Do You Want the Blue Pill or the Red Pill? - Part 1

College Search & Selection
1

IPEDS lets us look into The Matrix to see what is real. The data isn’t very granular. For instance, what I’ve found is university wide and and isn’t subdivided into colleges or classes. It is illuminating though. There are a lot of TAs at some schools. They have to be doing something. I’ll summon @Data10 for a better understanding of the data.

Below are schools that frequently come up when engineering is discussed followed by the total number of instructional staff (excluding TAs) and graduate teaching assistants. At the first 4 listed, less than 60% of the “teachers” are instructional staff. At the first three, there are more TAs than instructional staff.

Do you want the blue pill or the red pill?

Cal - Berkeley 2737 3481
University of Illinois - Urbana - Champaign 2669 2849
Georgia Tech 1220 1248
Purdue 2431 1804
MIT 1712 680
Stanford 4134 1130
Cal Poly 1289 100
Hose-Hulman 196 0
Harvey-Mudd 120 0
Olin 57 0

EDIT: I removed the prologue about Cal Poly. I don’t want my homerism to deflect from the discussion.

1 Like
2

My Purdue engineering has still never had a course taught by a TA but yes, the PhD candidates run the small group recitations and the labs. Across the board, D says they have been awesome.

I don’t buy the argument that TAs lessen the quality of the instruction. In many cases, D says they enhance the class.

There have also been studies done that show that students are more apt to remain in their major if they have TAs as part of their instruction.

7 Likes
3

Having been a TA for three years as a grad student, teaching labs and recitations, I’d like to think that my teaching was mostly beneficial to the undergrads of our department. All the PhD students and post docs had to contribute to our advisor’s teaching load and as we were an experimentally heavy department we all taught a lot during the year. Some of my fondest memories of grad school were when undergrads came by my office to ask questions and would stay for hours discussing the material but also the joys and frustrations of experimental science. All us TAs would encourage the undergrads to try research through summer work and would advocate for them to our advisor (the tenured professor leading the lab). Many stayed on for graduate work. I guess we were a pretty happy and cohesive group at a big public university. I would not think of any student being at a disadvantage for being taught by the whole group rather than only the lead professor (and he is an excellent instructor). Now that may not be the experience all undergrads encounter with TAs but being taught by professors only is not the miracle cure for excellent teaching. Some of my worst experiences as an undergrad were with celebrated and well known professors.

8 Likes
4

Student:faculty ratios should not include TAs…not on the common data set, not on IPEDS (which says to use the same logic as CDS I-2):

Report the Fall 2020 ratio of full-time equivalent students (full-time plus 1/3 part time) to full-time equivalent instructional faculty (full time plus 1/3 part time). In the ratio calculations, exclude both faculty and students in stand-alone graduate or professional programs such as medicine, law, veterinary, dentistry, social work, business, or public health in which faculty teach virtually only graduate level students.

  • Do not count undergraduate or graduate student teaching assistants as faculty.
1 Like
5

It wasn’t meant to be a judgment on teaching quality. Certainly full professors can be bad and grad students can be good.

Student teacher ratios ideally would follow the IPEDS guidelines, but they don’t always (I edited that in the original post).

Even sticking with the IPEDS guidelines, they don’t paint an accurate picture. The student teacher ratio at Berkeley is roughly double that of Harvey Mudd, about 10 more students per section on average at UCB than HMC. There are lectures at UCB though that are larger than the whole HMC student body.

6

Then what is the point? What is “a different story” if the mix is not meaningful?

1 Like
7

I didn’t say the mix isn’t meaningful. In my mind, it very much is, not so much due to the implied quality of teaching (which may or may not be accurate), but what you can infer about class size.

That said, I can tell you having someone teach a fluid mechanics lab and discussion that was at NASA for the first half of their career is different than having someone who just stepped out of undergrad. Again, the NASA alum may be a horrible teacher, but at worst they have additional experiences to draw on that a graduate student won’t.

