Difference between revisions of "Notes for My Book-in-Progress on Writing, Talking, Listening and Thinking"

From Rasmapedia
Jump to navigation Jump to search
(Other)
(CODING)
Line 133: Line 133:
  
 
==CODING==
 
==CODING==
[https://statmodeling.stat.columbia.edu/2020/08/10/know-your-data-recode-missing-data-codes/#comment-1407902  Know your data, recode missing data codes ] from  Andrew Gelman's blog, 10 August 2020:
+
*[https://statmodeling.stat.columbia.edu/2020/08/10/know-your-data-recode-missing-data-codes/#comment-1407902  Know your data, recode missing data codes ] from  Andrew Gelman's blog, 10 August 2020:
 
{{Quotation|  <html><img src="https://statmodeling.stat.columbia.edu/wp-content/uploads/2020/02/Screen-Shot-2020-02-19-at-10.17.50-AM.png"  width = 500px /></html>
 
{{Quotation|  <html><img src="https://statmodeling.stat.columbia.edu/wp-content/uploads/2020/02/Screen-Shot-2020-02-19-at-10.17.50-AM.png"  width = 500px /></html>
 
We had a class assignment where students had to graph some data of interest. A pair of students made the above graph, as a reminder that some data cleaning is often necessary. The students came up with the excellent title as well!
 
We had a class assignment where students had to graph some data of interest. A pair of students made the above graph, as a reminder that some data cleaning is often necessary. The students came up with the excellent title as well!
Line 155: Line 155:
 
The reason someone might use something like “-999” is so that when you import the dataset your stats software is much more likely to properly categorize each variable (e.g., string, integer, factor, etc.). Integers for missing values are far less likely to mess up this process whereas strings will interfere with it almost every time. Large negative integers are the standard for many professionally created datasets (e.g., General Social Survey).
 
The reason someone might use something like “-999” is so that when you import the dataset your stats software is much more likely to properly categorize each variable (e.g., string, integer, factor, etc.). Integers for missing values are far less likely to mess up this process whereas strings will interfere with it almost every time. Large negative integers are the standard for many professionally created datasets (e.g., General Social Survey).
 
}}
 
}}
 +
 
----
 
----
 
  
 
==Other==
 
==Other==

Revision as of 22:16, 24 November 2021

I'm coming back to the idea of revising my Aphorisms on Writing, Speaking, and Listening. In particular, though good writing never changes, technology does. I am turning the book chapter into a short book to be finished in 2021. In July 2020 we started what my son Ben calls The Club on my deck, young men who sit on the deck with me as I pontificate on such things as appellate argument, economics workshops, thinking about your objective in talking, the importance of having video shown on DC Circuit TV, Temin's 3 Goals of Research Presentation, marginal vs. average product of time spent on particular points, the usefulness of declining cross examination if the prosecutor has failed to prove his point with a witness, burden of proof, and how to gut a catfish (Benjamin had a good day at Lake Monroe when we first met). We got through only one page of the Tips for Online Seminars, so the Club may well be The Marching Chinese, but it worked well for all concerned. To be sure, there is still uncertainty over whether beer complements rhetoric better than port, but we hope further research will resolve that issue.

I will use this page to add notes of things I come across.  

See also:
How to Run Online Talks and Writing and Talking and Listening and Coding and Tables of Numbers and Figures and Diagrams and Social media and C. P. Snow, Good Judgement and Winston Churchill and Indefinite Pronouns and Writing Right Right Now. and Writing Style. and Rewriting Abstracts and Diagrams. and Careful Writing Requires Work.

Miscellaneous

  • The Communists probably wanted to use her to displace people like Frederick Douglass who were too bourgeois for their purposes.
  • Proper use of "like"?
  • "One of the things you said that really stuck out to me is to realize rejections (to journals, conferences, etc.) are another way in which I am getting my name and my work out there..." The story of Barry Himmelstein and rejections at poetry journals.

Graphics

  • From a patient's viewpoint: I am mystified as to why doctors would never give quantitative answers as to the probably our babies would be born on time, early, or late. This is an extremely common question, of great importance even though not life and death. The doctors themselves have the data at hand in the local hospital.
Suppose they did collect their hospital's data for the past five years of baby arrival relative to predicted due date. How would this best be communicated to patients?
:    Would a histogram of bars indicating frequency be a good way? Would bars made up of 100 stick figures be better? Visualizations are far far better than numbers for pretty much everybody, from the illiterate to statistics PhD's (remember the Chernoff faces anyone?).
Or would it be better to say, "We had 300 births over the past few years in this hospital. 100 were early, 50 were late, and 150 were within one day of the due date"?

Editing

Don't think you impress people who ask you to look at their writing by saying it's perfect. I can't help but look down on people who say that to me, because I can always find ways to improve, usually even obvious ways. I do try, though; I know most people can't write.


Writing

  • Standard maxim of writing: "Never use the passive voice--- except when you're trying to conceal who it was that did something bad."
  • How not to cite sources. Here's an example of how somebody in Twitter gave a source--good, unusual for twitter-- but put the link as an image file, so nobody can click it or even cut and paste, so the source is useless. You have to put yourself in your readers' place. Think!
  • For young scholars in econ and political science, and probably every field, one problem is they haven't read enough papers. As you get older, you get impatient with reading millions of papers,a nd you want things simple-- or at least, for the paper to quickly tell you what's important. And you get tired of technique. You want a quick answer, because you can't read all the papers you see, and you need to know if it's worth reading carefully. This is particular important if the author is not at a top-ten department or is too young to have a recognizable name.

