Category Archives: Teaching

Why don’t schools teach debugging, or, more fundamentally, fundamentals?

January 16, 2024 By in Academia, Culture, Essays, Teaching Tags: , , , , 8 Comments

If you find this piece worthwhile, consider the Go Fund Me that’s funding ongoing cancer care.

A story from Dan Luu, from back when he “TA’ed EE 202, a second year class on signals and systems at Purdue:”

When I suggested to the professor that he spend half an hour reviewing algebra for those students who never had the material covered cogently in high school, I was told in no uncertain terms that it would be a waste of time because some people just can’t hack it in engineering. I was told that I wouldn’t be so naive once the semester was done, because some people just can’t hack it in engineering.

This matches my experiences: when I was a first-year grad student in English,[1] my advisor was complaining about his students not knowing how to use commas, and I made a suggestion very similar to Luu’s: “Why not teach commas?” His reasoning was slightly different from “some people just can’t hack it in engineering,” in that he thought students should’ve learned comma usage in high school. I argued that, while he might be right in theory, if the students don’t know how to use commas, he ought to teach them how. He looked at me like I was a little dim and said “no.” 

I thought and still think he’s wrong.

If a person doesn’t know fundamentals of a given field, and particularly if a larger group doesn’t, teach those fundamentals.[2] I’ve taught commas and semicolons to students almost every semester I’ve taught in college, and it’s neither time consuming nor hard. A lot of the students appreciate it and say no one has ever stopped to do so. 

Usually I ask, when the first or second draft of their paper is due for peer editing, that students write down four major comma rules and a sample sentence showcasing each. I’m looking for something like: connecting two independent clauses (aka complete sentences) with a coordinating conjunction (like “and” or “or”), offsetting a dependent word, clause, or phase (“When John picked up the knife, …”), as a parenthetical (sometimes called “appositives” for reasons not obvious to me but probably having something to do with Latin), and lists. Students often know about lists (“John went to the store and bought mango, avocado, and shrimp”), but the other three elude them.

I don’t obsess with the way the rules are phrased and if the student has gotten the gist of the idea, that’s sufficient. They write for a few minutes, then I walk around and look at their answers and offer a bit of individual feedback. Ideally, I have some chocolate and give the winner or sometimes winners a treat. After, we go over the rules as a class. I repeat this three times, for each major paper. Students sometimes come up with funny example sentences. The goal is to rapidly learn and recall the material, then move on. There aren’t formal grades or punishments, but most students try in part because they know I’m coming around to read their answers.

We do semicolons, too—they’re used to conjoin related independent clauses without a coordinating conjunction, or to separate complex lists. I’ll use an example sentence of unrelated independent clauses like “I went to the grocery store; there is no god.”

I tell students that, once they know comma rules, they can break them, as I did in the previous paragraph. I don’t get into smaller, less important comma rules, which are covered by whatever book I assign students, like Write Right!.

Humanities classes almost never teach editing, either, which I find bizarre. I suspect that editing is to debugging as writing is to programming (or hardware design): essential. I usually teach editing at the sentence level, by collecting example sentences from student journals, then putting them on the board and asking students: “what would you do with this sentence, and why?” I walk around to read answers and offer brief feedback or tips. These are, to my mind, fundamental skills. Sentences I’ve used in the past include ones like this, regarding a chapter from Alain de Botton’s novel On Love: “Revealed in ‘Marxism,’ those who are satisfying a desire are not experiencing love rather they are using the concept to give themselves a purpose.” Or: “Contrast is something that most people find most intriguing.” These sentences are representative of the ones first- and second-year undergrads tend to produce at first.

I showed Bess an early version of this essay, and it turns out she had experiences similar to Luu’s, but at Arizona State University (ASU):

My O-chem professor was teaching us all something new, but he told me to quit when I didn’t just understand it immediately and was struggling. He had daily office hours, and I was determined to figure out the material, so I kept showing up. He wanted to appear helpful, but then acted resentful when I asked questions, “wasting his time” with topics from which he’d already moved on, and which I “should already understand”.

He suggested I drop the class, because “O-Chem is just too much for some people.” When I got the second-highest grade in the class two semesters in a row, he refused to write me a letter of recommendation because it had been so hard for me to initially grasp the material, despite the fact that I now thought fluently in it. My need for extra assistance to grasp the basics somehow overshadowed the fact that I became adept, and eventually offered tutoring for the course (where I hope I was kinder and more helpful to students than he was).

Regarding Bess’s organic chemistry story, I’m reminded of a section from David Epstein’s book Range: How Generalists Triumph in a Specialized World. In his chapter “Learning, Fast and Slow” Epstein writes that “for learning that is both durable (it sticks) and flexible (it can be applied broadly), fast and easy is precisely the problem” (85). Instead, it’s important to encounter “desirable difficulties,” or “obstacles that make learning more challenging, slower, and more frustrating in the short term, but better in the long term.” According to Epstein, students like Bess are often the ones who master the material and go on to be able to apply it. How many students has that professor foolishly discouraged? Has he ever read Range? Maybe he should.

Bess went on:  

Dan’s story also reminds me of an attending doctor in my emergency medicine program; she judged residents on what they already knew and thought negatively of ones who, like me, asked a bunch of questions. But how else are you supposed to learn? This woman (I’m tempted to use a less-nice word) considered a good resident one who’d either already been taught the information during medical school, or, more likely, pretended to know it.

She saw the desire to learn and be taught—the point of a medical residency— as an inconvenience (hers) and a weakness (ours). Residency should be about gaining a firm foundation in an environment ostensibly about education, but turns out it’s really about cheap labor, posturing, and also some education where you can pick it up off the floor. When I see hospitals claiming that residency is about education, not work, I laugh. Everyone knows that argument is bullshit.    

We can and should do a better job of teaching fundamentals, though I don’t see a lot of incentive to do so in formal settings. In most K – 12 public schools, after one to three years most teachers can’t effectively be fired, due to union rules, so the incentive to be good, let alone great, is weak. In universities, a lot of professors are, as I noted earlier, hired for research, not teaching. It’s possible that, as charter schools spread, we’ll see more experimentation and improvement at the ˚K –12 level. At the college and graduate school level, I’d love to see more efforts at instructional and institutional experimentation and diversity, but apart from the University of Austin, Minerva, the Thiel Fellowship, and a few other efforts, the teaching business is business-as-usual.

Moreover, there’s an important quirk of the college system: Congress and the Department of Education have outsourced the credentialing of colleges and universities to regional accreditation bodies. Harvard, for example, is accredited by “The New England Association of Schools and Colleges (NEASC).” But guess who makes up the regional accreditation bodies? Existing colleges and universities. How excited are existing colleges and universities to allow new competitors? Exactly. The term for this is “cartel.” This point is near top-of-mind because Marc Andreessen and Ben Horowitz emphasized it on their recent podcast regarding the crises of higher education. If you want a lot more, their podcast is good.

Unfortunately, my notions of what’s important in teaching don’t matter much any more because it’s unlikely I’ll ever teach again, given that I no longer have a tongue  and am consequently difficult to understand. I really liked (and still like!) teaching, but doing it as an adjunct making $3 – $4k / class has been unwise for many years and is even more unwise given how short time is for me right now. Plus, the likelihood of me living out the year is not high.   

In terms of trying to facilitate change and better practices, I also don’t know where, if at all, people teaching writing congregate online. Maybe they don’t congregate anywhere, so it’s hard to try and engage large numbers of instructors.

Tyler Cowen has a theory, expounded in various podcasts I’ve heard him on, that better teachers are really here to inspire students—which is true regarding both formal and informal education. Part of inspiration is, in my view, being able to rapidly traverse the knowledge space and figure out whatever the learner needs.

Until we perfect neural chips that can download the entirety of human knowledge to the fetal cortex while still in utero, no one springs from the womb knowing everything. In some areas you’ll always be a beginner. Competence, let alone mastery, starts with desire and basics.

If you’ve gotten this far, consider the Go Fund Me that’s funding ongoing care.

[1] Going to grad school in general is a bad idea; going in any humanities discipline is a horribly bad idea, but I did it, and am now a cautionary tale for having done it.

[2] Schools like Purdue also overwhelmingly select faculty on the basis of research and grantsmanship, not teaching, so it’s possible that the instructors don’t care at all. Not every researcher is a Feynman, to put it lightly.