8

Fluid mechanics is not my jam but solid mechanics is. Our group had so much lab equipment, each lab member was responsible for a bunch (with the help of our expert lab technician) and taught the labs requiring said equipment. Our advisor had no idea how all this equipment functioned and was not allowed near it unless accompanied by a postdoc or PhD student (unwritten lab rule :wink:). No one was allowed to teach right out of our undergraduate studies and the labs we usually taught were closely linked to our personal research. I’d say the knowledge was there and if any holes in understanding became apparent our prof’s door was always open for questions. Sometimes he also had to refer a student somewhere else (ex. to an electron microscopy expert down the hall).

I know your son went to CalPoly and that has a very different style of teaching than most research universities. My daughter also looked into it for engineering (was accepted) but decided against it. When I looked into it I loved what I saw particularly in my field of study. Mudd and Olin, Rose Hulman etc. all have different styles of teaching, but the reality is that most engineers are trained in large research universities. My undergrad and grad engineering studies were all at large research institutes (albeit in Europe) and for my style of learning it worked well. I love how you encourage kids to look into different schools for engineering and it’s great that the US has such a variety of institutions to choose from (too much variety it feels like sometimes) but your tone is very critical towards the large majority of our engineer-producing institutions. They might not work for some people but they work for most.

2 Likes
9

I think the tone you’re sending is really just exasperation with CC. Sorry!

My goal is to get students and parents to take their ranking blinders off. Too many people are beholden to that list and don’t have the slightest idea what the differences are even among the ranked programs, let alone the non-PhD granting ones. The undergraduate experience will be radically different at MIT, Stanford and Berkeley, but how does one know that if they are looking at a list and nothing else?

As for your advisor not understanding the lab equipment, that’s because that wasn’t their role in that paradigm. That’s also important to know. The people at the pinnacle of research universities aren’t primarily tasked with teaching. They concentrate on their research. How does that translate to the undergrad? It’s a give and take, the subtleties of which are lost in the rankings.

So that’s the challenge. You can take the blue pill and follow the rankings blindly, or you can take the red pill as you did and see the whole, only then deciding which path is right. As Part 2 aims to show, it all works out really no matter where they go.

3 Likes
10

TA’s are usually the professor’s lackies in training. Whoever is the best lackie will get a chance to go for the PH.D. It really depends on how well the lackie TA does the professor’s bidding. I’m sure the lackie TA is expected to fail a few people who don’t deserve it to keep up appearances of class well run. It’s sort of the job no one wants because you have to do bad things, but it’s those bad things that you do to show you sold your soul that gets you a step up.

11

Not my experience.

2 Likes
12

That was my experience based on one professor’s class I took at UCSB. It was a humanities class. The TAs were a bunch of students with MA degrees going for their PH.D. I’m sure the TA job was to test them to see if they would make a good professor.

13

No worries!!! I have always appreciated your posts. And yes, CC can be frustrating but I personally learned a lot from it in my efforts to help guide my daughter. And a lot of what I learned was what to avoid!!!

And when I talk about the majority of our engineers, I’m not talking about Stanford and MIT grads. I’m talking about Iowa State, Clemson, UC Davis, Mines etc. grads.

I appreciate the conversation.

2 Likes
14

Ok. Sorry. I have no experience in the humanities. Only in engineering. I don’t teach anymore (at least not in an official role) but as a TA we wouldn’t even dream of grading exam papers. Lab reports, yes.

I’m sorry you had this experience.

15

You can sometimes find greater granularity, including specific colleges or courses elsewhere. For example, Stanford University Explore Courses indicates that during Spring 2019-20, Stanford’s intro programming class CS 106A had 666 students enrolled, with 105 sections. When I attended, each section was run by a separate TA, although I wouldn’t be surprised if they’ve increased to a particular TA having 2+ sections for this class.

This type of model is common for popular intro classes at Stanford. They have huge lecture classes that are taught by a well known expert in the field, then are divided down in to smaller sections where students have the opportunity for more personal attention. In the example CS 106A class, the average section size appears to be approximately 666/105 ~= 6 students. CS 106A intentionally keeps the section size small because being the first CS class most students take, there are a lot of non-majors who need extra attention, as well as various groups that Stanford has been making an effort to not discourage in intro CS. This size should be small enough to give students the opportunity to ask questions and such, which wouldn’t be possible in a huge lecture class with 666 students. So having the high number of TAs is certainly better than no TAs, although some may find the large lecture class size problematic, regardless of how small the section size is and how many TAs are available.

