Slow Motion Change in Engineering Education

I wondered in my previous post, why have I met so many engineers who started out thinking that engineering wasn’t for them? My theory was that if so many people were discovering by accident that they enjoyed and were good at something they had previously felt put off by, that there must be something wrong with the way the field of engineering has been presenting and introducing itself to the uninitiated.

I’ve also been hearing recently about the various ways schools like Harvey Mudd and Carnegie Mellon University have been changing the way computer science is introduced in their curriculum. One of the key insights is that they spend more time on practical applications and on the purpose of what engineering is for. 

There’s a book that I picked up to explore this topic more, The Existential Pleasures of Engineering by Samuel C. Florman, first published in 1976 and then reissued in 1994. He writes in the original preface, “In this book I have not attempted to describe in any detail what engineers do. Rather, my interest is in how engineers think and feel about what they do, and in the more general aspects of what it means to be an engineer.” That’s exactly what I was hoping for.

In the second preface of the book, written by Florman in 1994, he notes that much of what he had written in 1976 was still relevant, which is saying a lot, considering that 1976 was the year Apple introduced the Apple I. In 1994, if you had a PC, it would have probably been running Windows 3.1. Michael Spindler was CEO of Apple. 

Obviously, there were many technological advances. “But,” he writes, “in most important ways engineering education and engineering practice have not been fundamentally transformed. In fact, one of the main characteristics of engineering since 1976 is that it has not changed as much as it should have changed.”

In what way? What should have changed? Florman criticizes what had become of American engineering schools and the faults of their graduates who, “…are poor communicators, have little understanding of business and politics, and are uninformed about foreign cultures and world affairs. At a time when we need engineers who are leaders, and leaders who understand engineering, we are not producing either.”

At that time in 1994, the number of bachelor's degrees in engineering had hit a peak in 1986, and had already declined by 18% by 1994. Attrition between freshman and senior years was about one third. For women and minorities attrition was even worse.

Why were so many dropping out? Academic failure? Poor job prospects? Neither explanation fit the facts. Florman writes, “Young people are dropping out of engineering school for the same reason they are shunning it in the first place: The program is laborious and in many respects disagreeable.”

This is interesting. This sounds exactly like what the changes being made today in the computer science curriculum are intended to correct.

We should help young people perceive how important technology is in the scheme of things. — Samuel C. Florman

Florman offers this prescription for what needed to change in 1994:

“Once the major problem has been identified, the solution seems stunningly obvious. We should stop looking at engineering school as a boot camp designed to eliminate all but the most dogged recruits. We should stop making the first two years the obstacle course they have become—consisting of calculus, physics, and chemistry. We should bring practical, creative, ‘fun’ engineering into every year, particularly the first, and teach mathematics and the sciences as enabling complements to engineering rather than isolated afflictions to be endured. We should help young people perceive how important technology is in the scheme of things. We should advise and nurture the students at every step along the way, paying particular attention to the needs of women and underrepresented minorities. Thus will we attract talented young people to engineering, keep them from dropping out, and at the same time improve the quality of our graduates.”

Reading these words in 2019, I find them shocking. How could it be that the same issues Florman highlighted in 1994 are the same problems that are being addressed today? Are engineering schools still trying to implement what 25 years ago was “stunningly obvious?”

Perhaps I’m just not familiar enough with the progress that has already been made? I need reader feedback on this. I haven't been an engineering school student lately, but I still remember the “weed-out” classes I encountered as an undergraduate.

In my case, it was Introduction to Electrical Engineering. I did a quick search to see if anybody else recalled that class in the same way, and I found the recollections of Cameron Campbell matching mine. He wrote regarding that class, “While all the classes were challenging, it was frustrating that at least some of them were clearly designed to scare away anyone who wasn’t in the top tail of the distribution... The worst offender by far was the required course in electrical engineering, EE 14ab.”

In that class we had grueling problem sets. The final product for each one was about 40 handwritten pages of math and circuit diagrams that usually took pulling an all-nighter to complete. There was a common room in one of the houses that was typically full of electrical engineering students at 4am the day problem sets were due.

I think I managed to gut it out for one quarter, but I didn’t re-up for EE 14 b. I clearly wasn't going to survive as an electrical engineering major, at least not in that program. I opted to stick with the more humane major of computer science, a decision that I ultimately have no regrets over.

Something has changed since then. The downward trend in overall engineering degrees in 1994 did bottom out and reverse (see graph below, top line).

Even my old nemesis EE 14 is gone. Instead, first year students interested in majoring in electrical engineering take this introductory course:

Actually, that looks like a lot of fun. 

Engineering education clearly has changed. But if we’re still talking about the same problem that needed fixing 25 years ago, then perhaps it is still true that it has not changed as much as it should have changed.

What do you think?