Even older scholars usually don't realize that it's often the diagram (or, for a theory paper, the numerical example) that sells a paper and makes it famous. The author has to to do the fancy stuff, too, but often the fancy stuff is too hard to read, so people skip it. Even if they read it, what they remember is the diagram or the numerical example. And if you want to persuade the reader that your result is correct, a diagram can be crucial, even if the reader, self-deceived, thinks its the fancy regression that has persuaded him.

{{Quotation| Yisong Yue @yisongyue When I was a PhD student, a professor told me:

"There are only three papers worth writing on a given topic: the first paper, the best paper, and the last paper."

It's obv a bit tongue-in-cheek, but I've often used the underlying heuristic to aid in choosing research directions.}


  • Andy Grewal (@AndyGrewal) tweeted at 11:08 PM on Sun, Jan 17, 2021: I'm plugging through many of them as I review admissions files. So many profs just give you a bunch of adjectives ("so-and-so is super, great, outstanding, etc. etc. etc."). I'd take one *illustration* over twenty adjectives any day.

(https://twitter.com/AndyGrewal/status/1351018599037759491?s=03)


  • In the intro, tell a few compelling anecdotes about bias, with inset paragraph quotes if you can. Then you can say, "This is thus an important question to investigate. But it shows the importance of looking at aggregate data, particularly when there are 80 million cases from which to cherry-pick 3 anecdotes. We will show that despite these compelling examples of bias, there is no sign of systematic bias in India."
  • In Table 2, the standard errors are all tiny and all the coefficients massively significant. Thus, you shouldn't report the standard errors, which add clutter. Just say int eh table note, "all t-statistics are of size 10 or more,s o we do not report the standard errors."
  • LSTM stuff: it woudl be good to have one example of how the machine learning works. E.g., you feed in Shivamurti, the machine notices that Shivamorti's are a lot like Shivamurti's after 10,000 iterations and starts classifying them the same. In both the paper and the presentation. Note 11 is good to have, but an example would convey the process better,a nd if you include both,e verybody will skip the footnote 11-style explanation adn just read your example. Maybe it should be in an appendix, a page or two long.
  • Instead of "Accused as a percentage of the population" use "Accused per 10,000 of the population", replacing 10,000 with whatever yields you numbers like 123 instead of .123 or 1.23.
  • Table A6, etc. My comment about accused per 10,000 is just for summary statistics. For your regressions, pick units so the regression coefficients come out looking nice. Instead of 0.008 you want the coefficient to be 8.47, which you can do by changing the units.
  • The "Effect Sizes...." figure on slide 45 is great! Make sure that is in the paper, and stays there even after the editor tells you to shorten it.
  • Put the refrences paper at the end, AFTER the appendices. Experienced readers like to flip to the references to see what the paper is about.

Talking

  • Latex works fine for making slides. Beamer is commonly used, but why? It uglifies your Latex with bad defaults. Powerpoint combines well with LaTeX, though. Tex your document. Then cut and paste what you want to put on your slide. Once it's on, Powerpoint makes it easy to resize appropriately,and lets you change the background color, insert photos, add plain text, etc. Best of both worlds.
  • Whenever you read a book, you should write notes about it. Mediawiki software is good for that. Keeping track of your notes is important too--- perhaps even more important than writing them up, since though writing them up is a good exercise, if you instantly deleted them, that removes much though not all of their usefulness. Not remembering where they are is the same as having deleted them. No-- I lie. Actually, not remembering where they are is much worse, because not only do you not have them available, but you waste considerable time trying to find them. This would not happen if we were fully rational, but there seems to be a systematic bias towards underestimating the amount of time and effort requried to find something as opposed to re-creating it. This is related to what George Stigler wrote about information--- that it is not a zero-marginal-cost good once produced, and in fact often it is easier to rederive information than to look it up.

Listening

  • The "gunner" in law school is the student who answers questions too eagerly and quickly. It is worth discussion how to be a good student-- when to answer a question, when not to answer, and when to ask a aquestion of the professor.

My daughter’s fifth grade teachers were tired of getting emails consisting of, “hi, i don’t understand the hw. can u help. thx.” So they had a lesson on it, and last night’s homework was to write an example of a proper email requesting help. Here is my daughter’s.


