Education: September 2005 Archives

Doug Kern explains why he thinks he washed out of engineering school, but while criticising his teachers (often justly) I think he too-hastily dismisses what may have been the real reason for his failure.

I am an engineering washout. I left a chemical engineering major in shame and disgust to pursue the softer pleasures of a liberal arts education. No, do not pity me, gentle reader; do not assuage your horror and dismay at my degradation by flinging a filthy quarter into my shiny tin cup. Instead, hear my story, and learn why the United States lacks engineers.

Not long ago, I showed up for my first year at Smartypants U., fresh from a high school career full of awards and honors and gold stars. My accomplishments all pointed towards a more verbal course of study, but I was determined to spend my college days learning something useful. With my strong science grades and excellent standardized test scores, I felt certain that I could handle whatever engineering challenges Smartypants U. had to offer. Remember: Kern = real good at math and science. You will have cause to forget that fact very soon.

Being "real good" at math and science isn't the most important thing for engineers. I know engineers who are brilliant mathematicians, but many aren't. I've taken years and years of calculus, but whenever I have to apply it to a difficult problem I need to scour the internet for pointers. As for science, most science classes are just memorization.

As Clayton Cramer recently wrote, it takes more to succeed as an engineer than just intelligence.

When my wife and I first met, she asked me what I did for a living. I explained that I was a software engineer. She was impressed, and assumed that it was a very difficult job, requiring exceptional skill. I told her that I thought you could almost teach chimpanzees to do it. I was exaggerating for dramatic effect, but my perception was that the skills that I had were really very widespread.

I've since found out that not ony can't you teach chimps to do this, you can't even teach a lot of very smart people to do this. My wife is a very bright, very thoughtful, very logical person. With my encouragement, she took a programming class when she was attending Santa Rosa Junior College. For reasons that I could not understand, this very smart woman that I am married to just didn't get it--and I've discovered that she is not alone in this respect. It is apparently somewhat harder to learn how to program, at even a very simple level, than I realized.

... Most important of all, from the standpoint of wage rates, most people seem to lack something fundamental that allows them to be effective software engineers. They may be able to write simple programs--or even write programs that should have been simple, and turn them into steaming piles of incomprehensible crud--but they will never be a software engineer. Hence, wage rates are pretty darn good for this line of work.

Engineering requires a certain attitude: a morbid cynicism mitigated by can-do; deep curiosity with ruthless practicality. Most people don't have it. As an engineer myself, I think it's pretty fun, and I wish more people thought like engineers... we might not end up with so many stupid messes. But then again, most artists wish others thought like them; it takes all kinds.

Mr. Kern goes on to lament what sounds like a rather pathetic engineering school at Smartypants U., but I don't think his experiences are atypical in America or abroad. The qualities that make for good engineers are often orthogonal to those that make for good teachers, which makes such combinations rare. The result is that most engineers, even highly-educated ones, are significantly self-taught; a person who can't bootstrap himself into engineering probably won't make a good engineer. Many of the best engineers I've known, however, weren't engineers by training or profession.

When it comes to secular theology, the worship of education is high in the statement of faith; and when it comes to education, dogma number one is the belief in the benefits of class size reduction. On the surface it seems logical that a teacher with fewer students can give more attention to each, and it seems logical that a student who receives more attention from teachers will learn more. Right? Well, apparently not. Despite America's devotion, class size reduction has little effect on learning.

Tennessee's Project STAR (Student/Teacher Achievement Ratio) - a study conducted between 1985 and 1989 - seemed to support that conclusion. Over 6,000 students were involved in the study in which children from 79 schools were randomly sent to large, medium and small classes. The study's conclusion was that smaller classes led to significant performance improvement, estimating that students who stayed in small classes for four full years - kindergarten through grade 3 - ended up 5.4 months ahead of their peers by the time all had entered grade 8. Moreover, benefits for minorities outpaced the positive affects for white students, producing nearly twice the gains. The STAR study seemed to substantiate all that pro-class-size-reduction forces had expected, and is now cited as fact by educators, academics and policy makers. ...

Harvard University's Caroline Hoxby argued that the methodology of STAR was lacking. Its biggest flaw was that study participants knew they were being studied and hence tended to work to achieve outcomes desired by the researchers. As Hoxby writes, "the schools in a class size experiment may realize that if the experiment fails to show that the policy is effective, the policy will never be broadly enacted. In such cases the schools have incentives that the fully enacted policy would not give . . . . the experiment alters the incentive conditions . . . In addition, some individuals temporarily increase their productivity when they are being evaluated."

Eric Hanushek, then at the University of Rochester, also examined the study's methodology, but pointed to different shortcomings. Among these were:

* Between 20 and 30 percent of students in STAR quit each year, leaving less than half of the original group by the study's end.
* The students who quit were disproportionately low performers, providing a statistical boost to smaller classes.
* No pretests were given to students at the beginning of the study, providing no baseline off of which to measure achievement gains.
* While students for the program were chosen randomly, teachers and schools were not.

In other words, STAR was bad science. The researchers found what they set out to find. Real-world experience in California -- at great expense -- has demonstrated that reducing class size has essentially no effect on student learning.

Inspired by STAR, the state of California forged a plan to implement a massive class-size-reduction effort. Starting in 1996 the Golden State set in motion a program to lower average state K-3 class size from an average of 28 to 20 students. Unfortunately for Californians, there have been no achievement gains as a result of reducing class size. According to a series of reports by the state commissioned Class-Size-Reduction (CSR) Research Consortium, class-size-reduction has not achieved anything like even the questionable gains reported in the Tennessee experiment. In its third and most recent report, the Consortium reported finding no evidence that class-size-reduction has produced improved scores, though it has indisputably cost a great deal of money and displaced many effective programs and teachers. While they found that "achievement has been increasing during CSR's implementation," the researchers concluded that there "was no strong association between differences in exposure [to reduction efforts] and differences in achievement effects during this period." In other words, there was no correlation between how long students were in reduced-size classes and changes in their test scores. And the cost to achieve so little? To date, an estimated $8 billion.

Go read the rest of that article and you'll see that countries all around the world have larger classes and higher performance than the United States -- if anything, it appears that large classes are beneficial, at least for some subjects. Why? Perhaps because large classes allow good teachers to work with more students. The paper goes on to argue that what really helps students is a reduction in school size.

It shouldn't be any surprise that the primary culprits behind the dogma of class size reduction are the teachers' unions. Smallers classes means more teachers with easier jobs, so the unions have a monumental incentive for advocating the failed programs. The $8 billion dollars wasted by California tax-payers went straight into the pockets of the unneeded teachers and their union organizers. What's more, larger classes with fewer teachers would mean that the best teachers could teach more children and the worst teachers could be dismissed -- a concept antithetical to the unions.