college program you'll be facing a lot of boring work with little meaning. The best case scenario is to win the lottery and have gone to a good HS and get into a good college. But even then you'll still have a crap ton of hard work in front of you.
My experience was that of going from having to spend a bit of time memorizing useless material to having to spend a great deal of time memorizing useless material. For me the only "culture shock" was the realization that I no longer had time to learn anything on top of the busy work.
So I switched to a fine art degree and now I know more about engineering than 90% of engineers that I meet. Go figure.
The math-science death march. Now that is a perfect name for it. It's exactly what I lived through and why I changed my major halfway through college from chemical biology to art and technology. Why, you might ask?
- being pushed through a system that treats you like a number (In two years, I never met my advisor or received any guidance from faculty)
- having your entire college career already planned out with no wiggle room or a chance to explore other interests (until the very last semester where there was one free elective)
- professors who spoke English poorly
- classes that were taught straight from books with no added knowledge or insight
And the list goes on and on. For all those who managed to make it through, more power to you. But it wasn't until I switched majors that I started enjoying the academic aspect of college.
I "made it through" with a double major (CS and math), but the issues you bring up in your post do not really describe my experience with either department (though when I was looking around at universities during high school, I did write off those that looked like they would have such problems -- #2 is the easiest to detect).
Absolutely, this is also the reason why women aren't in technology. The more social and intuitive gender could see that for years of slaving away in high school/undergrad and even grad school, being at the lowest totem-pole of sexual and social hierarchy (from U.S perspective) for a average financial and professional payoff.
Obama wants to boost STEM; however the policy-makers and educators fail to understand the competitive disadvantage of "home-grown" industry vs. foreign ones; today, scores of Chinese and Indian PhD candidates and post-doc's are willing to take less pay and more hours for a shot for a green card in US and the respect of academia/engineering accorded to them in their culture.
Compare that to an average American engineering undergrad student, who on hand sees that his fine arts/humanities peers are getting laid and party more often, his law/business/medicine peers having greater potential financial payout and the opportunity to manage the engineers down the track - all the while he's suppose to be fretting all night long about problem sets. What a gyp!
Sad thing is US society used to reward engineers who invented things and businessman who actually created things instead of MBAs and marketing exec's - downfall of any great civilization comes when the focus of society turns from production to finance (see Holland, Roman Empire, Great Britain and US). There's still a few opportunities though in engineering to make good money, IMO just don't drink the startup culture kool-aid and always do your own thing.
Law school is actually extremely hard, there is no grade inflation in law school, if anything grades are deflated because there area maximum number of As (and B's) given out (usually around 10-15% is the max for A/A- and 20% for B's) for each and every class. So you end up with a bunch of liberal arts majors who have never gotten a C and suddenly they're getting a whole heaping lot of them. They get complexes.
I'm just one data point, but I did spend one semester in law school (at Columbia) before taking a leave of absence (in good academic standing, B's in everything) and enrolling in a PhD in engineering (Berkeley).
Law school was not hard. At an elite school, it's very easy to get B's in your courses, which won't get you on law review, but will generally land you a job at a top firm from an elite school. Failure rates are astoundingly low. I would guess 97+% of the class gets a B or higher, and failing out is almost unheard of (again, at elite schools). Lower ranked schools, from what I've heard, do grade on the harsher scale you mentioned.
My PhD program at Berkeley, on the other hand, was a horror show of attrition and failure. It was so much more brutal than law school it's completely silly to compare the two. My dept at Berkeley said that 40% fail to get the PhD, but that's not counting people like me who were awarded masters degrees and so are considered to have achieved their degree goal. I'd guess that the failure rate is well above 50% - and keep in mind, this is for a very elite program that is extremely selective.
Graduate programs in math, science, and engineering are littered with the broken dreams of exceptionally smart people. It's exceptionally unusual for med, law, and mba students to fail out at elite programs. It's commonplace in top PhD programs.