I’ve also taken intro STEM classes at one of the SUNYs. My experience at SUNY was quite different. Rather than have a huge lecture class taught by a well known expert in the field, some of the classes were smaller and taught by a PhD student. For example, my MV calc class was taught by a PhD student in math. There were no sections and instead many relatively small classes MV calc classes taught by PhD students and no primary faculty lecturer.

In some areas, I don’t need a well known expert to teach the class. For example, I don’t need a well known expert in mathematics to teach me basic calculus. Instead I need someone who is a quality teacher. In this class, I thought the PhD student was a great teacher. He wasn’t burnt out and seemed to be really enthusiastic about having the opportunity to teach the class.

However, I also liked having the opportunity to hear well known experts in the filed in some of the Stanford classes. For example, Mark Zuckerberg used to guest lecture in some of CS 106A classes mentioned above (maybe he still does). I particularly enjoyed hearing MacArthur Fellowship winner Robert Sapolsky teach an intro bio class, who is a great and inspirational lecturer with amazing stories about darting wild baboons in Africa and measuring hormones after observing social interactions, combined with texts that are enjoyable enough that many including myself read them for fun.

Either the SUNY or Stanford system could be more or less desirable, depending on the particular student, instructor, and class. This level of detail is difficult to understand from just comparing the total number of TAs to instructional staff.

1 Like
16

Indeed. Those are the experiences that only a few schools can offer. The Feynman Lectures didn’t happen at Cal State Fullerton. :stuck_out_tongue_winking_eye:

That said, they are all available as MOOCs now, not just at Stanford, but MIT and many others as well. When Sebastian Thrun started perfecting the concept, in person attendance in his class plummeted. When he asked his students come finals time why they quit showing up, they said it was because the MOOC format was better.

Intro to CS at UCB is literally that for some students. It’s taught in Wheeler Hall, but Wheeler isn’t big enough to hold the 1000+ students enrolled. Some have to sit in overflow rooms and watch a video feed.

I’m in no way suggesting MOOCs are a replacement for in person schooling. There are lots of things a MOOC cannot replicate, labs, camaraderie, etc. We are asking a lot of money for what is essentially a MOOC type of experience. A student from Iowa without need-based financial aid can pay for all 4 years at Iowa State for what they’d pay to attend an elite private school. That’s if they paid full sticker at ISU.

There is a third way though, small lectures, mitigating the need for discussion/recitation, taught by instructors with terminal degrees. They will bring experience that a graduate student doesn’t have yet. As an example, one of my son’s professors took a sabbatical from NASA to see if he’d like to teach and never went back. He taught lecture and lab. That is a different experience. Certainly, they could be bad teachers.

Each model has strengths and weaknesses.

17

However, sometimes aggressive boosterism (for a particular college or type of college) has the opposite effect.

Also, why choose CPSLO when you could have gotten WUE tuition at other CSUs with similar instructional organization, such as CSUC, CSUN, CPP, CSUSac? (Or are there even lower cost in-state publics that offer similar methods?)

18

Indeed and I’m trying to be more mindful of that.

This isn’t a CP defense thread, but I’ll answer your question. The CSU mission is different than that of the UCs. The focus is primarily on teaching. The polytechnic schools take that one step further with their learn-by-doing approach pioneered at the University of Chicago Lab School. At WPI this is accomplished through multiple collaborative projects. At Cal Poly it’s through labs and clubs. What really sets Cal Poly SLO apart are the facilities exclusively for undergraduate instruction and support, the faculty they attract and what they do with those resources. There are over 80 discrete lab spaces in the CENG alone. Having visited a lot of schools, they seem able to combine the strengths of large and small schools.

All that said, students who will thrive will do so no matter where they go.

19

CPSLO is just one of many CSUs that offers engineering majors.

All ABET-accredited engineering major programs have learn-by-doing in the form of engineering design course work. Also, CPSLO is not unique among CSUs in being named a polytechnic, since there is also CPP.

1 Like
20

That’s why I said polytechnic schools, plural. We’ll have to agree to leave it at that though. I don’t want this to devolve into a discussion about an individual school irrespective of my past postings.