Why can't engineering be more relaxed?

sakky · October 28, 2006, 5:12am

I don’t see this as the same issue. Rigor has nothing to do with practicality, as myself and monydad have pointed out. Like he has said (and I agree), the most rigorous classes are often times the most theoretical, which generally have little true real-world practicality.

This actually gets to yet another problem I see, which I hadn’t discussed much yet, but I will now. The truth is that many engineering programs are rigorous ABOUT THE WRONG THINGS.

Let’s use Berkeley as an example. If you just want to be an EE (not a CS or a CompE guy, just an EE), you still have to pass the notorious weeder Computer Science Data Structures class. This is a NOTORIOUS weeder at Berkeley. Yet the truth is, data structures has little to do with the actual tasks of a practicing EE. EE’s simply don’t need to know more than, say, the first few chapters of the data structures book (and probably not even that). Like I said, this is not CS we’re talking about. I agree that CS people should know data structures. This is EE. If you just want to deal with circuits or fiddle with electronic devices as your career, frankly, you don’t really need to know what a radix sort is. You don’t need to know what a splay tree is. You don’t really need to understand randomized analysis. These topics have nothing to do with what an EE actually does in his daily job.

Now, don’t get me wrong. Are these things nice to know? Sure. I have no objection to EE’s learning these things who want to learn them. But the issue is why FORCE them to learn it, when they don’t need it? And, in particular, why weed people out of a particular profession because they couldn’t understand something that, frankly, they don’t need to know for that particular profession? Why do that? That’s like saying that you won’t let somebody graduate with an English degree because they couldn’t speak Chinese. What you using to weed people with has nothing to do with the actual job.

And again, this also gets to the notion of the ‘academic engineer’. I know some engineering students at MIT that have absolutely stellar grades and are true geniuses. But I wouldn’t want them to be practicing engineers. Sure, they’re brilliant math geniuses, and that’s why they do so well in the highly theoretical engineering coursework at MIT. But I don’t think they would do well in an actual engineering JOB. Theoretical coursework and actual job tasks are not entirely consonant, and in some ways, are actually orthogonal.

So by making things more rigorous, especially at the top schools, you are simply eliminating more people who could be engineers and encouraging the creation of more ‘academic engineers’ - guys who can brilliantly do the math, but can’t actually do the job of engineering. For example, making EE more rigorous at Berkeley would inevitably mean raising the standards of the data structures course, which would mean that even more people who could be perfectly good EE’s would be thrown out of the program because they couldn’t do data structures, which they don’t need to know anyway.

Then you are saying that you disagree with what MIT does with its hidden freshman grades, and what Caltech does with its freshman shadow grades (which are basically another form of hidden grades), and what Stanford does in general (which is to hardly fail anybody out). Is that right? If so, then you should come out and say that these schools are wrong.

See, there it is again, so I ask you, what about all of those lower-tier, no-name schools? They are churning out engineers who are not as good as engineers coming out of MIT. These engineers might cause your GTA game to be delivered slow, or your apartment complex to be razed because of bad design. So are you advocating the shutting down of these lower-tier schools because they produce relatively lower-skilled engineers (relative to the average)?

Maybe you recoil at the notion of my calling them “relatively lower-skilled”, but, that’s the whole point of an average - that some are above average, and some are below average. By definition, 50% of the schools out there are “below-average”, and 50% of the engineering graduates out there are “below average graduates”.

And that’s where we are going to disagree. I don’t see the assignment of letter grades as part of the TA’s job. The prof can ASK for advice regarding the assignment of grades, and the TA is perfectly allowed to give it. But I didn’t see any evidence of this occurring.

And it is precisely this sort of reactionary attitude that with which I disagree. First off, who says that you can’t cut it? That’s an arbitrary distinction. A Berkeley grad who “can’t cut it” could almost certainly cut it at a no-name school. The notion of not being able to cut it at the top schools is therefore a purely arbitrary notion. Berkeley arbitrarily decides that the students are to meet a certain threshold, and if they can’t, then that means that they can’t cut it in terms of meeting that arbitrary threshold. It doesn’t mean that they can’t cut it AT ALL.