How To Efficiently Advance Your Software Engineering Career [Guest Post]

October 11, 2023 By in Essays, Teaching Tags: , , , 3 Comments

This guest essay is by my former student Nathan Lin.

At age 30 I was able to retire from my software engineering career to pursue my interest in day trading. I had advanced rapidly as a software engineer, climbing through multiple promotion cycles. Combined with some good luck and timing, by my 30th birthday I had enough of a bankroll saved for the next 5 years.

I didn’t start my career in Big Tech, although I did end my almost decade-long career there. I’ve worked in various sized companies with as few as 50 engineers to over 30,000. What I discuss here should work regardless of the size of company.

I didn’t knowingly create this framework of expectation management at the time. Only by reflecting back on my career and identifying the recurring themes am I now able to put a name and a method to it. It’s largely inspired by Jake’s essay on How to get coaching, mentoring, and attention, which I read in college, and changed the trajectory of my academic and professional career.

My framework of expectation management is built on the same foundations, and is like a “Part 2” applied to software engineering. 

Cycles

A key part of the framework of expectation management requires you to understand cycles. Cycles continuously repeat themselves throughout your career, and it is imperative that you understand:

  1. What cycle are you at in your career? (The primary cycle)
  2. What cycle are you at in your specific job? (The secondary cycle)

Just as the seasons change and pass, so too does your location in the cycle change, and that location determines how you must manage expectation and performance. Note that the primary cycle also applies to other corporate 9 ­– 5 jobs:

  1. The novice (0-2 years)
  2. The intermediate (2-4 years)
  3. The expert (4-10 years)
  4. The guru (10+ years)

These cycles apply not only to you where you are at in your career, but also where you are at in your job. I discovered this first-hand as I followed the software engineering maxim of moving every 18 months to maximize compensation. There’s always a “rampup period” where, even though you may have a few years of industry experience, you’re given time to learn your new employer’s technology stack.

After that period, you’re expected to perform at a meaningful level. And after even more time, you’ll be expected to handle expert-level questions and solve tough problems. The cycle is compressed from years to months compared to the career cycle. I call this the secondary cycle:

  1. The novice (1-4 months)
  2. The intermediate (4-12 months)
  3. The expert (12 – 36 months)
  4. The guru (36+ months)

If you want to advance rapidly, you must know your location in both cycles, and destroy the expectations placed on you. The expectations are cycle-driven and tacitly acknowledged. The industry itself is aware of these cycles, as this is how roles and expectations are defined. However, these cycles are never really explicitly discussed, which is what makes exceeding expectations and rapidly advancing difficult. I’m trying to make tacit knowledge explicit. Most CS programs don’t include class called “the software engineer career,” but maybe they should.

The secondary cycle is reflexive, based on your location in the primary cycle. So, if you’re a novice who just graduated college as a 22-year old, you may be given 4 months at your first job to be brought up to speed. Some companies will even give slightly more. But as you progress to intermediate and expert, you’ll be expected to ramp up faster.

After all, if you’ve been in software for 8 years, why would it take you 4 months to ship your first meaningful pull request?

Expectations

Understanding the actual expectations is the last piece before getting to the framework. The framework itself is quite easy to apply once you understand cycles and expectations.

You must understand the cycle you’re in so that you figure out the expectations for both the primary and secondary cycles. They’re pretty much identical, so I’ll discuss the primary cycle:

  1. The novice – junior engineer
    1. Requires supervision and frequent check-ins
    1. Needs to be told what to do next
    1. Sees the 10 foot view
  2. The intermediate – midrange engineer
    1. Requires minimal supervision
    1. Capable of self-direction and identifying the next task in a larger project
    1. Sees the 100 – 1,000 foot view
  3. The expert – senior engineer
    1. Supervises intermediates and novices
    1. Plans entire projects to solve business needs
    1. Sees the 1,000 – 10,000 foot view
  4. The guru – staff/principal engineer
    1. Supervises experts, intermediates and novices
    1. Plans large-scale initiatives that break down into projects, coordinates with multiple teams to do so
    1. Sees the 10,000 – 100,000 foot view

Another way of viewing the cycle: at the novice level, you’ll always have someone you can ask for help. If you get stuck, ask your mentor/team lead/a senior engineer. But as you advance, you work on harder and harder problems and the number of people you can ask for help approach 0 at the expert and guru levels.

When you get to those levels, you are the one person you can approach for help, and you must ultimately figure out the answers by yourself. Note that you can (and should) consult other experts for ideas. Peers often spark new perspectives and ideas. In the end, however, you’ll have to do the heavy lifting yourself.  

The Framework

The framework is actually very straightforward:

  1. Find your location in the career cycle
  2. Find your location in the secondary cycle
  3. Exceed expectations

Now, you don’t necessarily have to exceed expectations. This really depends on your goals and how motivated you are. You don’t have to do it the way I did it, working 60+ hours a week to climb furiously. I had a goal in mind and I was determined to get after it. 

If you enjoy the 9-5 corporate life, there’s nothing wrong with meeting, instead of exceeding, expectations as the third step. The 9 – 5 corporate life can yield a great career, and if you know which companies have a healthy work life balance, you can have a very fulfilling job.

How do you exceed expectations? I only gave a high-level summary, so you may have some trouble with the details. But the rough idea can be found in Jake’s essay: prove to your team, especially your team lead and manager, that you’re here to get shit done. 

And the only way to prove that is to actually do it. This is where the rubber meets the road, and where the buck stops. Delivering results is the measure by which all top-level performers measure themselves and others by. So you must start speaking their language. The easiest way to do this is to deliver a small result as fast as possible, in your first week if you’re experienced, or your first month if you are a new graduate.

I’ve found fixing a low-risk bug, or improving test coverage especially reliable.

This is where the concept of cycles comes into play if you want to do this efficiently. You can quickly demolish expectations when you first arrive at a new job. Expectations are low, your team is giving you space to learn, so this is therefore the most logical place for you to make a first impression.

The beautiful thing is that this small result compounds. You’ve established a foothold in the codebase, and it’s now easier to work outwards from there. There may be some starting effort to establish the initial foothold, but it’ll pay off as the next pull request becomes easier. And as you continue to add code and make improvements, you’ll end up holding entire swaths of the codebase in your head.

Word on the team will travel fast. Good engineers are hard to find, and your mentor will mention your results to your manager. Your manager will begin to pay attention to your development. You may end up working on projects designed to promote you to the next level. And you may work with the heavy hitters on your team, and it won’t be a coincidence.

Skills compound over time. Mentorship opportunities open up to those most worthy of them. Life isn’t always fair, but over time, merit wins more often than not.

Promotions

If you want a promotion you must exceed expectations. At a certain point in time, you must begin signaling the attributes of the next rank above you, which requires two things. First, you must be proficient in your existing role, and second, you must begin to model the people on your team who are already at that next level.

Reverse engineer the next role, and ask yourself: “What does someone in that role do that I don’t?” So if you’re a junior engineer and you wish to become a midrange engineer, start proactively taking on tasks. Identify the next logical step to do without waiting for your mentor/team lead to give you the task. 

There’s an element of critical thinking here where you should ask, “What’s the next thing I should work on to efficiently move the project towards the finish line?”

Ship some code, then go back and refactor it, or move to the next task: “This is in a part of the codebase I was just in, I’m already familiar with these files.” 

Model the people who are already at the level you want to be at. If you think the way they think and code the way they code, then you’ll automatically begin operating at their level. They’ll see how you do it in code review and they’ll think, “Wow, she coded it the same way I would have. She’s pretty smart”. These people are also usually the ones your manager will solicit to determine if you’re ready. 

This brings me to my promotion rule: The more frequently you signal you are ready, the faster you’ll get promoted.

The Resting Rate

I haven’t talked about when to take a break. If you continue to smash expectations day after day, week after week, you’ll undoubtedly burn out (I speak from experience). When’s the ideal time to take a breather? There are actually a few places to strategically do so.

The first is after your team ships a major project. Everyone is happy: the project is complete, the deadline has been met. On most teams, the workload lightens substantially. Don’t do what I did and take on more work to keep padding your resume. There will be plenty of that down the line, so take a breather.

