Category: education

How to Write: Lessons From My Writing Class

Seth Roberts8 Comments

I just finished teaching an undergraduate class called Academic Writing at Tsinghua. One semester, pass/fail, about 10 students. The last assignment was list six things you’ve learned. Combining the answers, I came up with this:

1. Don’t tell readers what they already know. This came up a lot when I discussed how to write a personal statement. “Your university has an excellent program in X” — no, don’t say that.

2. To make your writing moving, focus on your own thoughts and emotions. Moving = evoking emotion. Evoking emotion was enormously important, I said.

3. Use simple words and sentences (don’t show off). As one student put it, “Received the blames from one class, changed all my GRE words into understandable words.”

4. Give examples.

5. Avoid boasting (say “I like X”, don’t say “I am good at X”).

6. Do not write about things that are “too big”.

7. Have clear connections between sentences. We spent several classes on the various ways adjacent sentences can be related.

8. Say things that are honest and true. In contrast to what you think your reader wants to hear. A friend asked for advice on her personal statement for a graduate school application. She sent me a revised version. I thought the unrevised more honest version was better.

9. Begin with something interesting.

I asked which of these lessons they already knew. The consensus answer was #1 (don’t tell readers what they already know) and #4 (give examples). Their personal statements flagrantly violated #1. One student said they had learned it, yes, but needed to be reminded.

Jon Cousins of Moodscope, in town for a Quantified Self conference, gave a guest lecture. From his talk the students came away with four main things:

1. Copy someone’s writing you admire.

2. Imagine your audience. Are they busy? Curious?

3. Write as you speak.

4. Revise after a period of time. Like a month.

Another of Jon’s lessons was use punctuation sparingly. An editor told him, “Using an exclamation mark is like laughing at your own joke.”

 

Assorted Links

Seth Roberts9 Comments

Thanks to Dave Lull.

Assorted Links

Seth Roberts2 Comments

Thanks to Casey Manion.

Assorted Links

Seth Roberts10 Comments
  • “Light” Ph.D. — a less expensive research degree
  • Umami Burger expands
  • A diuretic reduces autism symptoms. Does water balance influence brain function in people without autism?
  • This Amazon reviewer is almost always disappointed and his one-star reviews are fun to read. I suggest that ratings (book ratings, product ratings, etc.) compare the rating to other ratings given by the rater. A 5-star rating is more impressive if a rater’s average rating is 2 than if it is 5. I suggest percentiles. For example, rating = 5 (90%ile) is more impressive than rating = 5 (50%ile). I’d also like to know the average percentile across raters.
  • Lack of variation in heart rate predicts infection in neonates. The writer (Mike Loukides) is too surprised (“astonishing connection”). Many studies have found associations between too-little variation in heart rate and serious health problems.

Thanks to Adam Clemens and Patrick Vlaskovits.

Online Teaching Versus What?

Seth Roberts12 Comments

Is online teaching (e.g., MOOC) a big deal? In an essay (“Why Online Education Works”), Alex Tabarrok argues for the value of online education (meaning online lectures) compared to traditional lectures. A friend told me yesterday that MOOC was “a frontier of pedagogy”. No doubt online lectures will make lecture classes cheaper and more available. Lots of things have gone from scarce/expensive to common/cheap. With things whose effects we understand (e.g., combs), the result is straightforward: more people benefit. With things whose effects we don’t understand, the results are less predictable. Did the spread of sugar help us? Hard to say. Did the spread of antibiotics help us? Hard to say. It may have helped sustain simplistic ideas about what causes disease (e.g., “acne is caused by bacteria”, “ulcers are caused by bacteria”) reducing effective innovation. Do we have a good idea of the effects of lectures (or their lack of effect), or a good theory of college education? I don’t think so. Could their spread help sustain simplistic ideas about education? Maybe.

As books spread, the teaching of reading increased. Everyone understood that books were useless if people couldn’t read. The introduction of PCs was accompanied by user interface improvements. This helped PCs become influential– not restricted to hobbyists. Will online education be accompanied by similar make-it-more-palatable changes? I have heard nothing about this. Their advocates seem to think the current system is fine and if it could only be available to more people…

Online lectures will make much difference only if the cost and quality of lectures is the weakest link in what strikes me as a process with many links. It would be a coincidence if the link that can be most easily strengthened turned out to be the weakest link. For example, is the cost of lectures the main thing driving up the cost of college? That would be wonderful if it were true, but I haven’t seen evidence that it’s true. At Berkeley, for example, there has been enormous growth in the administrator-to-faculty ratio.