  • There are two parts to the skill of asking for help: manners, and substance. Manners are as important as substance. If I get an unmannerly request for help from a student, scholar, or administrator, I will very happily either ignore it or respond with a rebuke. I say "happily" because an arrogant request is one I can deny in good conscience and save myself the effort of helping. Indeed, it is my duty to refuse it, lest I encourage arrogance.
Mrs. Glaser's daughter has mastered the skill of asking politely. That is the most important and easiest skill to learn. As with most of Manners, it only requires notifying the person that they are gauche, uncouth, and liable to be despised unless they follow basic easy rules. Note that a big part of why they are likely to be despised is that it is so easy to be polite, and not knowing it shows that you must be a very ignorant person indeed. Most fifth-graders, being children, are indeed very ignorant, so this is not a moral fault, but it is one reason most people don't respect children or take them seriously. (I, by the way, take them far more seriously than average, precisely because I think about things like this.)
The second part of the skill, though, is substance. How should a person ask for help? Not just politely, but in a way that makes the recipient able to help them. Saying politely, "I need help" is close to useless, because the helper has very little idea of what you need help with. Sometimes you do need to ask for help like that, because you really are totally confused. Here, the requestor has done better. She says she needs help because "because I did not understand page 3 problem 2 of my math homework." That is not as good as she could have done, though, by a longshot. First, she needs help in English as well as math. She is old enough to know that she should write "because I did not understand problem 2 of my math homework, from page 3", which is what a person would say in natural language, rather than the crude "page 3, problem 2". More important, I think she really meant that she couldn't answer the question, rather than that she couldn't understand it, though I could be wrong there and she could answer it but not understand it, as is often the case with math--- the student can apply the formulas, but not really understand what is going on. This would be clear if she explained what she wanted the teacher to tell her about. As it is, she says the problem's instructions are unclear, but she gives zero hint as to which part of it is unclear. When asking somebody to help you with a problem, you need to give them somewhere to start. Her teacher could look at the problem and think, "Usually students are confused about part (c) and about what the word 'divisor' means" and answer that question, but that requires the teacher to put in extra effort and it might well be it was something else about the question that was confusing. It is much easier for the student to say what the student doesn't understand than for the teacher to have to figure it out, and student thinking time is an investment in future ability rather than a cost like the teacher's time, especially if it is an experienced teacher who already knows how students think.
All this is to say that learning to ask questions is a difficult and important skill. Part of it is learning how to be polite, and another part is learning how not to waste the teacher's time-- or the consultant's time, or the attorney's time, or the judge's time, since this skill comes up repeatedly over life. People who don't think about are enormously inconsiderate, coming for help and expecting the helper to invest enormous time to figure out things about what the requestor knows and doesn't know that the requestor could figure out with relatively little effort.

Social Media

Andrew Gelman's Statistical Thinking is an example of a blog with good comments. Usually heavy moderation is needed, as with letters ot the editor in a newspaper. Professor Gelman's blog is on a technical subject, and that keeps out the riff-raff.



Speaking

Have your own website, even if not your own domain, and have a file on there called temp.txt for miscellaneous stuff like this, e.g. if you want to post something on the web as a comment during a seminar. Should it be *.txt *.htm, or something else? *.rtf and *.doc are bad because the browswer won't display them and will ask to download; *.pdf requires processing when you write and this file is often for stuff you want to post within seconds.

Conferences

The first meeting we went to together was an AAAS meeting in Cleveland. This must have been about 1948, and as I recall we didn’t go to any of the talks. Instead, we went to the Cleveland Art Institute. Jimmie had heard they had a great El Greco collection. We asked where the collection was, and somebody told us to go to a particular room. We got to the room and there were no paintings on the wall. There were a stack of canvases on the floor. All the El Greco’s were on the floor, and we put on our own El Greco show by turning these canvases over and looking at them as close as our noses could get.

Another meeting that we went to that was very memorable was an IMS meeting at Ithaca. Jimmie came one way, I came another way, and we met very early in the morning at a lecture hall. We must have spent at most five minutes in the lecture hall and then we went outside and sat on the piazza. And we spent the rest of this beautiful spring day on the piazza. At this meeting, Jimmie introduced me to Egon Pearson. That was the only time I met Pearson. John Tukey was there



CODING

We had a class assignment where students had to graph some data of interest. A pair of students made the above graph, as a reminder that some data cleaning is often necessary. The students came up with the excellent title as well!


My comments:

I’m guessing, but I can imagine some foolish person decided that the best way to code “Missing” was not to use the string “missing”, because if he did that, then whoever ran a regression would have the app return an error like “Your data includes string values” instead of giving an answer. Instead, he had the bright idea of putting something like the integer “-66” or “-77” to represent missing values. I have to admit I might do that myself in a dataset for personal use, though I at least would use “-999”. Eyeballing the data, the user would see that “-66” was a nonsense value, and, being clever, would deduce that it must be the code for “missing value”.

This reminds me of the CDC, except the CDC isn’t that smart. In their online official list of covid19 deaths, check out the latest week. It will always be astoundingly low. Then look at the table notes. They explain that the data is added as it comes in from the states, and that takes 1 to 8 weeks. They don’t go on to say it, but of course this means their reported deaths numbers for the past 2 months are pretty much worthless. (Note: they do this for *totals*; it’s not just “average death rate” I’m talking about.)

This reminds me of something I like to say about music in church services. At our church, and others, we commonly have highly skilled musicians in the "praise band", who practice selflessly during the week (most are volunteers). The worship pastor has degree in baroque violin from a top five music school; the guy on the bass guitar has a degree in violin-making, and so forth. But who is the most important member of the worship team? — the guy who posts the powerpoint slides. If he messes up and doesn't put up the next slide with the words to the hymn for five seconds too long (as happens occasionally, if the leader decides to do the order differently than planned, or repeat a stanza, or just loses his place himself), the congregation is completely confused and the psychological effect is to pretty much destroy the entire hymn, no matter how skilled the musicians are or how much they practiced. The Slides Guy doesn't have to have a fancy degree, but he has to be sensible, quick, and utterly reliable-- and if need be, you could eliminate the entire praise band and still get by so long as he was there with his slides. The Data Guy is the same way.

My PhD advisor, Frank Fisher, has a good article where he says every litigation expert witness team needs a Data Guy (he was IBM's chief antitrust witness against the Justice Dept., and the Justice Dept's chief witness against Microsoft). See Econometricians, and Adversary Proceedings. Franklin M. Fisher. Journal of the American Statistical Association , 81 (June 1986), 277-286. http://www.jstor.com/stable/2289215 .

I like to go meta. This article appears on page 8 of the Google Scholar results for "Franklin Fisher" with 48 cites, just below "Identifiability criteria in nonlinear systems: A further note". Yet I don't think I talk about any other paper of his as often. (I did pick him as advisor partly because he was such a good writer, so there's a selection effect.)