The second is after advancing through a phase in the cycle. But doesn’t this look bad if you get promoted and your work output drops? No, because never in my career have I seen someone demoted or put on a plan right after they were promoted. Your manager and skip level have gone through weeks/months of work to promote you, HR is involved with your compensation increase, so demotion or similar would look terrible on all parties involved. 

Now, I’m not saying to slack off and do nothing. But take a breather. If it doesn’t get done today, that’s okay. Your career will last for decades. This is something I did not really understand. 

I thought that in order to advance, I had to signal I was at the next step every single day. I suspect this is something that plagues a lot of other high performers in tech, causing burn out.

It’s true that the more you signal you’re at the next level, the faster you’ll get noticed and promoted. But there’s also a ceiling to the amount of signal you give off. There’s a point at which all the additional signal you produce will be ignored.

This is due to factors outside of your control, such as bureaucracy or company/HR policy. So you should really push as close as you can to this ceiling, and then no further. Additional effort at this point will be wasted. If you want to make it to staff/principal level, this is really important. You don’t want to be burned down to the ground by the time you make it. 

Why It Works

Several key quotes from Jake’s essay have stood the test of time in my career and life in general:

“…professors (and others with knowledge and competence) are most inclined to help people who won’t waste their time.”

“If a student really wants to learn, the professor will usually help, but most students don’t—so the professor builds a wall between herself and her students to make sure that the only students who breach the wall are the ones who do care about learning.”

“Secondarily, your professor will often recommend reading to test your seriousness…If you go away and don’t come back, you’ve demonstrated that you would’ve wasted her time had she spent an extra hour talking to you outside of class and office hours.”

This is why I recommend you immediately fix a small bug or improve code coverage. You’re going to be asked to do this anyway. Everyone is busy, and no one has time to sit you down (unless you are a recent grad) and walk you through the codebase. If you do this proactively, you’ll have taken one more item off their plate and shown, in a small way, that you’re worth their time.

The wall to getting a top performer to mentor you might even be more difficult to breach than a professor. Every minute of time spent talking to you is a minute she could be spending on advancing her career for a big raise or promotion.

You’ll often be shooed away with a simple bug fix. And if you come back asking trivial questions that can be answered by the README or a cursory glance at the code, their guard immediately goes up. How are you worth their time if you demonstrate you’re not capable of the basics?

But if you come back with a pull request, and a follow-up question about the architectural design of the system, now you’re showing you’re worth their time. You’ve produced the result and you’ve dug deep enough into the code to ask a harder question.

You have shown you are worth their time, maybe they could learn something from you. 

“People who really want to do something… do it. Or they make changes so they can… But most people say they want to do something and then they don’t. Over time, others notice this (like me), and they start to assume that most people who say they want to do or know something are full of shit.”

This is why you must signal you are at the next level for a promotion, and why you must signal often. 

Do you know how many times a year your manager gets asked for a promotion? It happens all at once, during the annual performance review, and it won’t get brought up again until the next year. Most people will get rejected and complain to their spouse, “Ah my damn manager doesn’t want to give me that promotion”.

But if you make it a recurring theme in your 1:1s, and you ask, “How am I progressing towards the next level?”, then the question sounds very different. Don’t just ask your manager to be put on a project. Find a problem that needs to be solved, come up with a plan of attack, and then build a proof of concept.

People react in one of two ways when expectation does not match reality. When reality is worse, they’re angry and disappointed. When it’s better, they are happy and excited. And that’s why expectation management works, because you’re demonstrating that reality is better. 

Completing The Cycle

I don’t want to drag this out and turn it into career management advice, so I’ll end on a few notes on advancing as efficiently as possible. 

First, you have to advance through the cycle no matter what. You can’t remain a novice forever or you’ll get fired. At best you’ll never reach your potential. If you’ve been in the industry for 10 years but produce the results of a novice, you’ll be viewed with suspicion because you’re breaking the expectation in the wrong direction.

Second, you don’t have to push as hard as I did if you don’t want to. The key is to work backwards and figure out your goals, then match your signaling frequency to that goal. 

There will be so many more projects down the line. Don’t tunnel vision on padding your resume and working 24/7. It’s okay to rest, and relax, because this is the natural cycle of things. 

Third, you have to know when, where, and how much to signal. 

When you are fighting to make it to expert status, you will need to signal more often. You have the energy and vitality to do so when you’re young. When you reach expert status, the game is more about conserving energy. You can decide whether or not you want to reach true guru status, or stay at your current level. Both are acceptable.

Thanks for reading. I hope this helps you, both in your career, and in your life, the way Jake’s essay helped me. 

Most people don’t read carefully or for comprehension

January 31, 2022 By in Academia, Teaching Tags: , , , 4 Comments

Dan Luu has a great Twitter thread about “how few bits of information it’s possible to reliably convey to a large number of people. When I was at MS, I remember initially being surprised at how unnuanced their communication was, but it really makes sense in hindsight” and he also says that he’s “noticed this problem with my blog as well. E.g., I have some posts saying BigCo $ is better than startup $ for p50 and maybe even p90 outcomes and that you should work at startups for reasons other than pay. People often read those posts as ‘you shouldn’t work at startups’.” In other words, many people are poor readers, although “hurried” or “inattentive” might be kinder word choices. His experiences, though, are congruent with mine: I’ve taught English to non-elite college students, off and on, since 2008; when I first started, I’d run classes by saying things like, “What do you all think of the reading? Any comments or questions?” I’d get some meandering responses, and maybe generate a discussion, but I often felt like the students were doing random free association, and it took me an embarrassingly long time to figure out why.

After a semester or two I began changing what I was doing. An essay like Neal Stephenson’s “Turn On, Tune In, Veg Out” is a good demonstration of why, and it’s on my mind because I taught it to students recently (you should probably read “Turn On, Tune In, Veg Out” first, because, if you don’t, the next three paragraphs won’t make a lot of sense—and you’re the kind of person who does the reading, right?). Instead of opening by asking “What do you think?”, I began class by asking, “What is the main point of ‘Turn On, Tune In, Veg Out?’” Inevitably, not all students would have done the reading, but, among those who had, almost none ever have, or give, good answers. Many get stuck on the distinction between “geeking out” and “vegging out,” even though that’s a subsidiary point. Some students haven’t seen or dislike Star Wars, and talk about their dislike, even though that’s not germane to understanding the essay.

Stephenson says at least three times that Star Wars functions a metaphor: once in the third paragraph, once in the second-to-last paragraph—although that technically compares the Jedi to scientists, rather than Star Wars as a whole to society—and again at the end (“If the ‘Star Wars’ movies are remembered a century from now, it’ll be because they are such exact parables for this state of affairs”). Most students don’t know what a “parable” is, which also means I wind up asking what they should do if they come across a word they don’t know. It’s also not like the essay is long or using numerous complex words: it’s only about 1,300 words and it’s about pop culture, not some abstract topic.

The first few times I taught “Turn On, Tune In, Veg Out” this way, I wondered if I was getting unrepresentative samples, but I’ve done it many times since and have consistently gotten the same results. I think most high-school students, to the extent they’re being taught to read effectively at all, are being taught to skim a work for keywords and then vomit up an emotional reaction (I assign free-form, pass-fail student journals, and most take this form). Very few students seem to be taught close reading, although when I was still in grad school, I had a cluster of students who all had had the same junior or senior year high school teacher, and that teacher had drilled all of them in close reading and essay writing—and they were all proficient. She seemed to be the exception, not the rule, and I meant to send her a letter thanking her but never did. Teaching “Turn On, Tune In, Veg Out” usually takes somewhere between 45 minutes and an hour, in order to go through it and look at how the essay is constructed, how the sentence “What gives?” functions as a turning point in it, and other related topics. I tell students at the end of the process that we’ve not talked about whether they like “Turn On, Tune In, Veg Out” or not; the goal is to understand it first, and evaluate it later. Understanding before judgment: Internet culture encourages precisely opposite values, as I’m sure we’ve all seen in social media like Twitter itself.

At the end of class, I ask again, “What is the main point?” and get much better answers. I’ll sometimes do the same thing with other argumentative essays, and often the initial answers aren’t great. I posit that most students aren’t being taught close reading in high school, and part of that theory comes from me asking them, individually, what their high school English classes were like. Many report “we watched a lot of movies” or “nothing.” Sure, a few students will have taken “nothing” from excellent classes and instructors, but the answers are too uncomfortably common, especially from diligent-seeming students, for me to not see the pattern. In high school, few students seem to have looked closely at the language of a given work and how language choices are used to construct a story or argument. To my mind, and in my experience, doing that is a prerequisite for being a proficient writer, including on topics related to “social justice.”