Here are two arguments used to argue that online lectures are a big step forward:

It will help people in poor countries, like Zambia. There is a long history of people in rich countries misunderstanding people in poor countries. Several years ago I was in Guatemala. I heard about a school being built by a (rich country) religious group in a poor area. After two years, the American running it wanted to leave. No member of the community took it over. It disappeared. “Maybe they didn’t want a school,” said the graduate student who told me about it. Maybe few people in Zambia want online lecture classes. (I have no idea.) If so, the benefit will be small.

It will save labor. Each lecture will be viewed many more times. Saving labor is not always good. It is plausible that the growth of online lectures will mean fewer college professors. Colleges and universities are among the few places where people do research and almost the only places where they do unrestricted research. Most of the research is useless; a tiny fraction is enormously useful. At the moment, lectures subsidize research. By giving lectures, professors are allowed to do research. Fewer professors, less unrestricted research, less innovation. “Wasteful” lecturing might be labor we shouldn’t save.

One thing I like about online classes is the possibility they will connect people who want to learn the same thing, like ordinary classes do. They can help each other, encourage each other, and so on. I have no doubts about the value of this. (I find language partners — I teach them English, they teach me Chinese — way more pleasant and helpful than tutors.)

At Berkeley, I tried to find good lecturers. With two exceptions (Tim White and Steve Glickman) I failed. Almost all lectures, even those by brilliant researchers, were dreary. (A shining exception by Robin Hanson.) They suffered from a lack of stories and a lack of emotion. (At Tsinghua, things are worse. A friend who majors in bioengineering told me that 80% of her teachers lecture by reading from the textbook.) The power of professors over students in some ways resembles the power of doctors over patients. Just as there is little pressure on doctors to understand disease (if antibiotics have bad effects, it doesn’t harm the doctor who prescribed them), there is little pressure on most professors — at least at the elite research universities that produce online lectures — to understand education. At Berkeley, many professors say they teach their undergraduate students “how to think” or “how to think critically”. In fact, they were teaching their students to imitate them. The simplest form of education. This is neither good nor bad — it depends on the student — but it is the opposite of sophisticated.

A few months ago I assigned my Tsinghua students (freshmen) to read 60 pages of The Man Who Would Be Queen by Michael Bailey, a book full of stories and emotion. Any 60 pages, their choice. No test, no written assignment, no grade. One student told me it was the first book in English she’d ever finished. It was so good she couldn’t stop reading. My assignment had changed real-life behavior: what my student read in her spare time. Maybe it changed her tolerance of homosexuality and the tolerance of those around her. My assignment (not a textbook or academic paper, not a fixed reading) and evaluation (none) differed from conventional college teaching. Experiences like this make me wonder what fraction of important learning during college happens due to lecture classes. (In my case, the fraction was zero.) If the fraction is low, it suggests that online learning won’t make much difference.

Who is the Richest Person in China?

Seth Roberts6 Comments

If you open the American edition of Forbes, you will find articles about the richest people in America. If you open the Russian edition, you will find articles about the richest people in Russia. If you open the Chinese edition, you will find articles about the richest people in America.

A Russian friend of mine noticed this. He happened to know an sophomore economics major at Tsinghua. It is incredibly difficult to get into Tsinghua and the economics major is the most desirable major of all. To be an economics major at Tsinghua you need a test score that is in something like the top 1 out of 100,000. Staggeringly high. My Russian friend asked the Tsinghua economics major, “Who is the richest person in China?”

The economics major didn’t know. He seemed a little angry. “Why should I know? We’ve never been taught that,” he said.

 

The Emphasis on Education in China

Seth Roberts5 Comments

One of my students grew up and went to high school in Nanjing, population 8 million. Her acceptance to Tsinghua was such a big deal that when her acceptance letter reached the local post office they called to tell her. The post office also alerted journalists. When the letter was delivered to her house, there were about 20 journalists on hand. One of them, from a TV station, asked her to say something to those who failed.