Someone else commented:

The reason someone might use something like “-999” is so that when you import the dataset your stats software is much more likely to properly categorize each variable (e.g., string, integer, factor, etc.). Integers for missing values are far less likely to mess up this process whereas strings will interfere with it almost every time. Large negative integers are the standard for many professionally created datasets (e.g., General Social Survey).



Other

The reader is always right. A lot of readers are dumb. They're still right. The writer must try to make it simple.


  • Professor Marc Bloch's Strange Defeat, chapter 2 "One of the Vanquished Gives Evidence", p. 61.

No officer, I think, who was called upon to do duty with one of these regional organizations, or with a Subdivisional Area Group, will ever remember without a wry smile the incredible tangle of "measures', all duly numbered, which were supposed to be taken during the so-called "Alert Period' preceding actual mobilization. Dragged from one's bed in the middle of the night by a telegram which might read, for instance, "Measure 81 to come into force immediately', one would rush to the code-card which was always kept handy, only to find that "Measure 81' involved the implementation of all clauses contained in "Measure 49' with the exception of such of them as might have already been set in motion by the application of "Measure 93'— should the latter happen to have come into force earlier than its numerical place in the series seemed to warrant, and that, in any case, the two first paragraphs of "Measure 57' must also be acted upon. I pick on these numbers at random, and am not prepared, at this distance of time, to say that they are accurate. Those with whom I worked will, I think, agree that I have shown the whole business as being a good deal simpler than it actually was. Can one wonder that in such circumstances mistakes were of frequent occurrence? For instance, as a result of not reading instructions with sufficient care, the police authorities of Alsace-Lorraine proceeded in September 1939, to carry out a premature massacre of all the carrier pigeons of these departments. I do not say that the officers sitting far away in Paris, in a dark little office of the rue Saint-Dominique, adding figure to figure until the result was like a Chinse puzzle, were necessarily lacking in imagination. It was only that their particular type of imagination was not calculated to give them a clear idea of what it would mean to carry out their orders.


  • Include lots of summary statistics. This may not be a new contribution of your paper (unless you are using new data), but it could be the biggest contribution, and the one that would get you the most readers. Of course, give ample credit if you aren't the first to do it. Data description is an important scholarly contribution, even if a humble one. It is really more important than statistical analysis, even if it is easier.

WRITING. Rob Henderson tweets a quote from a study:

"there isn’t replicable evidence that animal abusers are more likely to commit violent crime...40% of female college students and 66% of male college students admitted to having abused animals" https://uk.sagepub.com/en-gb/eur/the-sage-handbook-of-evolutionary-psychology/book267401#contents https://uk.sagepub.com/en-gb/eur/the-sage-handbook-of-evolutionary-psychology/book267401#contents


That's a good example of how lack of confirmation by statistical research is completely unimportant, since it's hard to imagine how good data could be collected. Sometimes anecdotes are vastly superior to numbers. This comes up all the time. It's very common with studies of the effect of different kinds of upbringing, anything to do with self-reported crime, rare events (cf. The Lizard Man Constant), lots of epidemiology, lots of claims about moral or immoral behavior's effect, mental health... There are a lot of subjects where not only is most of the research incompetent garbage and biased, but you can't even think how a good statistical study could be done with feasible data, even if you had a million dollars to collect it yourself. On the other hand, many of those topics are perfectly amenable to study by other means-- introspection, case studies, anecdotal evidence, interviews, and so forth. As Aristotle said inBook I of the Ethics :

Our discussion will be adequate if it has as much clearness as the subject-matter admits of, for precision is not to be sought for alike in all discussions, any more than in all the products of the crafts. Now fine and just actions, which political science investigates, admit of much variety and fluctuation of opinion, so that they may be thought to exist only by convention, and not by nature. And goods also give rise to a similar fluctuation because they bring harm to many people; for before now men have been undone by reason of their wealth, and others by reason of their courage. We must be content, then, in speaking of such subjects and with such premisses to indicate the truth roughly and in outline, and in speaking about things which are only for the most part true and with premisses of the same kind to reach conclusions that are no better. In the same spirit, therefore, should each type of statement be received; for it is the mark of an educated man to look for precision in each class of things just so far as the nature of the subject admits; it is evidently equally foolish to accept probable reasoning from a mathematician and to demand from a rhetorician scientific proofs.


Common Knowledge and Referee Reports. In game theory and philosophy, that means A knows fact X and B knows it; A knows that B knows X; B knows that A knows X, and the whole infinite regress. In referee reports, it's common that referee A knows he wrote a scathing report; and author B knows that A wrote it by deduction from internal evidence in the report even though the report is anonymous (e.g. the referee says to cite A's obscure paper, or A is the obvious referee for a paper on that topic and the report has exactly the criticisms A would make). But A does not know that B knows who wrote the report, it may happen, or at least A can pretend that he does not know, when he meets B at a conference. And even if A knows, B cannot be sure that A knows that B knows that A wrote the scathing report, so B can pretend to A that he doesnt' know, over drinks at the hotel bar at the conference. And both know it is totally bad form to bring up the report-- or, at least, to ask who wrote it. (I guess author B coudl bring it up at the bar, so long as he said, "I don't know who wrote it, but I got the most scathing but useful report..." as a way of thanking referee A, who would then pretend not to know who wrote it.)