It’s not just “turn On, Tune In, Veg Out;” when I assign Orwell’s “Politics and the English Language,” I’ll get strange responses from students about how it’s so totally true that these days the English language is being used poorly. After enough of those kinds of responses, I began to open class by asking students to take 20 or 30 seconds to write down when “Politics” was written. In case you think this is a trick question, “1946” is displayed in huge font at the start of the version I’m using, and it’s repeated again at the very end. In the text itself, Orwell cites a Communist pamphlet, and he mentions the “Soviet press,” and such choices should be clues that it’s not contemporary. Nonetheless, if a third of a given class gets in the right ballpark—pretty much anything between “1930s” and “1950s” is adequate enough for these purposes—that’s good, which implies two-thirds of a given class hasn’t done the reading or hasn’t retained what I’d call an elemental idea from the reading. Students routinely guess “2010s” or “2000s.”

Right after college I taught the LSAT for two years, and the LSAT is largely a test of reading comprehension. I worked for an independent guy named Steven Klein, who’d started his company in the late ‘80s or early ’90s, before Kaplan and Princeton Review became test-prep behemoths. He and his business partner, Sandy, would marvel at the students who had 3.8, 3.9, sometimes 4.0 GPAs in fields like sociology, communication, English, or “Law, Society, and Justice” but who couldn’t seem to understand even simple prose passages. The students would get frustrated too: they were college grads or near college grads, who were used to being told they were great. The LSAT experience made me a sympathetic reader of the book Paying for the Party, or Beer and Circus: How Big-Time College Sports Has Crippled Undergraduate Education, both of which describe how most colleges and universities have evolved vast party tracks that require minimal skill development and mental acuity, but reliably deliver high grades. I think of those books when I read about the massive, $1 trillion and growing amount of outstanding student loan debt. Many college and university students would be better served with apprenticeships and vocational education, but as a society we’ve spent 40, if not more, years disparaging such paths and exalting “college.” Articles like “41% of Recent Grads Work in Jobs Not Requiring a Degree” are common. We have many bartenders and airline stewards and stewardesses and baristas who’ve obtained expensive degrees: I’m not opposed to any of those professions and respect all of them, but a four-year degree is a very expensive way of winding up in them.

The LSAT is a standardized test, and many schools still like standardized tests because those tests aren’t changed by how rich or connected or otherwise privileged a person is. Some Ivy-League and effectively-Ivy-League schools are doing away with the SAT, in the name of “diversity,” but that usually means they’re trying to give themselves even more discretion in “holistic” admissions, which tends to mean rich kids, with a smattering of diversity admits for political cover. “Race Quotas and Class Privilege at Harvard: Meme Wars: Who gets in, and why?” is one take on this topic, although numerous others can be found. The students who had gotten weak degrees and high GPAs were flummoxed by the LSAT; when they asked what they could do to improve their reading skills, Steven and Sandy often told them, “Read more, and read more sophisticated works. The Atlantic, The New Yorker [this was a while ago], better books, and do it daily.” I’d sometimes see their faces fall at the notion of having to read more: they were hoping to learn “One Weird Trick For Improving Your Reading Skills. You Won’t Guess What It Is!” When I’ve taught undergrads, they often want to know if there’s a way to get extra credit, and I tell them to do the reading thoroughly and write great essays, because I will grade based on improvement. This seems particularly important because many haven’t been taught close reading or sentence construction. I also see the disappointment in their faces and body language, because they think I’m going to tell them the secret, and instead I tell them there is no real secret, just execution and practice. A lot of school consists of jumping through somewhat ridiculous, but well-defined, hoops, and then being rewarded for it at the end, but real learning is much stranger and more tenuous than that. Sarah Constantin argues that “Humans Who Are Not Concentrating Are Not General Intelligences,” which is consistent with my experiences.

Many, if not most, English and writing professors also seem strangely uninterested in teaching writing or close reading. I get peculiar looks when I talk about the importance of either with other people teaching writing or English; one woman at a school I taught at in New York told me that social justice is the only appropriate theme for freshman writing courses. I know what she meant, and grunted noncommittally; I didn’t really reply to her at the time, although I was thinking: “Isn’t developing high levels of skill and proficiency the ultimate form of social justice?”

This is a long-winded way of saying that poor reading comprehension may be closer to the norm than the exception, and that may also be why, as Dan observed, very few bits of information trickle down from the C suites in big companies to the line workers (“I’ve seen quite a few people in upper management attempt to convey a mixed/nuanced message since my time at MS and I have yet to observe a case of this working in a major org at a large company (I have seen this work at a startup, but that’s a very different environment)”). I’d imagine the opposite is also true: if you’re a line worker, or lower-level management, it’s probably difficult or impossible to tell the C suite people about something you think important. Startups can disrupt big companies when a few people at the startup realize something important is happening, but the decision makers and the BigCo don’t.

I’ve also learned, regarding teaching, a message similar to what the MS VPs had learned: not much goes through, and repetition is key. One time, my sister watched me teach and said after, “You repeat yourself a lot.” I told her she was right, but that I’d learned to do so. Teachers and professors repeating themselves endlessly made me crazy when I was in school, but now I understand why they do it. I’ll routinely say “Do [stuff] for Thursday. Any questions?” and have someone immediately say: “What should we do for Thursday?” There’s a funny scene in the movie Zoolander in which the David Duchovny character explains to the Ben Stiller character how male models are being used to conduct political assassinations. He goes through his explanation, and then Zoolander goes: “But why male models?” The David Duchovny character replies: “Are you stupid? I just explained exactly that to you.” Derek Zoolander is a deliberately stupid character, but I think inattention is probably the most relevant explanation in the real world. Big tech companies like Microsoft probably have very few stupid people in them. Most students aren’t stupid, but I think many haven’t been effectively challenged or trained. It’s also harder for the instructor to teach close reading than it is to have meandering discussions about how a given work, which has probably been at best skimmed, makes students feel. I’ve written on “What incentivizes professors to grade honestly? Nothing.” There’s a phrase that floats around higher education about a rueful compact between students and teachers: “They pretend to learn, and we pretend to teach.” Students, I’m sure you’ll be shocked to know, really like to get good grades. I of course grade with scrupulous honesty and integrity 100% of the time, just like everyone else, but I have heard rumors that there’s temptation to give students what they want and collect positive evaluations, which are often used for hiring and tenure purposes.

Politicians appear to have learned the same thing about repetition and the limits of the channel: the more successful ones appear to develop a simple message, and often a simple phrase (“Hope and change,” to name a recent one: you can probably think of others) and repeat it endlessly, leaving the implementation details to staff, assuming the politician in question is elected.

When Paul Graham confronts readers mis-reading his work, he’ll often ask, “Can you point to a specific sentence where I state what you say I state?” It appears almost none do. Even otherwise sophisticated people will attribute views to him that he doesn’t hold and hasn’t stated, based on the mood his essay creates. In Jan. 2016, for example, he wrote “Economic Inequality: The Short Version” because he saw “some very adventurous interpretations” of the original. In April 2007, he wrote “Microsoft is Dead: The Cliffs Notes” because many interpreted his metaphor as being literal. I often teach a few of his essays, most notably “What You’ll Wish You’d Known,” and some students will report that he’s “arrogant” or “pretentious.” Maybe he is: I’ll ask a version of the question Graham does: “Can you cite a sentence that you find arrogant or pretentious?” Usually the answer is “no.” I tell students they could write an essay arguing that he is, using specific textual evidence, but that never happens.

I’ve told bits and pieces of this essay to friends in conversation, and they sometimes urge me to try and make a difference by making an effort to improve college teaching. I appreciate their encouragement, but I don’t run any writing or English departments and have a full-time job that occupies most of my time and attention. I like teaching, but teaching represents well under 10% of my total income, tenure-track jobs in humanities fields haven’t really existed since 2009, and adjunct gigs offer marginal pay. To really encourage better classroom teaching, schools would need to pay more and set up teaching systems for improving classroom teaching. The goal of the system is to propagate and perpetuate the system, not to disturb it in ways that would require more money or commitment. Pretending excellence is much easier than excellence. I’m okay with doing a bit of teaching on the side, because it’s fun and different from the kind of computer work I usually do, but I’m under no illusions that I’m capable of changing the system in any large-scale way. The writing I’ve done over the years about colleges and college teaching appears to have had an impact on the larger system that’s indistinguishable from zero.