I would agree with you for the most part. Top schools ("T14" for us law nerds and "the Top 14 law schools" for the lay people!) are fairly easy to get B's in given effort is extended, but A's are still hard to get. I don't know many T14 schools that have curved and capped classes. I wasn't in an extremely lower ranked school (top 30) but it was a hardcore school in terms of the curve. They've since relaxed a bit (now A's are capped around 18% I believe). I do know people that went to Boalt and got D's so it's possible but failing out is pretty unheard of in law school, period. Also the legal job market has shifted dramatically in the past 4 years so those B students at high ranked schools can't bank on an offer from a big firm based on the name of their school anymore--I know too many who are unemployed for that to be the case.
Funny thing is my husband did his UG in ME at Berkeley and that department ( and the College of Engineering in general) was a sheer nightmare. The unimaginable things that took place and the workload he endured (while working 20 hours at LBNL) blew my mind and that was just a BS. I can imagine a PhD would be insane.
But my point wasn't that one is harder, just that law is hard and it's not an easy thing to grasp because they ask you to learn and be tested in a way that is very non-traditional, at least in my experience. It makes babies out of people who thought of school as a cake walk. It was actually funny at times to see the emotional breakdown over their first B or C because they were so dramatic about it.
But I commend you on your choice to leave law school, probably smartest thing you ever did! ha.
"Difficult" is a slippery concept for me. Is it more difficult to study to be a CS major or to be a CPA? Because I tell ya, even though I have a BS and MS in CS and a Math minor, the minute I cracked open my roommate's CPA book, I was in freefall.
It was something about the presentation and the sheer vast number of seemingly unrelated rules and regulations. I realized it took a special constitution to digest that stuff. I think I could have gotten it, but I would have had a lot harder time than I did learning about parsers or some other CS-y thing.
Completely agree what's hard is all relative... if there's an easier path with a bigger payout why wouldn't you choose that unless your passions point you in another direction.
My undergrad took this to the extreme. Our lowest level math class was Calc 1 and to get into the class you had to either take Calc AB in High School or show up to school a couple months early and take it before Freshman year started. I knew people that would fail out of our engineering program and go to another and get a 4.0.
Me personally, I had to unlearn all the bad crap my Physics teacher in High School taught me. It took an entire semester to get back on track but was well worth the hard work.
You can see how much harder Engineering and the Sciences are, from my own experience I thought getting my MBA was a breeze compared to studying engineering. Nearly everyone else struggled and worked their asses off, they had zero preparation for all the math and logic problem you had to work through. The solution is to force EVERY major to take a few REAL math classes and REAL science course so people have a solid foundation.
The title of this article is really misleading - it's not that "It's Just So Darn Hard" it's that the teaching of science grinds everything that's interesting about science out of the curriculum. That's the point that everyone in the actual article makes.
>>
But as Mr. Moniz sat in his mechanics class in 2009, he realized he had already had enough. “I was trying to memorize equations, and engineering’s all about the application, which they really didn’t teach too well,” he says. “It was just like, ‘Do these practice problems, then you’re on your own.’ ”
>>
Engineers build things. So engineering education should have a heavy component of building things - that's what's exciting and motivating to engineering students. The theory should be in the service of the practice, and most classes get this completely backwards. So they lose students, and grump that kids don't want to work hard.
Engineering is hard, if you are not that smart and cannot optimize problem solving. If you can't see the shortcuts (e.g. zero force members in Statics problems), it takes longer. I struggled.
This is how engineering education was explained to me:
I may be biased, but I actually prefer the most elusive 4th year labs being done in 1st year, as the projects to be built are much more easier and it is way more rewarding when a teacher hands you a specification sheet and tells you how to use (for example) an AND and OR gate with absolutely no mention of the project.
You can truly call the creation your own, and if you really get stuck you can go talk with the very helpful teacher (or other students) and they will lead you into the right direction without directly giving the answer.
The less support you have the harder it is. I have a graduate degree in engineering, and it was hard, but a lot of that came from having to face difficult challenges on my own. Classes are short, and the hours spent at home studying are long. Getting time with the professor is difficult, and his hours are limited.
We need to give students more support, don't let them stress out and drop out, we need to be there for them. Sitting at home for hours staring at a badly written text book is not productive and burns kids out.