I’ll give you a poignant example. I know a guy at Berkeley who did extremely poorly in data structures. Yet the fact is, he actually has quite good knowledge of data structures. In fact, he had a part-time job tutoring data structures to some students at San Francisco State. So it’s not as if he didn’t know data structures. He knew it pretty well, and certainly better than the SF State kids did. The issue is that he didn’t know it well enough BY BERKELEY’S STANDARDS. But that is a purely arbitrary standard. Heck, if he had been going to SF State, he’d be cleaning up.

So when you say that somebody doesn’t cut it, I would ask “by whose standards”? According to Berkeley’s standard, he doesn’t cut it. By SFState’s standard, he easily cuts it. Yet neither standard is inherently more correct than the other. Whichever one you choose to adopt is a purely arbitrary choice. Hence, the notion of not cutting it is also an arbitrary distinction.

GLOBALTRAVELER · October 28, 2006, 9:12am

“The truth is that many engineering programs are rigorous ABOUT THE WRONG THINGS.”

Now THAT is something I agree with. I have been “at odds” with computer science curriculums for awhile now because of the lack of emphasis on practical applications.

Cheap plug: Some of you read my thread about “The only CS courses that you need are” thread.

See, the business schools/colleges/department got the message awhile back. They basically said to the engineering departments “Here…you keep your computer science major, we will develop our own major called Information Systems which has much less math and focuses on more practical applications”.

I have been in the I.T. field long enough to see plenty of I.S. majors get hired OVER CS/CompE majors. I won’t even lie and I know this is wrong but there were times when I had a vote in the hiring of a new employee and I chose the CS/CompE/Math major (and I was overruled) because I favored the person who had the “rigor” that I endured in college.

My point: I hope that the CS and Engineering departments get a grasp on real-world applications, not just “academia” applications. To be honest, I should NOT have to rely on Borders Book Store to gain knowledge on real-worl practical applications.

Dirt_McGirt · October 28, 2006, 10:31am

You are assuming that someone from a no-name school will start out with the same responsibility as someone from Caltech. That is not the case and I think this is what makes a good number of your arguments about the lower tier schools wrong. Yes they will be engineers, but they will not be given the same responsibility (right away) as someone from a top school.

Your example of the guy in the aerospace industry only means that he proved himself AFTER her got a job. I’m not saying that someone from a no-name school can’t rise to have great responsibility, I’m only saying that they would not be given as much responsibility initally with their only record of performance being their college history.

sesshomaru511 · October 28, 2006, 10:37am

''no-name school ‘’

name a couple of ‘‘no-name schools’’ in the Mid-Atlantic areas?

Dirt_McGirt · October 28, 2006, 10:40am

Then why are you advocating that universities should keep information from employers?

You are ignoring the fact that it is possible for Caltech to set their standards higher. Obviously, Caltech has higher standards for what an engineer must be than Alabama Tech does. Just the same as how Northrop Grumman has a different definition of what a competent engineer is than Johnny’s Toy Making Shop does. Caltech does not think that your hypothetical person is good enough to be an engineer. Some other school may well give them a degree, but that is not on Caltech’s shoulders at that point.

Dirt_McGirt · October 28, 2006, 10:46am

I’m not familiar with the details of the policy at those schools, but usually covered grades only mean that you get a pass/fail kind of grade. You can still tell when someone failed a class. This is meant to give people a period of time to adjust to college classes without having the agonize over the difference between an A and a B.

Also, take your quote and realize the power that is placed on “in certain instances.” You are saying no one should ever fail a class. Those schools try to give students a short adjustment period where they can get the hang of college. I think that is a little different from removing responsibility completely from the college experience.

Dirt_McGirt · October 28, 2006, 10:58am

Look them up yourself. I don’t see how it’s relevant to this discussion for me to spend my time looking through engineering schools in a certain part of the country that you are interested in.