Teachers and the income ceiling

October 15, 2019 By in Teaching Tags: , 2 Comments

The teacher pay gap is a myth? Maybe. This one has a different spin than the typical unthinking “teachers are underpaid” articles (it argues they’re not, and yes there is data to support the assertion), which is why I point to it specifically, but I think both sides are underestimating an important point about flexibility and possibility in any given career; teaching has a relatively low pay ceiling and even if a given teacher wants to work twice as hard and make twice as much money, that’s really difficult and potentially impossible. High IQ and high conscientiousness persons in other professions, meanwhile, can maximize incomes to a much greater extent than teachers. For example, as a consultant there are various avenues I can explore to further raise my income, but if I taught high school I couldn’t. In big metro areas, too, there are also many more employers to choose from than most teachers can choose from, since there’s usually one district, or a handful in nearby areas; at the same time, high rents in many superstar cities make teaching less attractive, so the “teacher pay” story is also a story about land-use regulations that raise the cost of housing—it’s just never discussed in those terms.

Teaching has other challenges: bureaucracy is a real problem and so is low morale, perhaps related to bureaucracy. In most sectors, if you don’t like the bureaucracy at one company you can switch to another, but this is considerably harder for teachers, since most districts are local monopolies. Financial returns to IQ also seem to be going up and at the same time we’re making housing in many cities and first-tier suburbs out-of-control expensive—so teachers are really taking it financially, from both the cost-of-living side and from the payment side. I have heard through the grapevine, for example, that pretty much every new public school teacher in Seattle says the same thing about how she got there: “My husband got a job at Amazon and…” That’s no doubt an exaggeration, but how much of one?

To use myself as an example, my parents moved from California to a boring suburb when I was a kid, and they left California because the cost of housing was so high. I checked how much they paid for the house in the boring suburb; in inflation-adjusted terms, they paid about $350,000 in today’s dollars. The same house is estimated to be worth about $700,000 today, so the cost has just about doubled in real terms, but that suburb forbids townhouses in the vast majority of its land, so no one can buy the house, knock it down, and put two houses on the same lot. If that were legal, we’d not have the housing crisis we do in many high-productivity cities and their suburbs. If we could lower the overall cost of living, struggles around teacher pay would seem less dire. Oregon, for example, has legalized duplexes and fourplexes statewide, so it will offer a natural experiment in lowering housing costs.

Let’s return to the original link:

shortages exist precisely where expected in a nationwide labor market that pays an increasing premium for STEM and other specialized skills. When researchers have examined the teaching vacancies that districts say they have trouble filling, they find that elementary, English, and social-studies teachers are not the problem. In fact, the Department of Education found that 20 states and the District of Columbia produced over twice as many elementary-education graduates as they had elementary-teaching positions to fill. At the same time, some districts struggle to fill STEM and special-education positions.

It’s hard to measure teacher “productivity” and yet almost no one bothers trying. So the most productive teachers are incentivized to move somewhere they can be paid in line with their productivity, which can’t be at most schools. Talk about an adverse selection problem! We have problems in data and definitions in both health care and teaching; rarely have two fields absorbed so much concentrated thought and produced so little change. The data and definition problems make it much harder to evaluate quality, inputs, and outputs than in other fields, especially those related to manufacturing. In most businesses, a major goal is simple (profit) and measurement is comparatively simple. The output of high-quality teaching may not manifest itself for decades, and very little of the value improvement is captured by the educator. By contrast, if you build a better widget, there’s a decent chance you’ll be able to capture more of the widget value beyond the cost of production.

Because we’re not measuring, or able to measure, teacher quality effectively, most schools also seem to pay based on years of experience. Anecdotally, I’ve been told that it can be very hard to get hired by another public school after 10 – 15 years of experience, because then you’re too expensive and most schools would rather hire cheaper teachers. So this is another way of ossifying the labor market.

There’s also a narrative about teachers working a lot of hours per week, but Bureau of Labor Statistics data consistently show that teachers work 40 hours a week. One example, although I have seen many other similar ones. Anecdotes consistently run one way, and data consistently runs another.

Overall, however, it seems that the number of people trying to have a completely honest conversation around this topic is not high, and many who argue that we should pay teachers “more” don’t know what “more” looks like. We may also see lots of composition effect problems, in which people who can get higher wages or better working conditions leave and get them, while those who can’t stay at school districts and take what they can get, which is at least consistent with some of the teachers I had in high school, but that may not be desirable at the societal level.

Teaching demands starting where comprehension ends

February 11, 2019 By in Links, Teaching Tags: Leave a comment

“How Craig Barton wishes he’d taught maths” is from Timothy Gowers‘s blog, and many sections are not unique to math; they apply to teaching almost anything. Like this:

I’m jumping around a bit here, but a semi-counterintuitive idea that he advocates, which is apparently backed up by serious research, is what he calls pretesting. This means testing people on material that they have not yet been taught. As long as this is done carefully, so that it doesn’t put students off completely, this turns out to be very valuable, because it prepares the brain to be receptive to the idea that will help to solve that pesky problem. And indeed, after a moment of getting used to the idea, I found it not counterintuitive at all.

In English, “pretesting” as such is often not possible, but it’s useful to attempt to gauge students’s knowledge and go back to wherever the student is confused—which may be very simple aspects of language, like parts of speech. I often had debates about this subject in grad school, when other grad students or professors would lament students’s weak grasp of “basics” or “fundamentals” like comma rules. The stern professors had a point, in that university students should know those things, but I would counter that, if students don’t know them, it’s useful to teach them, even in “advanced” classes. Sometimes students seem to have not been taught much of anything in high-school English classes. Many high-school English classes have devolved into discussions of feelings and vague hand-waving about a given book, and students emerge from them with few concrete skills.

To be sure, sometimes the opposite is true. While teaching in grad school, I had a series of students, all good writers, all of whom had been taught by a particular teacher in a particular high school, and she apparently really drilled students in close reading and essay construction, like someone out of “The Writing Revolution.” The results showed. I meant to send her a letter thanking her but never did. I would guess that she did a form of “pretesting,” albeit without multiple-choice questions, to ascertain students’s skill levels and then base each day in class on what students know. I used to do something similar at times, by doing quick yes/no questions based on raised hands, in order to get a sense of where students were. Now, reading “How Craig Barton wishes he’d taught maths,” I think I should have spent more time and energy on assessment.

In most if not all subjects, it’s not possible to teach (or learn) advanced topics without mastering fundamentals, so an instructor should go back to wherever someone lacks mastery and begin building up from there. If that doesn’t happen, students—in the broadest sense, even outside formal school—at most muddle through and at worst waste everyone’s time. It’s nice to see someone as eminent as Timothy Gowers coming to a similar conclusion.

The Case Against Education — Bryan Caplan

March 12, 2018 By in Academia, Books, Nonfiction, Teaching Tags: , , 20 Comments

The Case Against Education is a brilliant book that you should read, though you’ll probably reject its conclusions without really considering them. That’s because, as Caplan argues, most of us are prone to “Social Desirability Bias:” we want to say things that are popular and make people feel good, whether or not they’re true. Some true things may be socially desirable—but many false things may be too; the phrase “Don’t shoot the messenger” exists for a reason, as does the myth of Cassandra. We like to create scapegoats, and messengers are handy scapegoats. Simultaneously, we don’t like to take responsibility for our own ideas; and we like to collectively punish iconoclasts (at first, at least: later they may become idols, but first they must be castigated).