Now how we do this... more teachers? higher costs? Can maybe technology be the key to lowering costs while increasing support? I'm interested in working on this problem.
My school has teachers in their office for most of the week willing to help you out with all the problems you have, and it certainly helps but it isn't enough to fix everything (15% of students who start the program I was in end up graduating)
I concur that this is a very significant issue. One of the things they did not mention was that the way the course loads are designed, it is very easy for a student to get burned out. For example, pre-med students often take, in their first semester of college, Biology, Chemistry, Calculus, and then usually some fine art to clear requirements. Its the same thing the next semester. Then you're onto physics, organic chemistry, higher level biology classes. These are all incredibly difficult courses that require hours and hours of reading, studying, and practicing. Many students get completely burned out and drop pre-med because of it.
If spending 30-60 hours/week on studies causes medical students to burn out, that's a good thing. It's much better for them to burn out during year 1 of (cheap) undergrad than in year 1 of (expensive) medical school.
I certainly agree. There is also a different kind of burnout that comes into effect. It is very hard to focus on all science, all of the time. So you get burned out because you can't spend your time learning or doing anything other than science (most of which is in no way interesting and is simply a requirement that you have to take).
It isn't that the math is too hard, it's that the students want to run before they learn to walk.
I'm a 17 year old in high school right now and having taken Calculus AB and BC (I and II) the teacher was constantly skipping material that I think was very important to have a general understanding of Calculus. Needless to say, I read the book to learn everything she was skipping, and I was also the only one who got a 5 on the AP exam. My classmates weren't stupid, they had just learned to solve very specific problems rather than actually learning Calculus. The very specific testing curricula hurt actual learning.
Another issue lies in the fact that (at least at my school) calculus and other high maths are seen as only for the very smartest kids. We need to change that. If students are required to have been exposed to calculus they won't be in as much shock when they get to college.
The availability of higher classes is a big problem as well. I was super lucky to find a public school program that allowed me to take classes with a community college, because otherwise there would have been no classes for me to take past sophomore year and Calculus I. I'm taking Multivariate Calculus next semester, and I doubt many high school students have this opportunity at all.
I realize that I'm in the minority here. Not many high school students have the opportunity to take Calculus III their senior year, and that really needs to change.
"I was trying to memorize equations, and engineering’s all about the application, which they really didn’t teach too well" - Matthew Moniz from the article
He's absolutely right, and no first year design project is going to fix this if the rest of the degree is structured the same as before. I think the bigger problem is simply that university is the wrong way to teach engineering.
I graduated 10 years ago with an EE degree (power systems major) and have worked in the field since, so I'm going to pick on EE here. During my degree, I was mainly taught theoretical models of electrical systems, interspersed with contrived lab experiments.
Very few 18 year old kids are going to appreciate the standard equivalent circuit of a transformer or a synchronous machine, less so more abstract things like Fortescue's symmetrical components. Because no kid has ever had much of a chance to look at power transformers or machines in service. Sure, they've seen power lines, but how many have seriously looked closely at them? Or maybe even asked themselves why there are strange, ceramic looking things connected to them?
So the first time I saw a real synchronous machine was in a lab. But looking back, it was really a contrived situation, with everything looking like it came out of a Bob the Builder toy set, alligator clips and all. You would never see a machine hooked up like that in real life... but only work experience has taught me that.
When I graduated, I was clueless. And in the intervening years since, I came across countless graduates just as clueless as I was. It's almost surely a systemic problem. I only met one graduate who really had a clue, and it turned out that he was a qualified electrician before he did his degree.
Which brings me back to my point - IMHO the pure university system is the wrong approach to train engineers. Their premise is that you learn the theory first, then apply it in context after you graduate (with some lame attempts at "practical" teaching in between). This works for some, but I'd wager that for most, it's a waste of time and you'll end up having to learn it all twice.
I would prefer to see a hybrid apprenticeship - university system, similar to the way you train tradespeople, but with more coursework components. The work is aligned with the study (or at least the student-apprentice gets exposure to real environments) and modules don't necessarily have to be done in a set order. It could potentially even be set up on a competency basis (like Western Governors University).