WRITING. Sam Rosenwrote on Twitter  :

What do you use for managing your todo items across multiple projects? I want to stop using my inbox as the primary tool and interested to crowdsource some better methods. #EconTwitter

8:58 PM · Aug 11, 2020·Twitter for iPhone


My answer: I use plain text or *.doc files. I have 00todo.txt for an unordered list, 00calr.txt for calendear to which items of the day are copied, and 00ostrac.txt, 00flynn.doc in project folders for those projects. Avoid apps, anything to learn or that needs care in use.


Some Oscar Wilde quotes froma webpage that didn't have sources .

“It is what you read when you don't have to that determines what you will be when you can't help it.”

“An idea that is not dangerous is unworthy of being called an idea at all.”

“Experience is merely the name men gave to their mistakes.”

“The imagination imitates. It is the critical spirit that creates.”

“I was working on the proof of one of my poems all the morning, and took out a comma. In the afternoon I put it back again.”

"The only thing to do with good advice is pass it on. It is never any use to oneself.” (also wrong and discussable)

“Be yourself; everyone else is already taken.” (wrong, but to start a discussion)

“Words! Mere words! How terrible they were! How clear, and vivid, and cruel! One could not escape from them. And yet what a subtle magic there was in them! They seemed to be able to give a plastic form to formless things, and to have a music of their own as sweet as that of viol or of lute. Mere words! Was there anything so real as words?”


SOCIAL MEDIA. Twitter and blog comments. I should add a discussion of this. Don't do it just to cheer, or to express your opinion. Nobody cares about your opinion if you're anonymous. It's just venting. It's socially pernicious, pollution. Anonymous commenting has its place, but not to allow for poor writing. It is more like the usefulness of anonymous refereeing-- but there you will shame before the editor, at least, if you act like a scoundrel.


I recall Peter Diamond saying somethign like that nobody should write about methodology until age 40, at least. I'm over 60 now, so I'm entitled to, I guess. Joseph Epstein said soemthing similar. I think he said you shouldn't read Proust till you're over 60. Or you won't enjoy it till you're over 60, maybe.


https://www.rasmusen.org/blog1/liveblogging-the-return-of-in-re-flynn/ George Mason Professor Garrett Jones tweeted of my Rule 48 theory, “I’m sure your theory gets an asterisk or two.But I’m not sure your theory gets a high R-squared.” Nicely phrased. I will think about that. It’s a good way to think about explanatory theories. What he’s saying in econometrics terminology is that my theory is definitely true and explains a few cases maybe, but it doesn’t explain most of what happens.

That’s just one case. I’m sure prosecutors file Rule 48 motions now and then, and that they are granted in 99 out of 100 cases, perhaps 999 out of 1,000, perhaps 9,999 out of 10,000. Of course, any theory looks bad at explaining those cases, because in those cases, Rule 48 is just a bother. It makes the prosecutor draft a motion that’s routinely granted. But that’s like saying having parachutes for fighter pilots is a total waste of money 99% of the time, because planes hardly ever crash. Worse, even. Having a parachute doesn’t prevent a crash, but having Rule 48 prevents prosecutors from dismissing cases the Attorney-General wants prosecuted. It is *because* of Rule 48 that Rule 48 motions are so useless.

WRITING.

Editing as Mansplaining
Editing as Mansplaining

Rewriting Abstracts. One thing I often do in refereeing and in listening to papers being presented is to rewrite the author's abstract. This is a a good exercise even if you think the abstract is good. Then, you might imitate how Benjamin Franklin says he taught himself to write. Read the abstract carefully,to understand what is going on in the paper, and then put it away. Then write an abstract for the paper. When you're done, compare what you've written with the original. As Franklin said in his Autobiography,

About this time I met with an odd volume of the Spectator. It was the third. I had never before seen any of them. I bought it, read it over and over, and was much delighted with it. I thought the writing excellent, and wished, if possible, to imitate it. With this view I took some of the papers, and, making short hints of the sentiment in each sentence, laid them by a few days, and then, without looking at the book, try'd to compleat the papers again, by expressing each hinted sentiment at length, and as fully as it had been expressed before, in any suitable words that should come to hand. Then I compared my Spectator with the original, discovered some of my faults, and corrected them. But I found I wanted a stock of words, or a readiness in recollecting and using them, which I thought I should have acquired before that time if I had gone on making verses; since the continual occasion for words of the same import, but of different length, to suit the measure, or of different sound for the rhyme, would have laid me under a constant necessity of searching for variety, and also have tended to fix that variety in my mind, and make me master of it. Therefore I took some of the tales and turned them into verse; and, after a time, when I had pretty well forgotten the prose, turned them back again. I also sometimes jumbled my collections of hints into confusion, and after some weeks endeavored to reduce them into the best order, before I began to form the full sentences and compleat the paper. This was to teach me method in the arrangement of thoughts. By comparing my work afterwards with the original, I discovered many faults and amended them; but I sometimes had the pleasure of fancying that, in certain particulars of small import, I had been lucky enough to improve the method of the language, and this encouraged me to think I might possibly in time come to be a tolerable English writer, of which I was extremely ambitious.


I wouldn't recommend going so far as to turn the abstract into verse, though that might well give you deep insights into it. Franklin tell us his purpose in versifying was to expand his useable vocabulary, which is not so important for academic writing-- though do take his point that merely having learned a word at one time and putting it in your deep memory is not the same as having it at hand in your RAM for ready use.