Caplan is an iconoclast but a data-driven one, and that’s part of what makes him unusual and special. And, to be sure, I myself am prone to the biases Caplan notes. Yet, as I read The Case Against Education, I couldn’t find many holes to poke in the argument. The book blends data and observation / anecdote well, and it also fits disturbingly well with my own teaching experiences. For example, Caplan notes that students find school boring and stultifying: “Despite teachers’ best efforts, most youths find high culture boring—and few change their minds in adulthood.” While “school is boring” seems obvious to most people, it’s also worth asking why. Many of the reasons Caplan gives are fine, but I’ll also add that “interesting” is often also “controversial,” and many controversial / interesting instructors will take heat, as I argue in “Ninety-five percent of people are fine — but it’s that last five percent:”

Almost no teacher gets in trouble for being boring, but a teacher can get in trouble or can get in trouble for being many values of “interesting.” Even I’ve had that problem, and I’m not sure I’m that interesting an instructor, and I teach college students.

It’s easy for outsiders to say that teachers should stand up to the vocal, unhappy minority. But it’s less easy to do that when a teacher relies on their job for rent and health insurance. It’s also less easy when the teacher worries about what administrators and principals will do and what could happen if the media gets involved or if the teacher gets demonized.

Despite the fact that no one actively wants school to be boring, the collection of forces operating on the school experience pushes it towards boredom. Many people, for example, are very interested in sex and drugs, but those topics also excite many students and parents, such that it’s difficult to say much that’s true about them in school.

As Caplan says, however, boredom is almost a feature, not a bug. Boring classes allow students to signal traits that employers value, like conscientiousness, intelligence, and conformity. Even if reading Ethan Frome is boring, being willing to tolerate Ethan Frome is important to people who would not themselves read Ethan Frome.

Caplan argues that most education is actually about signaling, not skill development. It’s notable how little we in as a society have improved education in the last two decades, when the Internet has opened up many new learning and signaling opportunities. Caplan has a theory about why: using weird counter-signaling efforts itself signals non-conformity and general weirdness (“‘alternative’ signals of conformity signal nonconformity”). So we’re stuck in a negative equilibrium.

He might be right. That said, I wonder if we’re just seeing a lag: twenty years is a long time by some standards, but in the history of education it’s a relatively short time. The problems with contemporary education also seem to argue that many employers would be well-served to ignore the signals sent by degree and search for alternate signals instead. Google claims to be doing this, but I don’t know of any researchers who’ve audited or studied Google’s internal data (if you do, please leave a pointer in the comments).

The people who most need to read this book are probably educators and high school students. The former probably won’t read it because it punctures some of the powerful myths and beliefs that keep them motivated. The latter probably won’t read it because high school students read very few books, and the ones most likely to read The Case Against Education are probably also likely to gain the most from higher education. So it’s another of these books that’s caught in a readerly catch-22.

Here is a Claudia Goldin paper, “The Race between Education and Technology: The Evolution of U.S. Educational Wage Differentials, 1890 to 2005;” as one person said on Twitter, “I agree with @bryan_caplan that the wage premium from education mainly comes from signaling, rather than learning vocational skills. But – I also believe widespread, generalist, higher ed can be a very good thing (as explained in [“The Race Between…”]).”

I also wonder about this: “employers throughout the economy defer to teachers’ opinions when they decide whom to interview, whom to hire, and how much to pay them.” Do they? Do most employers require transcripts and then actively use those transcripts? It seems that many do look for degrees but don’t look for grades.

One question, too, is why more people don’t go into various forms of consulting; smaller firms are less likely to be interested in credentials than larger ones. I do grant writing for nonprofits, public agencies, and some research-based businesses. Zero clients have asked about educational credentials (well, a few public agencies have superficial processes that ask about them, but the decision-makers don’t seem to care). Clients are much more interested in our experience and the skills demonstrated by our website and client list than they are in credentials. And when we’ve hired various people, like website programmers or graphic designers, we’ve never asked about education either, because we don’t care—we care if they can get the job done. In restaurants, I’ve never stopped a server or hostess to ask if the chef went to cooking school. So smaller firms may offer some respite from degree madness; if there is a market opportunity for avoiding expensive college and the credentials race (for individuals), it might be there.

Yet at the same time, I feel (perhaps wrongly) that school did help me become a better writer. “Feel” is a dangerous word—it’s hard to dispute feelings but easy to dispute data—yet I don’t know how else to describe it. When I read other people’s writing, especially other people’s proposals, I often think, “This helps explain why I have the job I do.” It’s possible to get through college and learn very little about writing. Occasionally managers will learn that I teach writing and say, “Why can’t college graduates write effectively?” An excellent question and one that requires 10,000 words of answer or no answer at all. But the alternative—not taking any writing classes—often seems worse.

Caplan also conducts many fascinating thought experiments, of sorts, although perhaps “contextualizes common practices and ideas” may be more accurate:

The human capital model doesn’t just imply all cheaters are wasting their time. It also implies all educators who try to prevent cheating are wasting their time. All exams might as well be take-home. No one needs to proctor tests or call time. No one needs to punish plagiarism—or Google random sentences to detect it. Learners get job skills and financial rewards. Fakers get poetic justice.

Signaling, in contrast, explains why cheating pays—and why schools are wise to combat it. In the signaling model, employers reward workers for the skills they think those workers possess. Cheating tricks employers into thinking you’re a better worker than you really are. The trick pays because unless everyone cheats all the time, students with better records are, on average, better workers.

Makes sense to me. I sometimes tell students that, if they manage to get through college without learning how to read and write effectively, no one comes back to ask me why. No college offers partial refunds to the unemployable who nonetheless graduate. The signal is the signal.

Many of you will not like The Case Against Education too because it is thorough. Caplan goes through his arguments, then many rebuttals, then rebuttals to the rebuttals. If you want a book that only goes one or two layers deep, this is the wrong book for you and you should stick to the Internet.

Many books also fail to convincingly answer the question, “What should we do about the problem identified?” Caplan doesn’t. He argues that public spending on education (or “education:” as much of what seems like education should be called signaling) should be eliminated altogether, while simultaneously acknowledging that this is only slightly more likely than someone jumping to the moon.

Caplan fulfills many of the conditions of myth, but probably not enough people will read this book to truly hate him. Which is a pity: as I said in the first line, the book is brilliant. But socially desirable persons will reject it, if they consider it at all. And the education machine will press on, a monstrous juice press squeezing every orange that enters its maw. Once I was the orange; now I am the press.

One other answer to “What education does?” may be “to keep options open” and “provide a base from which to build later.” Without some writing and numeracy skills, it’ll be hard to enter many careers; while school may do a lousy job of building them (as Caplan demonstrates), if the alternative to school nothing (i.e. Netflix, hanging out, and partying), school may be a better option than nothing.

As for optionality, I think of my friends, many artistically inclined, who got to their mid or late 20s and around that time got tired of working marginal jobs, struggling to pay rent, working in coffee shops, crashing on friends’ couches, etc. Things that seem glamorous at age 20 often seem depressing five or ten years later. Many of them have gone back to school of various kinds to get programming or healthcare jobs. In the former case, math is important, and in the latter case, biology and some other science knowledge is important. Those who blew off math or bio in high school or college struggle more in those occupations. So maybe education is about keeping at least some options open—or more options than would be open for someone who quits school or begins vocational ed in 8th grade.

Finally, education might be an elite phenomenon. We educate everyone, or, more realistically, attempt to educate everyone, in order to get a relatively small number of elite people into position to drive the entire culture forward. The people at the pinnacle of the scientific, technical, artistic, and social elites got there in part because they had access to education that was good enough to get them into the elite spheres where it’s possible to make a real difference.

I’m not sure I’m in those elite spheres, but I may be close, and at age 15 I probably didn’t look like such a good bet. Yet education continued and here I am, engaging in the kinds of conversations that could move the culture forward. If I’d been tracked differently at age 15 that might not’ve happened. Yes, the process is horrendously wasteful, but it’s useful to give many people a shot, even if most people go nowhere.

To be sure, I buy Caplan’s argument, but I’ve not seen this angle pursued by others, and it at least seems plausible. I also don’t know how one would measure the “education as elite phenomenon” argument, which is another weakness of my own point.

Still, I’ve become more of an elitist because of my involvement in the educational system, which shows that most students are in fact bored and don’t give a damn. When I started grad school I thought I could help students become more engaged by changing the nature of the short journal assignments: instead of just writing for me, students would start blogs that they would read and comment on. Education would become more peer-driven and collaborative. The material would seem relevant. Right?