I don't know if this scheme would work, but I sure know that universities conceived for training academic researchers do not do such a good job of training practicing engineers.
I also think it's quite difficult for an 8th grader (I'll argue few develop a deep aptitude in math unless they start early) to understand why anything beyond basic math is important and worth learning. Further, I think we would be better off teaching the majority of students basic calculus and statistics in lieu of calculus. Get some stats and probability under your belt and you might even develop a clear motivation to learn calculus!
A co-worker, Asian-born, says that she knows an awful lot of ABC (American-born Chinese) kids majoring in business. This I think is not uncommon for groups either getting to the states or making their way into the professional classes. One generation does the hard stuff to get a job that pays well; the next generation notices that a) the business school is easier, and b) the MBAs get to lay off the BSEEs and send their jobs offshore.
As an MBA I can tell you, that the degree is not as highly valued as it once was. A STEM degree is much valuable even is rolls that were once filled by business types (some finance rolls, operations, etc.)
A MBA is a relatively easy Masters, but I think it's relevance is rapidly fading.
There are plenty of "soft skills" and it is good to have some academic work done in management, ops, marketing, finance, etc. However, this does not necessarily mean that this knowledge & these skills can't be acquired either on the job or through proper mentoring.
I'm not bashing my degree necessarily, it's just that there are too many MBAs @ the moment. In the end, if you are interested in management you can do better by reading Drucker than attending management courses, or read Porter & Ansoff (side note: Igor Ansoff is great example of what I am babbling about) if you are interested in marketing, etc.
Be careful. I once thought that the MBA degree was worthless, too. However I recently found that a friend of mine is earning 30% more than me at the same job, same company, because she has an MBA and I chose to get an MS Economics. We were both hired directly out of grad school with comparable work experience.
Whether or not the MBA teaches you anything, businesses still believe it does.
I looked at the graph, and it looked like only 2.4% of students are majoring in computer science. I would have thought that was much higher with all the buzz about social networks, startups and VC money falling from the skies. I guess I live in a bubble when it comes to that kind of news, at least to some extent.
I'm used to seeing students follow the money, like they did with engineering in the 1980's. I had to get a 3.5 GPA to get accepted to the EE program at my school. The school used GPA as it's valve to restrict students, which had it's drawbacks, but demand was so high for EE back then, they had to raise the GPA that high.
I also wonder if it isn't the calculus the keeps CS majors from completing, it's the recursion, pointers and such that really tied my brain in knots.
Likewise with EE, which was my major - I handled calculus, but eventually I ended up solving systems of equations with complex numbers and doing math beyond freshman calculus.
Of course, I gained problem solving skills that I use throughout my life and a career that's been great. I'm not sure how to better communicate those benefits to an 18-year-old in a way they'd understand.
Why would the legend of Facebook motivate someone to finish a degree in CS? Zuckerberg dropped out!
Steve Jobs dropped out, Bill Gates dropped out, I believe Steve Wozniak saved himself the trouble of dropping out by never going in the first place... Larry and Sergey have big-time CS degrees, but they also aren't as famous.
Let's assume (apparently falsely, according to Alexis Ohanian's anecdotes) that PG is universally famous among techie 18-year-olds. PG literally wrote important books in CS. But does PG encourage everyone who applies to YC to get their CS degree as a top priority? Methinks not. He wants you to build something. To the extent that studying engineering and math doesn't feel like building anything, it doesn't feel like a hot new SV startup.
There's a big demand for software talent, but how are teenagers supposed to grasp that getting a difficult CS degree is the best way to acquire or demonstrate that talent? Especially when we're not even very sure that's true? On the one hand, I've got a lot of Ph.D.-trained colleagues in my job (though, perhaps tellingly, none are in CS) but on the other hand several of our most valuable colleagues have degrees in music or English, and others are too young to have a degree at all.
The arrogance of a lot of the comments is depressing. What would you think if a Humanities major said that Math, Science, etc. were easy because they were simply formalisms. Simply gather the inputs, put them through a fixed series of steps and receive your precious "objective answers". And remember a mere computer (and we all know they are simply stupid, passive machines) can solve math problems.