GLOBALTRAVELER · October 28, 2006, 11:13am

I hate to be party-downer type of guy on this topic, but I still feel way too much emphasis is placed on the “name” of the college. Just at my employer alone, we have Cal-Tech, a few Cal/Cal-State at Whatevers, some U-Texas grads and even a Carnegie-Mellon grad and they ALL report to a chain of folks from schools that you would not even think offered a Calculus course…much less engineering degrees.

sesshomaru511 · October 28, 2006, 11:21am

‘‘Look them up yourself. I don’t see how it’s relevant to this discussion for me to spend my time looking through engineering schools in a certain part of the country that you are interested in.’’

I just want to know what makes an eng. school a ‘‘no-name school’’?

for example would you consider Temple University a ‘‘no-name school’’?

Dirt_McGirt · October 28, 2006, 12:59pm

I’m not trying to insult schools, we were just comparing schools so we had to refer to them as something. I guess just compare a top 20 school to a 3rd tier school.

I have no idea where Temple fits into the big picture, sorry.

morfinx · October 28, 2006, 4:53pm

I do love Sakky’s perennial “well I know someone who… so that’s why…”. Excellent supporting evidence :)

Carry on.

sakky · October 28, 2006, 5:24pm

But you have to ask yourself why that is the case? Why, just because you’re from Caltech, you will probably be placed into a position of higher authority? It is because, as I said, employers have limited information and are therefore relying on the brand-name of the school as a signal of quality.

But if employers rely on brand names as signals of quality, then we have to agree that employers also rely on GPA also as a signal of quality. And that’s where the misinterpretation of engineering GPA’s comes into play.

Look, the truth is, it would be nice and idealistic to think that employers have perfect information and perfect rationality and not need to rely on market signals. That is essentially what you (and wrprice) are advocating - that non-engineering employers should simply learn how to properly interpret engineering grades. The truth is that employers are imperfect human beings working with incomplete information working under the tenets of bounded rationality. Labor markets are riven with asymmetric information, which is why market signas like grades and brand names are so important in labor markets. But with that comes the possibility of misinterpretation of signals.

Hence, my opinion is that if employers are going to consistently misinterpret a particular signal (i.e. engineering grades), then the best thing to do is simply not provide that signal. It’s like telling a guy who is travelling to Turkey not to put his feet up, because if you do, you will be misinterpreted.

Because, like I said, employers consistently misinterpret it.

But again, the question is WHY does Caltech have a higher standard. Again, this gets down to the notion of brand names. Caltech has a brand name (a market signal) to protect. If Caltech were to consistently produce subpar (relative to Caltech’s standards) graduates, then Caltech’s brand name would decline.

Now, don’t get me wrong. I am not saying that there’s anything wrong with that. The Caltech standard is arbitrary, but that doesn’t mean that it’s wrong. It just means that it’s arbitrary. Caltech has a strong brand name to protect. Fine. I accept that.

But then, like I said, you have to ask why brand names matter in the first place. Like I said, if employers really had perfect information and perfect rationality, then they wouldn’t have to need to rely on brand names. They would know exactly how good of an engineer you are and would be able to slot you into the exact job that fits you.

But of course we know that this is not realistic, as employers do not have perfect information and are not perfectly rational (nobody is).

So let me trace out the logic for you. You say that employers are going to give somebody from Caltech a better job just because he’s from Caltech. I agree. That’s because the Caltech brand name serves as a market signal to tell the employer that he should give that guy a better job. But by the SAME LOGIC, non-engineering employers are going to see a guy with supposedly subpar grades and give that person a bad job (or no job at all), even if those bad grades came from engineering. They’re not going to know or care that this guy has bad grades only because he tried out engineering and was weeded out. All they’re going to see is that this guy has bad grades, so we shouldn’t hire him, or if we do, give him a bad job.

Don’t you see, Dirt McGirt, it’s the SAME THING. In both cases, employers are relying on market signals to combat the information asymmetry. The problem is that in the latter case, the signal is misinterpreted. That is why I say that if the signal is consistently being misinterpreted, then the best thing to do is to simply not send that signal.