After a semester or two of reactions that ranged from indifference at best to massive hostility at worst, I stopped and went back to the usual form of short written responses, printed, and handed in. That was easier on me and on the students, and it still at least exposed students to the idea of writing regularly. A few may have continued the practice. Most probably didn’t (and don’t). I learned a lot, maybe more than students, and I also learned that I’m a weirdo for my (extreme) interests in writing and language—but my own time in the education system and my own friend set had to some extent hidden that from me. Now, however, it’s so apparent that I wonder what 24-year-old me was thinking.

Caplan helps explain what I was thinking; many people who go into various kinds of teaching are probably optimists who themselves like school. They’re selected for being, in many cases, passionate weirdos. Personally, I like passionate weirdos and misfits and the people who don’t fit well into the school system (I’ve been all three). But I seem to be unusual in that respect too, though I wasn’t so weird that I couldn’t fit into the convention-making machine. A good thing, too—as Caplan notes, it’s individually rational to pursue educational credentials, even if the mass pursuit of those credentials may not be so good for society as a whole. Correlation is not causation, as you no doubt learned from your statistics classes and still understand today.

Here is a good critical review, not wholly convincing in my view, but worth thinking about.

Ninety-five percent of people are fine — but it’s that last five percent

April 24, 2017 By in Academia, Culture, Teaching Tags: , 6 Comments

“How Airline Workers Learn to Deal with Passengers” reminds me of something I’ve noticed about teaching and consulting: 95% of people are fine, but that last 5% can occupy disproportionate time and mental energy.* There’s a temptation to become somewhat armored against that last 5%, which negatively impacts interactions with the vast majority of normal, reasonable people.

A lot of public-facing professions seem to have this problem, including emergency medicine doctors, cops, retail workers, and public school teachers. Because the bottom 5% can be noisy and time-consuming, a kind of misanthropy can set in, as one begins to think the few represent the whole—even if, intellectually, one knows it does not. Mental, psychological, and emotional armoring can reduce one’s overall effectiveness; this is particularly obvious in teaching, in which person-to-person connection plays a stronger role than it does in consulting.

Something about the human mind seems to make one negative interaction stand out more than 10 positive or normal interactions. I didn’t realize this at first, so a small number of negative experiences affected me much more than they should have. There’s a kind of crowding-out effect going on. Now, when I have to deal with someone who is unreasonable or at least not representative of the whole, I try to actively, consciously remind myself that they don’t represent the whole. Behind every irrationally unhappy person there are probably 49 to 99 normal people who aren’t giving me unwarranted grief.

Colleges in particular have been in the news lately, and a lot of people have read stories about crazy social justice warriors or censorious students—like the Middlebury College thing or the Halloween costumes at Yale brouhaha. Sure, these stories are in fact outrageous, but, again, they’re also salient because they’re unusual. Because they’re unusual, they make the news (and these kinds of events do represent a problem, though the problem tends to be overstated).

Friends and acquaintances sometimes ask me about the activists students suppressing speech. Yeah, I do, a little, but not very much. The vast, overwhelming majority of college students, seem to want what college students have always wanted: to learn something; to get by; to get a job when they’re done; to get laid; to learn something about themselves and the societies they live in; to make friends; to individuate from their families. You could add other items. Many students feel a vague sense of worry about being excellent sheep, and that worry is itself a sign of intellectual health. Most students, if they’ve thought about free-speech issues at all, vaguely support it. But a minority of well-organized and angry activists can make a lot more noise and news than the silent majority! As Nassim Taleb says, “The Most Intolerant Wins: The Dictatorship of the Small Minority.” A motivated minority, especially given the college-complaint apparatus, can create a lot of bureaucratic hassle.

Professors and other students often feel a kind of chilling effect, and one related to my essay, “How do you know when you’re being insensitive? How do you know when you’re funny?” Similar issues play out in many fields beyond teaching and consulting. One angry, unreasonable, or irrational customer or client drowns out a lot of generically happy or satisfied ones. Or consider “I was a landlord: This is what it taught me about people.” Landlords have to be prepared for worst-case scenarios, and that preparation bleeds into their everyday scenarios and interactions.

Teaching, especially at the K – 12 level, suffers from defensive posture problem. A teacher who tries to be honest and interesting risks the ire of his or her angriest, unhinged, or most ideological students (and, even worse, their parents). Almost no teacher gets in trouble for being boring, but a teacher can get in trouble or can get in trouble for being many values of “interesting.” Even I’ve had that problem, and I’m not sure I’m that interesting an instructor, and I teach college students. Students who complain about school being boring get told that school is supposed to be boring. Students who complain about school being interesting (or “offensive,” or whatever) get much more attention.

It’s easy for outsiders to say teachers should stand up to the vocal, unhappy minority. But it’s less easy to do that when a teacher relies on their job for rent and health insurance. It’s also less easy when the teacher worries about what administrators and principals will do and what could happen if the media gets involved or if the teacher gets demonized. It would be helpful for more administrations to make public statements like the University of Chicago’s, confirming a commitment to free speech and open inquiry.

Social media probably amplifies many of the problem traits described above by allowing the least-reasonable people to organize, scream, and (not infrequently) lie. I don’t know what, if any, solution exists to these problems, apart from most individuals to attempt to be as reasonable as possible and not succumb to the noisy but unhinged minority. Not much of a rallying cry, is it? No one, or almost no one, rallies for free speech and free inquiry.

* You can change the ratios some; I doubt the number of problem students reaches 10% in most scenarios, and I also doubt that the number declines below 2% (among professions that face the general public).

Why hasn’t someone tried to build or fund a very low-cost, very high-quality college?

April 3, 2017 By in Academia, Teaching Tags: , , , , , 5 Comments

As the title asks, why hasn’t someone tried to build or fund a very low-cost, very high-quality college? Or, if they have, what school is out there and has tried this?

It seems like a ripe strategy because virtually every (even slightly) selective school is pursing the same prestige strategy—high sticker prices, nominal discounts via “scholarships,” tenure-based faculty selection system, extensive administrative bloat, and so on. Yet even as schools relentlessly copy each other, news about outrageous student loan burdens is everywhere and probably affecting the choices made by students, and student openness to alternatives. At the same time, college tuition has been outpacing inflation for decades, and everyone knows it. Education is a component of the “cost disease” that is afflicting other sectors too. The number of college administrators has grown enormously (though that may not be the prime factor behind public-school cost increases). Students used to be possible to work summer jobs and graduate with little or no debt; schools in the 1960s or 1970s don’t appear to have been dramatically worse at education than schools today, and in some ways they may have been better, yet today colleges are many times more expensive. Schools trumpet their commitment to nominal education quality, the same way airlines trumpeted their commitment to passenger comfort, before deregulation forced airlines to compete on price and other metrics too, and anyone who’s been to a modern college knows that real commitment is “quality” is more rhetorical than real.

College costs and debts have soared, and at the same time the number of PhDs granted far outstrips the number of tenure-track or teaching jobs, so the workforce is available. Most universities and even many colleges care far more about research, much of which is bogus anyway, than teaching. Many universities don’t care about teaching at all, as long as the professor shows up to lecture, isn’t drunk, and doesn’t trade sex for grades. I hear many, many grad students and early professors lament the way their schools don’t care about teaching. There’s a surplus of cheap PhDs out there who’d desperately like to be professors. While professors who only teach two or three classes per semester complain relentlessly about all the “work” they supposedly have and how “busy” they allegedly are, it could be very easy to get professors to teach far more than they currently do at most schools, further reducing costs.

In short, the supply of faculty is there, and the supply of students ought to be there. With the setup above, let me repeat: why hasn’t anyone attempted to start a teaching-focused college with low tuition and extremely high-quality academics? I’m thinking of a school with a mandate to minimize the number of administrators, sports teams, and other boondoggles. One could even eliminate tenure, and thus ensure that PhDs hired today won’t still be on the payroll in 40 years. The school could highlight “proof of knowledge” over seat time as a metric; to my knowledge, there’s nothing intrinsic about four or five years of seat time. Students who study hard could and should spend less time in the seat.

Some of the situation I’m describing sounds like a community college, but I’m imagining a school that still draws from a national applicant pool and still maintains or attempts to maintain an elite or comprehensive academic character. Think of a liberal arts school but scaled up and with fewer administrators. If I were a billionaire I might try to do this; stupendously rich people loved endowing schools in the 19th Century, but that seems to have fallen out of fashion. Still, it worked then, so perhaps it could work now.

