2005 October 17 Monday
De Facto Privatization Seen For Many State Universities
Percentages of budgets of state universities coming from state and local taxes have declined in recent decades.
The share of all public universities' revenues deriving from state and local taxes declined to 64 percent in 2004 from 74 percent in 1991. At many flagship universities, the percentages are far smaller. About 25 percent of the University of Illinois's budget comes from the state. Michigan finances about 18 percent of Ann Arbor's revenues. The taxpayer share of revenues at the University of Virginia is about 8 percent.
The cost of in-state tuition has gone up far faster than inflation.
The average in-state tuition nationwide for students attending four-year public colleges increased 36 percent from 2000-01 through 2004-05, according to the College Board, while consumer prices over all rose about 11 percent.
This is an argument for greater automation of education as a way to reduce labor costs. Why not record lectures on high resolution video and let kids watch lectures before they even set foot on university campuses? For many topics which have very objective material such as physics and math tests are automatable. Software could generate variations on test questions and automatically score tests as well.
State appropriatons haven't declined much but state enrollment numbers have increased rapidly.
"The air is filled with this rhetoric about privatization, but the evidence doesn't support it," Mr. Callan said. He noted that in straight dollar terms, state appropriations for public universities have not fallen much across the nation in recent years. They totaled $67 billion in 2001, $70 billion in 2002, $69 billion in 2003 and $69 billion in 2004, the last year for which nationwide data is available.
But because enrollments surged during those years by more than 1 million students, or 11.8 percent, per-student appropriations dropped more steeply than at any time since the early 1980's, to $5,721 from $6,874, according to the State Higher Education Executive Officers.
Again, automate. Automation would make education much more convenient and potentially much more rapid for anyone smart enough to focus intensely on a single subject. Watch lectures any time of the night or day and any day of the week. Watch many lectures quickly in succession. Take tests when you are ready rather than when a class's tests get scheduled.
State universities could film every class in their physics, math, chemistry, and engineering departments for starters. The most objective material with exact numeric and formula answers lends itself to automated testing. Upper division math classes which require proofs which can vary in approach and still be correct are harder to grade. But most of the more quantitative fields lend themselves to automated testing. On some topics such as economics and accounting some material can be tested in formats that lend to automation and some can not. Enough material could get tested automatically to allow big savings in labor costs and therefore in tuition costs.
I certainly am not opposed to expanding automation to higher education but as you hint at
the real big bucks spent on 'education' are not at the university level but the primary
and secondary school levels.
Now what do we get for this massive expenditure of public funds for K-12 education? Not
much, so far as I can tell. Kids can mingle with and even emulate the vices and
uncouth behavior of the underclass. Now why should mom and dad work their butts off to
buy a nice home in a decent neighborhood only to have their children still have to mingle
with the underclass? Mom and Dad would pack up and move if the parent/s of the same trash
their children must sit with in class with were to move next door to them.
With modern educators having abolished 'tracking' and surrendered control of their schools
to the underclass and with public education itself having become little more than the
study of the catechism of political correctness who in their right mind would want to send
their children to such 'institutions'. Yes, there are still good public schools but they
are increasingly rare and limited to those suburbs where, because of high cost and distance
from underclass neighborhoods they have managed a sort of 'academic quarantine' from the
decay afflicting most school districts. They are even 'attacked' for their 'exclusivity'
and efforts are made, not to restore other schools to their level, but to drag them down
to the LCD of the worst.
Automating K-12 education, where 'professional' instruction is rarely necessary is the
way to go. Even in my public school days, which fortunately were both long ago and limited
in duration I well remember the 'TV set' being brought in by the 'teacher' and turned to
WETA, which, in those days stood for Washington Educational Television not Clifford the
Big Red Dog! A lot of school districts have their own cable channel now. I've even caught
some of the programming. The math instruction is pretty good. A lot better than chalk and
blackboard demonstrations of the conventional classroom and a whole lot cheaper.