My purpose in rewriting abstracts is different, though. It is twofold. First, since I need to referee the paper or to sit through an hour and a half of a seminar on it, I want to understand the essence of what is going on. I am already trapped, so the abstract is not serving its usual function of telling me whether I ought to read the entire paper or skip it, but having been enticed into the author's lair already, with no hope of escape, I want to be prepared for the monsters or treasure within. If I have to rewrite the abstract-- or, indeed, if I have to write one from scratch-- I have a good start for going further. Usually this involves reading more than the abstract, of course, and the worse the abstract, the more of the paper I have to read to be able to write a good abstract. If the abstract is perfect, I can read it, and that is that. If it is very good, I can read just the abstract and figure out what it should say and improve it, maybe a lot, maybe just fixing a word or two. Ordinarily, though, I must read further. For a typical good paper, I just have to go on to skim the introduction, which will tell me what the paper is about. Maybe I will also have to read the conclusion. Or I might have to go on to read just the modelling assumptions, or just the theorems or just the regression tables or just the paragraphs immediately after the main results. If it is not a very good paper, I may have to read the entire paper before I find that it doesn't really have much to say.

Many referees do this, perhaps a majority, though they don't think of it as rewriting the abstract. They think of it as explaining to the editor what the paper is about, and they write a paragraph or two at the start of their report summarizing the paper. Authors need to read those summaries carefully, because it shows what expert readers think your paper is about, and if it's different from what you think it's about, you should know that. Either you have to rewrite so your readers understand, or you are wrong about what your paper is about. We smile at that thought, but it is quite often true that the paper doesn't understand what the best part of his paper is, perhaps because he started off writing up one idea but ended up coming across a different idea but didn't realize its importance. Don't be Captain Ahab, looking for Moby Dick and ignoring perfectly profitable other whales he meets on the way. Worse, not infrequently scholar thinks he's actually succeeded and caught Moby Dick but actually he's caught a big worthless carp that he loves and a couple of small but delicious bluegill that are hidden under it in the bottom of the bucket.

So, one reason to rewrite the abstract is to understand the paper, and a corollary to that is that if you show your rewrite to the author, he can tell if he's written his paper in such a way that you misunderstand it. The other reason is to help the author and everyone else to understand the paper. If you have improved the abstract, it can replace the original. You have done your good deed for the day. The author can use it instead of his own, or at least take whatever bits are improvements and ignore the edits that are worsenings.

As with any comments, abstract comments are worth getting. There is some cost to reading comments, but it is negligible, especially considering that one need not eat an entire egg to discover that it is rotten. A math metaphor conveys the point. Suppose your paper is the set P = \{-1, 0, 4, 7\} and your objective is to maximize the sum of the top four elements (10, currently). The comment set is C = \{-9, -7,-5, -2, 0, 1\}. Clearly, your commenter is either not as smart as you are or has not devoted as much time and energy to his comments as you did to your paper, since his 4-element score is -22, and it's definitely not publishable. But do not scorn his comments. If you read them and accept what is good, your paper will be P' = Max4(P, C) = \{0, 1, 4, 7\}, with a score of 11, not just 10. The scholar should not despise comments like set C. He should yearn for them. If your correspondent finds a typo for you, he has improved your paper. If he says he can't understand a sentence but can't say why, and you are stimulated to find a way to write it better, he has improved your paper. Look back to Alex Perkins's Twitter comment above. It was perhaps meant to be derogatory, though it is polite enough to be perfectly appropriate and as an economist I actually find it a compliment to be told that I "take mansplaining to an entirely new level", because I read that as saying, "Uses logical thinking to say things that are correct and useful but which I would never dare say, or even conceive of saying." But the author is a biology professor at a good university, a smart guy, and it conveyed to me that many people dont understand the usefulness of comments or the importance of abstracts and that this would be a good hook for the discussion I have just written. (We will leave the war between economists and epidemiologists for another day.)


WRITING STYLE. @vreshetnikov tells us:

I often use ?…? (heavy quotation mark ornaments, U+275D, U+275E) as "irony/sarcasm quotes". I cannot say for sure when and why I started to do that, but it seems kinda natural for me. Anybody else using them for that purpose? Or at least, understand them as "irony quotes"?

I think I associate them with oversized items that are often used by clowns for comical effect.



WRITING. Professor Jesse M. Shapiro writes in Four Steps to an Applied Micro Paper

Step 1: Aspirational introduction. 1. Write an introduction for the paper you aspire to write. Feel free to make up your results, within reason. 2. When you think you are done, ask yourself: if I write the paper outlined in this introduction, will I be happy with it? 3. If so, move to step 2. If not, stay on this step. Chances are that if imaginary results don’t excite you, neither will real ones.

He has lots of other good stuff there too.


WRITING. "Clearly", and "As I have shown". "As I have shown" is better. It is a challenge to readers, a dangerous rhetorical move. "Clearly" is a sneaky one, often meaning, "I can't explain why I think this, but ..." It is used that way in math sometimes too, for the hard or tedious bits in a proof.


WRITING. Indiana Policy Review:

It wasn't too much longer before the AP prohibited the use of “illegal immigrant” to describe immigrants who were here illegally or "Islamist" to describe attacks that were done in the name of Islam. The AP explained that the terms were dehumanizing. The AP Stylebook had gone to war with the core purpose of journalism, i.e., to accurately describe what its reporters can see and hear.



Captain Professor Sir: Some Lessons from Michael Howard by Beatrice Heuser:

Sitting in an antique armchair in his exquisitely decorated office at Oriel College (I have a vague memory of pastel colors including light green, grey, and pink, which he also sported in his ties), an ornate 18th-century golden clock ticking away above the fireplace, he shared with his student a cup of tea or a crystal glass of sherry as well as his recipe on how to write a good lecture or chapter. It was derived from the old Oxbridge essay style: Do all your reading, then retire for the evening with a good glass of red wine. In nocte consilium: Rise early, write the whole thing in one go, and then go back to your notes to insert the footnotes.