It may be that schools are really selling prestige and status, and consequently a low-cost, high-quality teaching school would be too low prestige and low status to attract students.

Still, and again as noted previously, pretty much every school, public or private, is pursuing the exact same prestige, admissions, and marketing strategy. With one or two exceptions (CalTech, University of Chicago—okay, there are a few others, but not many), they don’t even try (really or seriously) to distinguish themselves, and almost every school competes for the same BS college rankings. Such a uniform market seems ripe for alternate approaches, yet none are being tried or have taken off, though there are some small scale efforts, like Minerva.

What am I missing?

* Maybe it was easier to start colleges in the 19th Century, when regulation was nonexistent and complex subsidies of various kinds weren’t available. In the 19th Century, many colleges were also founded with the explicit intent of saving students’ souls, so perhaps the lack of religiosity in today’s billionaires and/or most of today’s students is a factor. God is an underappreciated component of many older endeavors.

* Current schools might just be too damn good at marketing for others to break in. Plus, college is a huge investment, which engenders a certain amount of conservatism in choices. Given costs, though, I think there’s room for experimentation here.

* Maybe there are efforts afoot and they’ve either failed or are too small for me to have noticed.

* Current schools are pursuing a complex price discrimination strategy, in which the sticker price is paid by a relatively small number of students, and much of the study body receives “scholarships” that are really tuition discounts. Maybe this system is more appealing to students and possibly schools than a transparent, everyone-pays-$5,000-per-year strategy.

* Students by and large pay with their parents’ money or pay with loans, so many an unbundled version of a school really is less attractive than one with lots of administrators, feel-good projects, fancy gyms, etc. Despite schools’ rhetoric to the contrary, they’re obsessed with attracting and retaining rich kids, so that market may be where all the juice is.

* Billionaires who might fund this are busy doing other things.

* The number of “good” or at least weird and different students who would try such a school is not great enough (given the current cost of college and the number of students out there, I find this one hard to believe, but it isn’t impossible).

I’m guessing number four is most likely, but maybe there are other features I’m missing.

Circumstances under which going to law school can make sense

December 22, 2016 By in Academia, Teaching Tags: , , 1 Comment

The reasons you should avoid law school are well known and I won’t repeat them here, but the other day I was explaining to a former student why she shouldn’t go to law school and she asked a perceptive question: Who should go? Under what conditions should a person go?

The answer is “almost no one” and “almost never,” but law school can be okay in a handful of circumstances:

* People who have already worked in law firms, probably as a paralegal but maybe under other circumstances, and who thus understand what the day-to-day life of a lawyer is like. That firm should have a job waiting and ready to go for the person before the person starts law school.

* People who have family (or close family friends) in law firms who can set the law school applicant up with a job straight out of school. If your uncle has a firm and wants you to take over that firm, law school can make sense.

* People with a very specific sense of what they want a law degree for and what they want to do with it—for example, people who desperately want to fight for voting rights, or immigrant rights, or something along those lines, and are convinced that those fights will be their life’s work, regardless of other challenges.

That’s really it; if I’m missing something, leave a comment or send an email. Law school mostly works for people who don’t need law school and only need the credentials that law school entails. There is a reason why most lawyers learned the craft on the job as apprentices, and law school only became a requirement in the post-World War II-era as a way of raising the salaries and status of then-existing lawyers.

Even going to highly ranked schools doesn’t make sense because, while you may get a big-firm job straight out of school, you’ll still be shackled to the work by student loan debt slavery, and you’ll still have to be a lawyer at the end (which most people don’t really want to do), and you’ll still probably not make partner (which means that you’re mostly working to line someone else’s pocket).

Don’t go to law school.

Geek Heresy: Rescuing Social Change From the Cult of Technology — Kentaro Toyama

November 17, 2015 By in Academia, Books, Nonfiction, Teaching Tags: , , , Leave a comment

My review on Grant Writing Confidential is actually germane to readers of The Story’s Story, too, so I’ll start by directing you there. The book’s central and brilliant point is simple: for at least a century various people have imagined that better technology and the spread of technology will solve all sorts of social ills and improve all sorts of institutions, with education being perhaps the most obvious.

Geek_heresy2There are many other fascinating points—too many to describe all of them here. To take one, it’s often hard to balance short- and long-term wants. Many people want to write a novel but don’t want to write right now. Over time, that means the novel never gets written, because novels get written one sentence and one day at a time. Technology does not resolve this challenge. If anything, Internet access may make it worse. Many of us have faced an important long-term project only to diddle around on websites:

Short-term pleasure often leads to long-term dissatisfaction. That intuition underlies the psychologist’s distinction between hedonia and eudaimonia. Pleasure-seeking hedonism is questionable, but maybe long-term eudaimonic life satisfaction is good.

One sees these issues all over. Porn remains ridiculously popular (though some consumers of it are no doubt fine). Many people drink soda despite how incredibly detrimental soda is to health, and in my view how bad soda tastes compared to, say, ice cream. TV watching time is still insanely high, though it may be slightly down from its previous highs. There are various ways one can try to remove agency from the people watching porn while drinking soda and keeping one eye on a TV in the background, but the simpler solution is to look at people’s actions and see revealed preferences at work.

Most people don’t have the souls of artists and innovators trapped in average everyday lives. Most people want their sodas and breads and sugars and TV and SUVs and all the other things that elite media critics decry (often reasonable, in my view). Most people don’t connect what they’re doing right now to their long-term outcomes. Most people don’t want to be fat but the soda is right here. A lot of people want a better love life but in the meantime let’s check out Pornhub. Most people want amazing Silicon Valley tech jobs, but Netflix is here right now and Coursera seems far away.

And, to repeat myself, technology doesn’t fix any of that. As Toyama says of one project that gives computer access to children, “technology amplifies the children’s propensities. To be sure, children have a natural desire to learn and play and grow. But they also have a natural desire to distract themselves in less productive ways. Digital technology amplifies both of these appetites.” I had access to computers as a teenager. I wasted more time than I want to contemplate playing games on them, rather than building the precursors to Facebook. Large markets and social issues emerge from individual choices, and a lot of elite media types want to blame environment instead of individual. But each individual chooses computer games—or something real.

It turns out that “Low-cost technology is just not an effective way to fight inequality, because the digital divide is much more a symptom than a cause of other divides. Under the Law of Amplification, technology – even when it’s equally distributed – isn’t a bridge, but a jack. It widens existing disparities.” But those disparities emerge from individual behaviors. People who want to be writers need to write, now. People who want better partners or sex lives need to quit the sugar, now. One could pair any number of behaviors and outcomes in this style, and one could note that most people don’t do those things. The why seems obvious to me but maybe not to others. The people who become elite developers often say coding is fun for them in a way it apparently isn’t to others (including me). Writing is fun to me in a way it apparently isn’t to others. So I do a lot of it, less because it’s good for me than because it’s fun, for whatever temperamental reason. Root causes interest me, as they do many people with academic temperaments. Root causes don’t interest most people.

Let me speak to my own life. I’ve said variations on this before, but when I was an undergrad I remember how astounded some of my professors were when they’d recommend a book and I’d read it and then show up in office hours. I didn’t understand why they were astounded until I started teaching, and then I realized what most students are like and how different the elite thinkers and doers are from the average. And this is at pretty decent colleges and universities! I’m not even dealing with the people who never started.

Most of the techno-optimists, though—I used to be one—don’t realize the history of the promise of technology to solve problems:

As a computer scientist, my education included a lot of math and technology but little of the history or philosophy of my own field. This is a great flaw of most science and engineering curricula. We’re obsessed with what works today, and what might be tomorrow, but we learn little about what came before.

Yet technology doesn’t provide motivation. It’s easy to forget this. Still, I wonder if giving 100 computers to 100 kids might be useful because one of them will turn out to be very important. The idea that a small number of people drive almost all human progress is underrated. In The Enlightened Economy Joel Mokyr observes that the Industrial Revolution may actually have been driven primarily by ten to thirty thousand people. That’s a small number and a small enough number that the addition to or subtraction of a single individual from the network may have serious consequences.

This isn’t an idea that I necessarily buy but it is one I find intriguing and possibly applicable to a large number of domains. Toyama’s work may reinforce it.