That position is a caricature - and completely wrongheaded, but it's not much worse than many of the statements here about Humanities. There is no simple move from "non-objective answer" to "easy", nor - even more - from "non-objective and many possible answers" to "anything goes random subjectivism". That's just sloppy thinking.
tl;dr If you find yourself comparing an entire academic field to a degree in basket weaving, you've gone off the rails.
When the article says, "We’re losing an alarming proportion of our nation’s science talent once the students get to college", and the reason is they can't hack hard math, it seems to me we are not "losing our nation's science talent", because those students weren't part of the small number that are capable of original research to begin with.
A better question to ask would be to look into why those who ARE capable choose to do other things, like become Quants for Goldman Sachs.
Wringing hands over why those who were never capable to begin with give up isn't worth worrying about any more than we should worry about the clumsy overweight and weakling kids who didn't make it into the state football team when they got to college.
The problem may not be that you are losing those capable of original research (though I am sure that many of them are lost), but that not all scientists need to be able to do that research. For each researcher, you need people capable of assisting, people in factories making the tools and reagents and measuring devices, etc.
These may be considered not as prestigious as research, but they are just as valuable nonetheless.
Great to see a reference to the WPI plan (my alma mater).
I think one of the advantages of an engineering school is that you are all in the same boat with needing to get through the tough classes (and maybe are less distracted by business majors with more time on their hands). I had a tightly nit group of friends who helped each other get through the harder classes like signals & systems, cryptography, etc.
WPI is quarter based so there is intense focus on each subject over a seven week period. If things go truly off the rails and you snowflake it is possible to recover (with semesters the concept of failing seems almost inconceivable).
It could also be the pace at which the information is conveyed. The lovely information is dished out at such a fast pace that students who would like to absorb the info at a deeper and more intuitive level are left behind while the equation-memorizers win even though they would not be able to describe WHY something is so. Secondly, science has become so compartmentalized that students with a "big picture" view of what science can do get turned off by classes that start off by focusing on microscopic aspect of some part of science.
It's probably telling that when I see the phrase "Science Majors" I don't consider Computer Science to be a part of that group.
Of course, the article also includes Engineering, etc.
I started as a physics major, changed to computer science after 3 years. Mostly because I knew physics would require a PhD to do anything interesting, and CS would just take a couple more years.
Math wasn't really a factor, but there was of course a lot less of it in CS, as compared to physics anyway.
I disagree; I think the tl;dr is a combination of (1) Math is hard AND (2) Teaching methods and student support optional are optimized for a minor subset of the student population
Point 1 is fixed; for most people, math is hard, and that's that. Regarding the second point, though, I imagine that there is a certain aspect of social engineering that could take place that would increase retention, without changing the curriculum at all.
Yeah it's hard. Pre-med students dropped out of my physics classes because the lack of grade inflation (higher risk = higher reward though)
I think STEM degrees should actively be subsidized a bit more than other degrees, even at the cost of other degrees, at state universities for a few reasons. First and foremost, they're probably the most likely to be retaken, and secondly to increase the demand and attractiveness of a STEM degree.
There's also the problem that it's made to be intentionally difficult. I know several very bright people that went to CalTech and transferred to "easier" schools like UCLA, Berkeley, and Brown. These were often students who were considered not only the top science/math students in their class in HS, but in some cases considered the top sci/math student the school had seen in years.
"rather than losing mainly students from disadvantaged backgrounds or with lackluster records, the attrition rate can be higher at the most selective schools"
I wonder if this is a visible effect of the fact that the most "prestigious" schools typically hire faculty solely on their research record and care not one whit about how good or interested they are at teaching...
For CS, at least, the problem is that people arrive at college knowing nothing, and the solution is to make CS/programming part of the core curriculum in high school, alongside math and science (replace home economics or something).
How many English majors would there be if everyone matriculated into college illiterate?
this makes no sense. english majors aren't majoring in how to speak english, they are honing their reading analysis and craft of writing. cs students aren't majoring in how to write computer code, they are learning how to apply their mathematical skills cultivated throughout school.