From Twitter:

Replying to @NoahMHorwitz I had the experience of trying a case in front of a visiting, retired Texas CCA judge. I made sure all my case law was an opinion he wrote or concurred in. He started laughing the 2nd time I did it. Needless to say, I won all the evidentiary rulings. If they laugh, that's fine, so long as you win.


"Aposiopesis is the rhetorical device of breaking off in the middle of speech. The sentence or thought is unfinished and the end left to the imagination of the interlocutor or audience," says the entry at literarydevices.com. "Aposiopesis always occurs midway through a sentence or thought that is left unfinished." Thus, this word does not apply to a long pause after a sentence meant to allow it to sink in. What is that called? (Nor is it Apostrophe-— breaking away from one person to address another.) From Exodus 32:

31 And Moses returned unto the Lord and said, “Oh, this people have sinned a great sin, and have made themselves gods of gold. 32 Yet now, if Thou wilt forgive their sin — and if not, blot me, I pray Thee, out of Thy book which Thou hast written!”


Epistrophe. Repetition of a word at the end of successive clauses or sentences.

Fisking.

{{{1}}}

See, for example, "A fisking of Indiana University Provost Lauren Robel's memo about me".

Gibe is a good word that we neglect. Pronounced as jibe, the alternative spelling. Merriam-Webster says:

to deride or tease with taunting words. GIBE implies taunting either good-naturedly or in sarcastic derision. ... hooted and gibed at the umpire... First Known Use of gibe 1567, in the meaning defined at intransitive sense... perhaps from Middle French giber to shake, handle roughly.


Rapidreadresponse. A real-time response commentary on a legal opinion or new story using boldface to indicate the interesting bits to the reader and some other obvious font difference (Courierfont, italics, 14 pt) for the author's commentary. MS Word makes it easy. See, for example, rasmusen.org/special/mandamus/InreFlynn-DCCircuit-june24-2020-boldfaced.doc . See fisking above; a fisking might be done as a rapidreadresponse or it might be done much later than the document commented upon was written; and a fisking is always adversarial and contemptuous, whereas a rapidreadresponse usually is not.

Wunderwerk. A German word useable immediately in English without translation, meaning wonderful piece of work, I expect. Anyway, that’s how I’ll use it, and who is master, me or the word? (allusion to Alice in Wonderland)


WRITING. Weisbach advice, Grumpy Economist, John Cochrane, April 16, 2020.

I encouraged Mike to have a longer discussion of what not to do. Much writing and self-improvement consists of editing, recognizing simple mistakes and fixing them. "Write clearly" is hard to follow. "Don't tell us why it's important before we know what it is" is easier to follow. I think the next draft will have more of that.

[from Weisbach, commenting] Blog readers will be interested in knowing that John sent me 17 pages of suggestions the other day, which goes well beyond what anyone can reasonably expect to receive.

[anonymous comment] One "rule of thumb" I learned at U Chicago (I'm thinking it was from Stigler) was that your research will remain relevant for approximately as long as the age of the articles in your footnotes. (Reference articles written in 2010, you'll be read until about 2030).



Louella Vaughan on Twitter:

Set a reasonable target for the day (1,000 words works for me).

NO leaving the office (lunch excepted) until the 1,000 words is written.

Small treat for hitting target early.

A few late nights/missed social events tends to fix any writer's block.



Marc Steen on twitter:

There are two types of stories that you can tell (in your thesis or paper): Look, everybody thinks X is complex, but it’s very simple, I’ll show you. Or: Look, everybody thinks Y is simple, but it’s very complex, I’ll show you.

As a rule of thumb any piece of prose can be compressed ~20% without losing information. This advice should only be applied iteratively in case of emergency.


Apply the 20% cut iteratively till you get tired. Actually, in doing this, you may find no meaning ever existed: if you do it iteratively, you may find it still works, with infinite iterations. That means you had no information to start with. So another way to look at it is: iterate till you start losing information.

Note that there is usually a tradeoff, at some point, between clarity and information, and that points comes before the point where you should stop cutting. You can usually cut 20% with zero loss of information and with gain in clarity. At some point beyond that--call it point B— you will start losing information, but while you are still gaining clarity. Keep cutting beyond point B: it's worth losing a little information to gain a lot of clarity. At point C, the loss from information reduction begins to exceed the gain from increased clarity, in terms of value. Stop there. If you do keep going, eventually you will reach point D, where not only are you reducing information, you are also reducing clarity.

It is also important to distinguish between precision and clarity. In most writing, it is clarity that matters, not precision. Precision is pedantic, exact, bloated, and boring. It is fine for computers-- quite clear to them-- but bad for human readers. For that matter, it would be bad for Martian readers too, or for our future robotic masters. Intelligence involves the screening out of irrelevant information, whether it be human intelligence, extraterrestrial, or artificial. I almost wrote "divine", but I suppose someone omnipotent has enough spare brainpower not to bother with screening, and omniscience makes it irrelevant anyway.