Lest you think I’m off base here, let me point out, again, that this is PRECISELY what MIT and Caltech do with their hidden/shadow freshman grades. MIT has a system of dual-transcripts, one internal (can only be seen by MIT faculty/staff), and one external (this is the one sent to outside parties). The external one has much less information than the internal one, and the biggest piece of missing information are you failed freshman grades. Essentially, you could fail ALL of your freshman classes at MIT, and nobody outside of MIT would ever know. Caltech has a similar policy with shadow grades (which are basically like MIT hidden grades). By all means, read about these policies if you don’t know about them.

<a href=“http://www.mitadmissions.org/topics/learning/why_you_cant_fail_a_class_freshman_year/[/url]”>http://www.mitadmissions.org/topics/learning/why_you_cant_fail_a_class_freshman_year/</a>
<a href=“http://web.mit.edu/uinfo/academics/grading/grades/features.html[/url]”>http://web.mit.edu/uinfo/academics/grading/grades/features.html</a>

So you say that universities should not keep information away from employers. But some universities, as we speak, are ALREADY DOING THAT. Like I said, MIT is already denying information about failed MIT freshman grades to employers. Are you saying that MIT is wrong to do that?

I don’t think MIT is wrong. I don’t think Caltech is wrong. If these schools can deny information to employers, then why can’t other schools do the same? Why not? Employers evidently still hire MIT and Caltech eng grads in droves, despite being denied some information about their grades. So is it really so radical for other schools to adopt similar policies? I don’t think so - I am merely proposing solutions that some schools have already successfully enacted.

Wrong. In the case of MIT, see the above links. No evidence of failed freshman grades will appear on your external transcript. The only future “employer” who will know about your failed grade is MIT itself (i.e. if you decide to work at MIT as a staffer). Otherwise, nobody will ever know that you failed a freshman class at MIT.

Again, please, by all means, if you don’t believe me, read the policies and see for yourself. I have pulled out the relevant quotes. If you don’t believe me, I have also provided the links so you can read all about it.

"As a freshman you are graded Pass/No Record for your first semester and IAP, and ABC/No Record for your second semester. The “No Record” grades of D and F are recorded on your grade report, displayed in WebSIS, but are not available on your transcript. "

“As a sophomore you may opt to take one subject per semester as Exploratory, which permits you to change the status of the subject to Listener after the end of the term through Registration Day of the next term. This allows you essentially to drop the subject after you get the grade.”

<a href=“http://web.mit.edu/uinfo/academics/grading/grades/class.html#frosh[/url]”>http://web.mit.edu/uinfo/academics/grading/grades/class.html#frosh</a>

"While the freshman year curriculum is very demanding, the grading system is designed to eliminate unnecessary pressure and competition. In the first term, all subjects are graded on a “pass/no record” basis. That means if a student does work at the A, B, or C level, a “pass” is recorded on his or her external transcript; if a student should happen to do work at the D or F level, the external transcript will show no record of the student ever having taken the class.

Why does MIT launch freshman year with this unusual grading system? Because it eases the transition to college, allows students to adapt to doing MIT-quality work, gives them flexibility to explore academic, research, and social opportunities at the Institute, and deemphasizes grade competitiveness while emphasizing learning for its own sake. These lessons stay with students throughout their years at MIT and are a big part of the MIT culture.

The second term at MIT is graded on an “A/B/C/no record” basis. While students do receive letter grades on their external transcripts for courses passed, students still have the “safety net” of having no record made of classes in which they happen to receive a D or an F. In this way, students have their entire first year to get used to the workload and standards of MIT, without serious penalty."

<a href=“http://www.mitadmissions.org/topics/learning/why_you_cant_fail_a_class_freshman_year/[/url]”>http://www.mitadmissions.org/topics/learning/why_you_cant_fail_a_class_freshman_year/</a>

And then of course you have Stanford which solves the problem by basically almost never failing anybody, something that I am sure Stanford people like Imblue can attest to.

But my point is this. MIT is, right now, denying some grading information to employers. So if MIT is already doing that, then why is it really such a radical suggestion for other schools to also deny some information to employers? Nobody seriously disputes that MIT is a top engineering school, and yet here they are, denying some grading information to employers. Nor is this policy particularly new - the sophomore exploratory policy is new, but the freshman hidden grades policy has been an MIT policy for at least several decades, if not longer.

sakky · October 28, 2006, 5:28pm

Well, hey, other people are using anecdotes to debate me. So why can’t I do the same?