That's almost exactly my point. Being an English major isn't about literacy, but literacy is a prerequisite. Being a CS major isn't about programming, but programming is a prerequisite. Literacy is taught from a young age; programming isn't.
Maybe i am just weird,but i double majored in Math&Computer science primarily because it is easier to prove the teacher wrong.The arts are just too subjective to be sure the teacher is fair.
Math isn't "too hard". It's about grade inflation. When my parents were in school, a 2.5 grade-point average was around the median, and 3.0 was good enough to get into graduate school.
Math and science courses still give real grades, with Bs and Cs for average work. Liberal arts classes have been inflated at most universities, to the point that getting a C is essentially failure. People get this perception that math is "harder" because of grading skew.
Also, with top employers refusing to look at resumes with GPAs below 3.5 (regardless of university, major, etc.) it makes sense that people would gravitate toward the "easier" majors.
I don't know that it's grade inflation.. I think it's the fear of being wrong.
Look at Moniz (in the article), he bailed on engineering and went to an English/Psych where the classes are "a lot more discussion based."
In math, there are right and wrong answers. In physics, there are right and wrong answers. In engineering - where math & science hit the real world - there are "best answers given the requirements" but it's a similar concept.
In English & Psychology, not so much. There are points of view, interpretations, theories, speculation, and working models, but there are very few "right answers" and lots of "reasonable interpretations."
> Look at Moniz (in the article), he bailed on engineering and went to an English/Psych where the classes are "a lot more discussion based."
The impression that I got wasn't that he bailed for discussion courses because he was afraid of being wrong, but rather that the discussions in English classes at least gave him some outlet to feel like he was applying what he was learning.
There's a very weird death march of lectures for 2-3 years in some engineering programs during which you're expected to cram in a ton of knowledge but given zero opportunity to see how it will be applied in the real world. Compared with my own CS background it looks horrific; if I had been expected to spend my sophomore and junior years doing nothing but sit in lecture halls doing theory courses and "practical" courses in which I memorized syntax for programming languages, or facts about OpenGL, or the ISA of a MIPS CPU, etc without being allowed near a computer to actually build some kind of project using that knowledge, I'd have fucked off into Philosophy or English, too. At least it would have made time for me to make stabs at using what I was learning to do philosophy instead of simply regurgitate facts. But luckily for my future employability, CS isn't so lost to bad pedagogy that it's thrown out most of its practical learning opportunities.
My primary take away from this piece, and talking to people who have burned out of physics and engineering programs, has been that a lot of said programs are simply structured in a way almost guaranteed to try the patience of a lot of otherwise bright people who just wanted to feel like they were making progress towards being able to accomplish something more than regurgitate facts on midterms.
I can't help but agree. I went through a pretty grueling 4-year engineering program and felt very much the same. The only times I've actually enjoyed the experience was in project-based courses where I can actually go out and build.
The rest of engineering is just an endless series of lectures, tutorials, and labs, with little relation to reality. Even when the applicability is obvious, the crushing courseload of your typical engineering program means you can put little of it to real use.
In the end I decided my "education" was more important than "school", so I actively sought to minimize the impact of some courses (read: I skipped a lot of classes) so I can give myself time to, well, build things. And thank heavens for that - I could've ended up being just another clueless engineering grad. Head full of theories and zero experience applying any of it.
Funnily enough, generally speaking, the people who aced their way through an engineering program were also the ones that spent the least time building things, and at the end of the day are the last people I want working on my team. There are exceptions of course - insanely smart, superhuman people who were able to ace the courses and find time to apply it, but they are exceedingly rare. Whenever a company throws up some high "minimum GPA to be considered" on a job posting I have to chuckle. They are probably filtering out the best candidates.
I had the misfortune of spending some time in a program like this and switched to fine art promptly. I think it's a copout to say that math is just "too hard". The problem is that engineering and math programs are poorly constructed and taught abysmally. These days I study math and engineering on the side because I can learn more effectively by reading books than sitting through 4 years of slog.