The real reason automation has not fully shown its effect in the educational system, is because the quality of the recoding is still not good enough. It is not just the high resolution video, but the style of recording that matters. It turns out that for recorded lectures, the lecturing style must be adapted also. But as time progresses, there will be better educational techniques, and although some universities like Harvard, Stanford, MIT, etc intentionally make the quality of the recorded lectures less than the quality available to their own hand-picked students, competition will gradually heat up. What we need is more internet-based universities that have a strong infrastructure for this system, and when the infrastructure is available, these internet universities can always hire moonlighting Harvard professors who will lecture exquisitively for them, and the university cannot legally block this. Similarly, the best textbooks in the world will become more readable and better illustrated in the future, once again thanks to multimedia, and this way people will be able to study at home at their own pace. Note that in big universities like MIT, the basic classes are so crowded and the lectures are so fast paced that there is almost no time for students to ask questions during the lecture, and there is minimal discussion during the lectures. Additionally, in top universities, the professors do not have too much office hours (and besides, only the bad students really need to ask questions during the office hours of basic courses, only in advanced graduate courses there is need for very detailed office hours. When Einstein was a patent clerk in Zurich, he was reading the books of Kirchoff on electricty and magnetism, and he got some inspiration from carefully reading the same text over and over again. The secret is to find a few good books and to read the same book many times under the microscope. The shortage is NOT in the quantity of books and lecturers, but in the number of GOOD students who are really creative and smart. This is why we need to change the immigration laws to modify the annual green card lottery that grants 50,000 free green cards to random foreigners every year: If we modify the green card lottery to allow ONLY those people with an IQ at least 139 to register for the lottery.
From what I remember about college, there were a lot of jobs students could do that were done by employees. Gardener, maid, cafeteria lady, secretary, etc... all these non-essential roles can be filled by students for far less money. There are some smaller schools where students perform such roles. I've also long thought that summer should be just another semester. That way students who want to graduate earlier can do so, and students can also work while in school easier.
There also needs to be more financial accountability. I've read recently about the president of some college in Cali giving her lover a cushy job. There was a report about some dean spending 25 grand on some gold medallion to wear for graduation. I'm starting to think there's a lot of graft.
In today's newspapers, there was an article about the effects of De Facto privatization in top state universition. It does NOT necessarily follow that this De Facto privatization in state universities is responsible for the fact that the tuition fees are rising dramatically even in public universities, but this is precisely what is happening. As a result, because of the fact that the tuition in public universities is increasing, and because the cost of living and many small intangible expenses at the university level are rising, getting a PhD from a top university is no longer as profitable as before, UNLESS the student joins the private sector after graduating. This is one more reason we can expect the public sector to fall behind the private sector in the future a lot more than it already is. In any case, the stratification will keep increasing until the world becomes a feudal society. The upper and lower class will NOT necessarily be demarcated by financial net worth, but by the cognitive DNA. For instance, if there is a poor but genius student, Harvard and Stanford will almost certainly find the funds to give that student a full scholarship, because such schools have plenty of money, such schools are competing among themselves in order to get the best students who will attain power in the future, so that these students will make a contribution to the schools as a feedback one generation later. Competition is brutal not only among students, but also among such top schools.
State universities will not automate to any great extent, because they function as centers for vetting elites for left sympathy. If you have people getting BA or graduate degrees via testing, in more than small numbers, independent minded types would be allowed past the vetting. These schools need to be privatized de jure, in order to disestablish a professoriate which is working to destroy civilization.
Automation is good. Applying the scientific method is better. Combining the two could be magic.
The education system we have is primarily based on sorting people. There is some merit to the idea that placing smart people around other smart and better educated people will generally lead to more smart educated people. But the fact is that those smart people would probably have learned close to as much by being around slightly less smart and educated people. This is to say if you can get into MIT what would the real educational difference be between going to MIT and a second rate school? What does MIT actually do to educate people that is different or better than the second rate school? Nothing.
Do the professors use scientifically engineered text books, course material, and tests? Can a given professor give you reliability and repeatability results for the tests they use? Do they have longitudinal studies showing that the practices and materials they use are better than alternatives? Nope. Basically they wing it, they make things up as they go along.
There are no statistical process controls used to determine if the methods being used are effective or not. If a student or a large number of students fail a test, does the school or professor try a different approach or do anything to help the student figure out what went wrong and how they can correct it? Almost certainly not. Classes are designed to weed people out, not help them learn.
Top schools are often seen as competitive and stressful places. Does being under stress help people learn? Nope stress inhibits learning. Why make the environment stressfully then? Because the goal is to sort not educate.
Rob Sperry wrote:
"real educational difference be between going to MIT and a second rate school? What does MIT actually do to educate people that is different or better than the second rate school? Nothing."
Although it is true that truly creative geniuses will get excellent education in any relatively good university with some decent resources, it it not true that "sorting people" so that all the bright people are concentrated in a few elite schools is simply a prestige oriented system. The truth is that when truly talented people are put together, they also interact in very interesting ways. For instance, the 2 geniuses behind Google, were student friends in the graduate school of Stanford, and they dropped out of the PhD program with a Master's degree to open that totally unexpected company that will probably bury Microsoft. It turns out that these two Google scientists had complementary talents, and together they were also able to bring hundreds of very talented scientists who were in the same geographic areas, thanks to their connections. But additionally, in top places like MIT, the level of education is qualitatively different, in the sense that not it is not that they only have a fast paced way of lecturing at MIT, but more importantly, they give very difficult homework assignments that even the professors in lesser schools will have trouble solving at gunpoint. Some of the undergraduates, graduate students and prossors are so good in in the top 10 schools, that they actually form a team together, and this yields a higher quality education not found elsewhere. Simply giving away the lecture notes will not make the other universities catch up with the top schools, although I fully agree 200 % that the other universities will benefit dramatically from this exchange of information.