Look, I’ll drop my anecdotes if others drops theirs. Cool?

sakky · October 28, 2006, 5:53pm

First off, again, what is so radical about the notion of somebody not failing a class? Like I said, that is essentially what Stanford does right now through inflated grading, and nobody disputes that Stanford is an elite engineering school. Almost nobody actually fails a class at Stanford, not even an engineering class. It’s difficult to get an A, but it’s almost impossible to actually FAIL. So if Stanford can get away with it, why can’t others?

Secondly, I don’t believe I am advocating the taking away of responsibility. Like I said, instead of failing a class, just don’t give credit for that class. If that class is a required class to graduate, then that student needs to take the class again, and if he can’t do it, he will never graduate. I think that places sufficient responsibility on the student. You don’t put in enough work to pass the required classes, and you won’t get a degree.

Personally, I think the best model I have seen is the Stanford model. I believe engineering schools should be more like Stanford, in terms of giving out relatively high grades, but also enforcing high initial selectivity to ensure a high quality graduate. Stanford is living proof that this model can work and work extremely well. Stanford is probably the best engineering school in the world other than MIT, and even that is somewhat debateable. But in any case, if there is one engineering school in the world that could potentially dethrone MIT, it’s Stanford. Stanford’s great competitive advantage as an engineering school is that it combines an elite program with enlightened grading, and this will draw those top engineering students who don’t want to subject themselves to the savage grading of other programs. Why put yourself through hell if you don’t have to?

Now, let me be clear. I have no personal affiliation with Stanford. In fact, my personal affiliations are with several of the other programs mentioned here. I am simply identifying a policy that I believe to be good. I believe that the way that Stanford runs its engineering program is the way of the future. MIT is even starting to lighten up with its introduction of sophomore exploratory classes (which basically let you drop a class after you have already received a final grade), and both MIT and Caltech are far less harsh schools than they were in the past (as evidenced by graduation rates which are far higher today than they were even a few decades ago). I believe that other schools should do the same, or else they are going to lose top students to those schools that do change.

GatorEng23 · October 28, 2006, 7:54pm

How can you ensure that those students who failed the course know the material? The problem with your idea is that it can blow up badly. In an ideal world, students would come into a class ready to work hard and learn. If you have a system of easy grading and no ‘F’ grades, then not many students are going to take the coursework seriously. The reality is that grades motivate students to do the work and to learn the material.

Your system can only work with small, private engineering schools. It may work at Stanford, where it is a safe bet that the students learned the material since you know they are smart to begin with. But do you really think you can apply this system to a huge, state public university, where there are thousands of engineering students? Students are lazy as it is with the current system, could you imagine how much more unmotivated they will become if their department issues a “no failing and easy grading policy”? I say, whose up for some more football?

sakky · October 28, 2006, 8:27pm

How can you ensure that those students who failed the course know the material? The problem with your idea is that it can blow up badly. In an ideal world, students would come into a class ready to work hard and learn. If you have a system of easy grading and no ‘F’ grades, then not many students are going to take the coursework seriously. The reality is that grades motivate students to do the work and to learn the material. 

Your system can only work with small, private engineering schools. It may work at Stanford, where it is a safe bet that the students learned the material since you know they are smart to begin with. But do you really think you can apply this system to a huge, state public university, where there are thousands of engineering students? Students are lazy as it is with the current system, could you imagine how much more unmotivated they will become if their department issues a “no failing and easy grading policy”? I say, whose up for some more football?

You’re exactly right - which is, again, another nail in the coffin of the presumed ‘public safety’ gambit. After all, if we REALLY wanted to maximize public safety, we would close down all of these huge state engineering programs that have lots of lazy students. That would clearly be a highly effective way to protect the public from dangerous engineers - just shut down those mediocre engineering programs. But nobody, least of all me, is proposing that. Hence, public safety must not be the real issue.