My undergrad in EE is from Rose-Hulman where we spend most of our time focused on applying things to the real world. Every year except freshman, we worked on real projects from local companies, alumni, etc, etc. As a junior, I was assigned to a senior ME project because they needed some understanding from our area. I wasn't the only one drafted either. That term we spec'd out a new workflow for CDs returned to Columbia House (their main facility was in town).
I had a prof (that looked like Santa Claus) and told us point blank that the half life of our studies was about 3 years and that his goal wasn't to focus on the facts but to make sure that we learned how to learn.
Science / Math courses force you to very objectively face how good you are at the subject being taught. Fields which are more subjective leave your expertise as a much less defined concept and hence do not force the student to confront the limit of their abilities.
Sure, but that doesn't mean that everyone switching out of STEM courses "failed to measure up" or was "afraid of facts".
If I attempted to teach Computing Science to people by having them sit in 400 person lecture halls while I read K&R and CLRS at them for 4 years, any number of bright people who could succeed in comp sci would rightly get fed up with the uselessness of my shitty curriculum and drop out or switch majors. And yet a surprising number of Engineering curriculums are structured this way, because it's cheap and easy for the professors.
I recently started teaching technical subjects and here are some points from the other side of the classroom
a) Every class has outliers on either side of the performance spectrum and teaching is a resource allocation problem on part of the teachers time on how to deliver maximal benefit to the students at large.
b) The number of people who are simultaneously entertaining, inspiring and educational is very few. Look around you and shortlist the number of people whom you consider extremely knowledgeable about some subject, now remove from that list the number of people who have great people skills. Of course people skills can be improved over time but it takes real effort and time.
c) As a teacher the best one can do while teaching challenging material is spend time in the first part of a lecture developing the motivation for getting students interested. Usually that leaves little time for interactivity in the latter part of the lesson.
d) Classes where students ask no questions are very hard to make interesting. When this is happening your teacher is probably feeling like a comic doing an act where no one is laughing. Next time youre stuck in a class which seems overly dull try to help out your teacher by asking intelligent questions.
e) In some challenging material there is simply not enough time to do it justice. In those cases the best a teacher can do is do a non interactive session where they do a brain dump hoping that students will be motivated enough to follow up with the material in their own time.
f) There is also the philosophical question about who gets to teach -
1) someone who is at the pinnacle of their field and hence can earn a livelihood by teaching people so that they can contribute with original research.
2) someone who is a great educator.
g) I think truly great teachers occur with the same frequency as truly great comics because both fields have something in common.
I mean, look: math is hard. It takes a lot more mental effort to solve math problems than it does to read books and do research. That's not to say that insight and hard work aren't important in the humanties, but the barrier is lower. It just plain is.
If you're taking a class to learn about 16th century west african colonialism (or whatever other "serious" history class you want), you start at the top of the reading list and work your way through. If you hit a concept or fact you're unfamiliar with, you look it up in your other books or wikipedia.
At no point in that process do you stop and realize "Crap, I'm stuck." In math (and the numerate sciences more generally), that's the normal state of things.
Math is hard. I think we spend too much time lying to people and telling them that it's easy, and far too little time explaining how much fun it is.
Funny, my fear of being wrong is what led me to engineering. I didn't like the writing-based classes because I couldn't tell whether I was right before I let anyone else see my work.
My experience has been quite difference, but I suppose it's all relative based on what you're comparing it against. The amount of effort it takes to get an A/B in a liberal arts class and an A/B in a math/science class are lightyears apart. The liberal arts majors I knew would (maybe) do a couple hours of homework a night and party on the weekends. The engineering students would spent most of their nights doing homework or working in the labs, with more of the same on the weekends -- to get an equivalent grade. The classes that STEM majors need to pass are significantly harder than what liberal arts majors are up against and therefore require more work and effort.
It can be the opposite though. With Science, Engineering & Math, it's possible to get a 99 or 100, and isn't that difficult if you know you stuff. And once you know your stuff, you can stop studying because further effort has very steep diminishing returns.