I have commented here before about the need for automated education. It will certainly help offset the difficulty as most people find their jobs becoming useless (for a human) through increases in automation and robotics.
There seems to be a drastic underestimation of what it takes to teach someone. There is a great deal of research in Developmental Psychology about the entire process. There are also huge initiatives in schools like Carnegie Mellon University (one of, if not the, top computer science universities in America) for automated education, or even educational aides.
It turns out to be a hard problem. I'm sure it will be solved soon enough, but it will involve plenty more AI than a video of a lecture. Today, approaches attempt to build cognitive models of the possible steps involved in solving a problem, perhaps eye-tracking gleans at which part of a problem the student is looking, and the system can see where the student errors.
I wonder if there could be an experiment done just for computer science, as the open source community is so active. First, someone should make a tutorial about how to make a tutorial in flash. This might include a short audio-book with text on screen, animated demonstrations, and editable figures. This could then be used to create other tutorials, everything from teaching the basics of any programming language to the depths of a beowulf cluster and MPI.
Why doesnít someone start this? Iíll start by learning flash. This seems like a good resource: http://hotwired.lycos.com/webmonkey/multimedia/shockwave_flash/tutorials/tutorial8.html
Any other resources?
Also see my post of a couple of years ago Accelerate Education To Increase Tax Revenue, Reduce Costs.
Why learn Flash? Do you really think it is a good way to present course material?
Aside: At the moment I'm learning C#, ASP.Net, ADO.Net, and MS Visual Studio 2003. I already know Java, C++, and lots of other APIs.
As for how hard it is to learn: Part of the learning is watching of lectures. So why not record them? Some people are smart enough to learn quite fast just from lectures. A lot of the asking and answering of questions can be done via discussion boards, usenet lists, and other electronic venues. I use these resources all the time with my programming work.
Some people will still need one-on-one tutoring. But the smarter the person the less the need for that sort of thing. What I want to do is to free the smartest from the institutional structures of how education is structured now with scheduled lectures and scheduled tests.
Flash can imbed audio & video, and uses light-weight vector graphics. It is accessible through a browser, and the files aren't that large (unless you have a large video). Rather than having a video of a lecturer with a white board, I imagine a voice-over on text and animated graphics. Imagine a visualization of the stack for a language with control over memory, like C. It would make much more sense than what I've seen in actual lectures :). I still havesn't seen a good explanation for virtual function tables in C++ (i.e. a good presentation, not just acurate content).
Also, flash can be composed by non-programmers non-video-editor, which is huge.
Further, those CDs that come with many expensive college books should come with better content. I've always been very disappointed with what I've found there.
As for learning, I agree, we should be using recorded lectures. Arnold Kling has mentioned this before (though I can't find the article now).
It does take a certain level of maturity to self-teach through lectures and other resources. Automated education for younger students will probably be different.
A lot of lectures are poor quality. But if enough lecturers get recorded for a given topic then some presentations will be of good quality. Calculus is taught thousands or perhaps even tens of thousands of times a year and year after year. Surely some of those presentations are great. Get enough recorded and some will become known as good.
It seems to me that attending lectures, or watching recorded lectures is not necessarily the best way to learn. A GOOD textbook is perhaps the best way to learn because it is much more organized and structured in a complete fashion than a lecture. If the student does not have the reading skills to go through a good book, then even a good lecture won't help. In fact, in some courses that I took, the professors required the students to read the chapter and to do the exercises before coming to class. There was no lecture, but only a discussion of the problems in the class. Only in very advanced courses where there is no textbook available, would a lecture be superior to a textbook. The function of the lectures in universities, is to force and threaten the students into submission so that they are scared enough to really finish reading the chapter and to do the homework exercises. In fact, one very good way of giving a degree for top universities (for those students who were not accepted due to various reasons), might be to offer course packs for students to study at home, and then to give final exams during the summer at the university. If the "outsiders" can get high grades in the final exams offered during the summer (after having studied during the whole year), then they can get a degree from that school.
The bottom line: There is simply no substitute for good reading skills and high mental aptitudes.
I always found a good lecture in combination with a good text improved my retention and comprehension. Of course, sometimes neither were any good.