However, more to your point, I would say that your concerns can be addressed in myriad other ways than just tough grading. Like I said, instead of just failing people, just don’t give them credit for the class at all. If the person needs that class to graduate, then he won’t be able to graduate. Seems to me as if the problem is solved. Those students who are lazy won’t be able to earn credit for their required engineering classes, and hence won’t be granted degrees.

Another way is to enforce an ‘exit exam’. I know some schools already do this by forcing seniors to have passed the EIT exam before graduation. This is, in effect, an exit exam. You don’t pass, you don’t graduate. One could combine this with another comprehensive school-wide entrance exam (as opposed to the state-wide EIT), and if a person doesn’t pass this exam, again, he doesn’t graduate.

Yet another way would be to enforce an ENTRANCE exam into the engineering major. If you want to declare engineering, you have to pass a comprehensive (and difficult) test about numerous engineering prereqs. You don’t pass, you don’t get to declare engineering as a major.

Many schools, like Berkeley, already restrict the entry of students who want to switch into engineering. For example, if you come into Berkeley as a Lettters & Science student and decide that you’d rather do EECS, you don’t just get to switch over just ‘like that’. You have to apply for entry into the EECS program via an internal transfer process, with no assurance that you will get in. In fact, most of these internal transfer applicants are denied entry. The flaw that I see is that the process uses lower-division grades to determine whether can switch in. I believe things would be better run via an entrance exam instead.

Another way would be to implement MIT’s policy of “pass/no-record” for all engineering weeder courses, but to determine whether you can enter the major. So if you fail an engineering weeder, you aren’t able to enter the major, but no record of that failure is printed on your transcript, which allows you to freely choose another major with a clean slate. If MIT can do this, why can’t others? Obviously nobody wants to fail the class and hence waste their time in that class. But if you do fail, then you are free to major in something else.

The point is, there are many ways to skin this cat. It’s not like other options don’t exist. Rather, it’s that engineering programs don’t WANT to pursue those other options. Like I said, there’s a big difference between “can’t” and “don’t want to”.

And if certain schools don’t change, then I am afraid they are going to keep losing students to those schools that do change. For example, if you’re a prospective engineering student who gets into both Stanford and Berkeley, then the fact that Stanford has relaxed engineering grading and Berkeley has harsh engineering grading gives you a strong reason to prefer Stanford. Why undergo pain if you don’t have to? It’s like choosing to have an operation with anesthesia, or without anesthesia. Who, other than masochists, would choose the latter? So, what will happen is that the top students will migrate to the enlightened schools. Even MIT is changing (if slowly) via the implementation of sophomore exploatory grading, which basically allows you to drop a class after you’ve taken it and seen your final grade. Even Berkeley, I hear, is becoming a smidge more relaxed in its grading. This is the direction that engineering is undertaking, and I think that’s a good thing. I just wish that more programs would get with the program, no pun intended.

aibarr · October 28, 2006, 8:40pm

This makes no sense at ALL.

Dirt_McGirt · October 28, 2006, 11:06pm

You just keep saying the same things over and over again. I’ve stated that I don’t think some of your examples get completely to the heart of the issue and what my feelings are on the subject. I’m not going to keep arguing the details with you. I’m done arguing about it. I disagree with you.

wrprice · October 29, 2006, 11:54pm

If you flood a medium with noise, no one can claim the signal isn’t there. (Doesn’t mean it’s useful any more.) I’m outta this thread, Sakky. There’s no point to arguing with someone who’s dead-set on shouting to the world that he thinks he’s right.

I, too, disagree with you.

pippyvan · October 30, 2006, 11:26am

I’m not sure if there’s a right or wrong here. As Sakky said, “there’s more than one way to skin a cat”. I agree more with the traditional methods of grading, that is a normalized grading scheme, rather than the methods sakky has been talking about. But it doesn’t mean I’m wrong or right. However, there’s a difference between explaining your views and stubbornness and that line has been crossed to the point where this thread has become a shouting battle.

gg you win.

Why can't engineering be more relaxed?

CONNECT WITH US