The Physical Spacing Effect: New Way to Learn Chinese Works Shockingly Well

Seth Roberts17 Comments

Two years ago I taped a bunch of Chinese flash cards (Chinese character on one side, English meaning on the other) to my living room wall (shown above). I’ll study them in off moments, I thought. I didn’t. It was embarrassing when guests pointed to a card and said, “What’s that?” But not embarrassing enough.

A few weeks ago, I can’t remember why, I decided to test myself: how many do I know? About 20%. I’ll try to learn more, I thought. I was astonished how fast I learned the rest. It took little time and almost no effort. I didn’t need “study sessions”. I glanced at the array now and then, looked for cards I didn’t know yet, and flipped them to find out the answer. After a few days I knew all of them.

I had been using conventional methods (flash cards studied in ordinary ways, Anki, Skritter) and an unconventional method (treadmill study) to learn this material for years. In spite of spending more than a hundred hours on each method, I had never gotten very far. I might get to 500 characters and backslide due to lack of study. Treadmill walking while studying made studying much more pleasant, but I found I would still prefer to watch TV rather than study Chinese. Maybe part of the problem was too many days skipped. After you skip four days, for example, you have a discouragingly large number of cards to review. Plainly these methods work for others. They didn’t work for me.

After my success with the two-year-old cards, I put up another array (8 x 13). I already knew about 40% of them. I learned the rest in a day or so. Then I put up a 10 x 12 array. I tested myself on them one morning. I knew 30 of them. I studied them during the day for maybe 30 minutes in little pieces throughout the day. The next morning I tested myself again (about 12 hours from the last time I had studied them). Now I knew 105 — I had learned 75 in one day, almost effortlessly. That day I studied for a few minutes. The next morning I tested myself again. Now I knew all but one of them.

I did not notice any facilitation of learning when I studied flash cards while walking around. In that case, unlike this one, (a) they were in roughly the same position relative to my body and (b) had no consistent physical location. I noticed the same facilitation of learning during a Chinese lesson in a cafe. I was having trouble remembering three Chinese words (e.g., the Chinese word that means graduate). I wrote each of them on a piece of paper with the English on one side and the Chinese on the other. I put the three pieces of paper at widely-separated places on the table. I studied them briefly, a few seconds each. That was enough. Five days later I still remember them (having used them a few times since then). This happened in a place (a cafe) with which I wasn’t familiar, unlike my living room. Maybe the general principle will be it is much easier to learn an association if it is in a new place.

It’s very early in my use of this method, but I doubt it’s a fluke. It connects with several things we (= psychology professors) already know. 1. The mnemonic device called the method of loci. You put things you want to learn in different places in a well-remembered landscape (e.g. different places in a building you know well). Usually the method is used to learn lists, such as the digits of pi or the order of cards in a deck. You place different items in the list in different places in the imagined place. Then you “walk” (in your imagination) through the imagined place. The method dates back to ancient Rome. 2. The power of interference. Thousands of experiments have shown that learning X makes it harder to learn similar Y. X and Y might be two lists, for example. The greater the similarity, the bigger the effect. What you learn on Monday makes it harder to learn stuff on Tuesday (proactive interference); what you learn on Tuesday makes it harder to remember what you learned on Monday (retroactive interference). To anyone familiar with these experiments, my discovery has a simple “explanation”: spatial interference. 3. Evolutionary plausibility. The study of printed materials (e.g., books) is so recent it is hard to imagine our brain has evolved to make it easy. In contrast, thousands and millions of years ago we had to learn about things in different places. Learning about food and danger in different places was especially important. When language arrived, the necessary learning (at first, attaching names to objects) is quite similar to my learning because the named objects were in different places. The study of vitamins and to some extent my work (especially the power of morning faces) show how hard it can be to figure out what we need for our brains and bodies to work well. How non-intuitive the answers may be.

These results suggest a new mnemonic device: Stand in front of an empty wall and imagine on the wall the associations you want to learn, each association in a different place like flashcards. This is a fast way of putting each association in a different place.

 

 

 

Assorted Links

Seth Roberts13 Comments

 

Thanks to Chuck Currie, Grace Liu, Alex Chernavsky and Dave Lull.

Assorted Links

Seth Roberts2 Comments

Thanks to Bryan Castañeda.