On the other hand, it was pretty much impossible to get a grade above 89 in an English class. No essay is perfect, it's always possible to improve an essay, so a perfect grade was never given out.
I put in a lot less effort than my friends in humanities and received much better grades than them. Many of my friends in Engineering dropped out, but the ones who didn't received excellent grades.
I go to a fairly prestigious engineering school and it works the opposite way here. Math and science grades are always curved. You have no idea what grade you are going to get for the class until you get it. The grades you receive are more often than not more than you've earned. Either, the grading scale is lowered (90 and up is an A, 86-90 is A-, etc.) or your grade is boosted at the end. For example, I got a B in an organic chemistry class when it should have been a C-.
On the other hand, every humanities class I've taken, I've gotten the exact grade I have earned.
The secret is to make high school a lot HARDER.
Not wanting to do a 'when I was young rant' but I learned the calculus at age 15-16 US grade 9-10. Then in our 'senior' high schools we only studied core subjects eg. maths+physics
Now I lecture to physics undergrads who have never been taught calculus. So not only do we have to teach them intro maths techniques but they have never really studied physics - they have been shown fun and interesting demonstrations and given a simple hand-waving 'explanation' - but without being able to follow the maths you can't explain it.
I'm sorry but the answer to fixing the science and engineering problem isn't more fun science games in high school, or pretty computer demos of experiments - it's concentrating on lots of hard boring maths earlier and earlier.
Why on earth would you want to make something HARDER when faced with a large group of people who already think it's too hard or non-relevant? It's not a weed-out contest.
I have an undergrad degree in EE. If someone could find a way to make today's undergrads learn the same material, but have a much easier time of it, why would anyone oppose that? Do I really need to feel bitter because I struggled but a whole new bunch of kids breeze through the material?
Sure, we don't want to waste time, but if the "fun science games" can lead to better understanding of the math behind it so much the better. I once used a weight hanging on a long string held up to my nose as an explanation of conservation of energy: "bet that this weight won't hit me in the face when I let it swing out and back?" "Why didn't it hit me?" "What can I change so it will smack me in the face?"
It's a "game," but they did learn something useful from it.
Math is to brain what lifting weights is to muscles. We teach math not because of some useful skills but to increase brain capacity. For the same reason languages used to be taught in the past. As same as with weight lifting it has to be hard to get results otherwise it's a waste of time.
If the problem is that large number of students enter STEM but drop out because they aren't equipped - then you have to prepare them better.
Two options - one is to make STEM degrees 6 or 8 years and teach 2-4 years of what should be high school physics at the start.
The other is to spend the time in high school teaching them the 'hard' stuff - when they don't have the option to drop out - rather than trying to 'interest them in science' by showing them only pretty/fun things.
It was the soph/jr/sr years where it was very difficult to tie what you are learning to anything real and concrete. The worst part is that you are being asked to remember facts/processes/patterns for a test, then you rarely have to go through them again. So you end up forgetting most of what you learned. Since you didn't know what it was used for in the real world, you end up forgetting about something you didn't even care for in the first place.
The courses I enjoyed, I really enjoyed. At a school on the quarters system, you end up taking about 50 total courses and I enjoyed < 10 of these. My junior year I built a 4-bit processor on a breadboard using discrete parts for everything but the control system (fpga). This was one of my best memories because I understood the real-world applications of a processor and I got to sit in a lab and build it myself.
It's a broken system. It's a long grind thats not effective at getting you up to speed as a professional or getting you interested in your career. Its also terribly inefficient at improving your problem solving and analytical thinking. It still improves your problem solving/analytical thinking through brute force and I think thats a necessary process, but there has to be a better way.
If there was a way to combine the rigors of academic courses (to improve analytic thinking/problem solving ability) with the hands-on experience of vocational learning, I would be all for that. It is more applicable to software than some of the more traditional engineering disciplines, but I think it could work for all of them.
(admittedly my situation was poor because I chose the wrong major. I was an EE who was taking all CE electives and I moved into software after I graduated)
As for not being able to graduate enough STEM majors, ffs let's drop the jingoism and bring in more foreign students.