In some contexts, however, precision is more desirable than clarity. The Tax Code is an example. It is not meant for easy reading. It is meant to cover all possible cases, and to provide an algorithm for coming out with answers. Tax Law is the only law school course with problem sets. (Slight exaggeration? Tell me, readers.) When I started writing on tax law with Professor Mark Ramseyer, who had been a tax lawyer in Chicago in his youth, he said, "Eric, you have to understand that tax law is different. There are right and wrong answers." And I found it is by far the hardest kind of law to understand, because it is so logical and precise, and you have to understand that terms are defined very precisely but very differently when used in different places. (For the results of our efforts, see "Can the Treasury Exempt Companies It Owns from Taxes? The $45 Billion General Motors Loss Carryforward Rule" J. Mark Ramseyer and Eric Rasmusen, The Cato Papers on Public Policy, Vol. I, Article 1, pp. 1-54 (2011) edited by Jeffrey Miron, http://www.cato.org/store/books/cato-papers-public-policy-paperback. It was "real science", because we had to conclude that General Motors didn't cheat on its taxes after all-- but the bailout was still scandalous!) Similarly, in instruction manuals for equipment repair, it often makes sense to be precise but unclear. If you are precise, that means a reader can go through the instructions blindly, without understanding what he is reading, and obtain the desired result. In most writing, though, you are trying to get something across to the reader--information or an idea-- rather than just providing an algorithm for him to use to put data in and get an answer out.

----

Writing to Persuade—Review of "Legal Writing: A Judge's Perspective on the Science and Rhetoric of the Written Word"

When I asked him about what impressed him about the writing of great advocates he’d studied, he said that he was struck and surprised at how easy it was to read. It wasn’t the magical turns of phrase, but that you never had to pause. Clarity, not personal style, was the trademark.

----  

Whistler, the painter: "Two and two the mathematician would continue to make four, in spite of the whine of the amateur for three, or the cry of the critic for five"


Hephsiba R Korlapati on Twitter: Write with any quill. And Edit with full will. --Hephsiba R Korlapati IAS (@hephsyyy) ">May 26, 2020

Write with any quill. And Edit with full will.

Write first with any quill, Then edit with full will. To wait will paralyze-- But endlessly revise.

This can be paired with

The best is the enemy of the good.

but

The good is the enemy of the best.



WRITING STYLE. There's argument in law reviews over whether to require "Smith et al" instead of citing "Smith, Jones, and Anderson".

@AlaLawReview

the et al. rule is silly, saves almost no words, and has discriminatory effects. Abandon it!

Quote Tweet

It's the tip of the iceberg. Many, many, law abbreviations are stupid in an age when typesetting is cheap and readers' time is expensive. Basically: any abbreviation a reader must look up or that reduces info content is an abomination.


Jorge Luis Borges, Collected Fictions, translated by Andrew Hurley.

…In that Empire, the Art of Cartography attained such Perfection that the map of a single Province occupied the entirety of a City, and the map of the Empire, the entirety of a Province. In time, those Unconscionable Maps no longer satisfied, and the Cartographers Guilds struck a Map of the Empire whose size was that of the Empire, and which coincided point for point with it. The following Generations, who were not so fond of the Study of Cartography as their Forebears had been, saw that that vast Map was Useless, and not without some Pitilessness was it, that they delivered it up to the Inclemencies of Sun and Winters. In the Deserts of the West, still today, there are Tattered Ruins of that Map, inhabited by Animals and Beggars; in all the Land there is no other Relic of the Disciplines of Geography.

—Suarez Miranda,Viajes de varones prudentes, Libro IV,Cap. XLV, Lerida, 1658



Articles to Incorporate.

Daniel S. Hamermesh "The Young Economist's Guide to Professional Etiquette", 1992, JEP.

Weisbach, Michael (2020) A Field Guide to Economics:A Scholar’s Introduction to Research, Publishing, and Professional Development.

A tweet on econometrics writing.

``How to Read a (Quantitative) Journal Article." Prepared for Sociology 210 Greta Krippner September 21, 2000.

``An Interview with Amitabh Chandra, Editor of the Review of Economics and Statistics." by David Slusky, American Society of Health Economists Newsletter Issue 2019:4.

``Writing Tips for Ph. D. Students," John H. Cochrane http://gsbwww.uchicago.edu/fac/john.cochrane/research/Papers/ June 8, 2005.

Code and Data for the Social Sciences: A Practitioner's Guide Matthew Gentzkow Jesse M. Shapiro.

Franklin M. Fisher (1986) [ http://www.jstor.com/stable/2289215  ``Statisticians, Econometricians, and Adversary Proceedings,]
Journal of the American Statistical Association , Jun., 1986, Vol. 81, No. 394

(Jun., 1986), pp. 277-286.

Gelman, Andrew (2009) ``Advice on writing research articles,

30 July 2009.

Gelman, Andrew (2014) ``How to read (in quantitative social science). And by implication, how to write, 2 December 2014

Jonathan B. Berk, Campbell R. Harvey, and David Hirshleifer (2017) [https://doi.org/10.1257/jep.31.1.231 doi=10.1257/jep.31.1.231 How to Write an Effective Referee Report and Improve the Scientific Review Process] Journal of Economic Perspectives—Volume 31, Number 1—Winter 2017—Pages 231–244.

``How to Read in College Staying Afloat: Some Scattered Suggestions on Reading in College, {\it Easily Distracted Culture, Politics, Academia and Other Shiny Objects}.


Hans-Martin von Gaudecker Effective programming practices for economists: Slides

A Black executive assistant quit after Condé Nast CEO gave her an English language guide as a gift



See also:
How to Run Online Talks
Writing
Talking
Listening
Coding
Tables of Numbers
Figures and Diagrams
Social media
C. P. Snow, Good Judgement and Winston Churchill
Indefinite Pronouns
Writing Right Right Now.
Writing Style.
Rewriting Abstracts
Diagrams.
Careful Writing Requires Work

---------------------------------