Saturday, September 20, 2014

Boots on the ground in studying cities: Street View or data analysis are never enough.

When I wrote you about using a close-up aerial view, I was making a suggestion. Not a demand. I thought it helped me understand where I was. And the birdseye view in Bing is even more helpful.

My goal is to give you the knowledge and skills to understand places in context. Maps, frontal views and other views you make by being there (over a fence, adjacent places,...), Street View, aerial and birdseye views, all help. What you are trying to do is to become fluent in all these media and methods.

I always have other professors asking me why I send you to places, since Street View has done it. But in fact Street View is partial, has no sense of the life of a place, may well be outdated, etc. More generally, you really have to go look and be there, in the city, if you want to understand what to do--whether you are a developer, a planner, etc. The deluge of visual data we now have is no substitute for boots on the ground, so to speak. Similarly, when you learn to do statistical analysis of a data set as in the Census or a survey, you still have to ask if it makes sense in terms of the people you are studying, and it is vital to do fieldwork and interviews to see if your analysis has ground truth (as they say in aerial surveillance). Good regression coefficients may be statistically nice, but your model may have missed important stuff. A good R-squared is no substitute.  As someone once said to me, about rats in psychology, you've got to talk to your rats.

Friday, September 19, 2014

Focused Paragraphs, Questions: What's the Big Idea?

In reading over students' discussion of others' work, I have some suggestions that may be more generally useful.

1. You have good ideas but they are often buried by lots of stuff. If you write a paragraph, make sure you put a topic sentence at its beginning that states the main point--which you may only discover after writing most of the paragraph.

2. The same for the questions you may ask at a seminar. You may be thinking "out loud" as you write the questions, but after doing that, underline the main question or questions. They need to be sharp and not very long. You can put all he context and qualifiers after the questions, or rescue the main question from the middle of your question paragraph.
3. What this all says is that you want to examine your work for the main points. So if you have written a paper, you want to then write an introduction that says the main ideas in a sensible way. Put differently, could a busy person get your points by reading the first sentence of your section or paragraph, etc. Is the introductory two pages sufficient for them to figure out your conclusions etc. 

What you want to do is not point what's wrong with work, so much as figure out what the speaker should do next. Or even better, what should be her main contribution? (If the paper is truly vulnerable, think about how to make it stronger.)

One other point. At the recent Health Economists conference here in June, there was an opening session of Peter Orszag and Casey Mulligan. Mulligan had just launched a book on how the responses to the Great Recession actually harmed economic growth. Orszag said that if we want to get at the 1/3 of health expenditures that are wasted we needed not only deal with supply, we'd have to deal with insurance. Very different approaches: Mulligan examined things as a first-rate economist. Orszag examined things in terms of what you needed to do to tackle a big problem. Neither was much moved by the other--since Mulligan is concerned with getting the numbers right, and Orszag was concerned about transformative change and politics. They were not talking past each other, they were talking about different things.

Tuesday, September 16, 2014

Asking for Help. Graduate Student and Post-Doc. It not "internalization of the aggressor," but taking care of your students as you wish you had been cared for.

The research literature suggests that a graduate advisor or a postdoctoral advisor can make a big difference in the training of a student and beginning researcher. Some advisors are proactive, others less available, others impossible.

But you have to ask for help, be persistent, and not be sidelined. If your advisor proves unhelpful, you need to find another, pronto! If you don't understand something, or need to learn something which you believe everyone knows or that should be easy to learn, but it is not for you, find someone to help you. If people put you off, keep looking.

Some advisors take it upon themselves to train people, others may leave it to postdocs to do the training of the student. But postdocs are so involved with their own survival and careers, they may be of little use. You have to persist. You need to make lists of what you don't understand, make sure you are not a tool in a research program, but are being educated.

Usually there is lots more to learn after you have completed your coursework and exams. And that is true after your PhD and even later. Your advisor should in effect be demanding, making sure you have learned. But many practice sink-or-swim, figuring it was good enough for them, so it will be good enough for you ("internalization of the aggressor"). Nonsense.

If you are stuck in a bad place, it's almost always better for the long term to get out of it if you can find another landing place, even at a different institution. The cost of not doing so may well impact your future career, and your sense of your competence.

Of course, you could decide this is not for you. That's a different question. Here we are concerned with the education and training and discipline you can learn from an experienced teacher. And it is not a matter of independence from your advisor--it's a matter of having a supportive person watching out for you.

Some are independent souls, able to do it all on their own. I know of very few.

As for advisors--you have to be proactive, work with your students and postdocs, be specific and demanding, and be open to mismatching and mistaken placements. It's simple: would you want you children being treated as you treat your students?

Saturday, September 13, 2014

By the fall of 1966, I was prepared to abandon mathematics and to turn to some other life, a first step being a year or two in Turkey with my wife and children, as a prelude to an existence whose exact form was undetermined.

 Robert Langlands is a distinguished mathematician and here he is describing a mathematically low-point in his life.


I, myself, was searching for a general notion and had despaired. By the fall of 1966, I was prepared to abandon mathematics and to turn to some other life, a first step being a year or two in Turkey with my wife and children, as a prelude to an existence whose exact form was undetermined. I, who had never been anywhere outside of English-speaking North America, returned to the study of Russian and began the study of Turkish, frivolously daydreaming of a trip to Turkey --- with wife and four small children --- through the Balkans or through the Caucasus. In the end we arrived in Ankara by a more banal route. Even with the Russian and Turkish, I had time to spare and began, as an idle amusement, to calculate the constant term of the Eisenstein series for various rank-one groups. I had, curiously enough, never done this before. I discovered rather quickly a regularity of which I had been unaware. It was described in the lectures delivered at Yale some months later and included in Part 3 of this collection. The constant term, or rather the second part of the constant term, the part that expresses the functional equation was there denoted M(s) and given at the very end of §5 as a product that I write here as
i=1rξi(ais)ξi(ais+1)(1)
r being a small integer, often 1, and ai being a positive number. Suppose, in order not to confuse the explanations, that r is 1. The issues arising in the general case are treated in the references. It is the relation expressed by (1) that suggests and allows the passage from the theory of Eisenstein series to a general notion of automorphic L-function that can accommodate not only a non-abelian generalization of class-field theory but also, as it turned out, both functoriality and reciprocity. It was the key to the suggestions in the Weil letter. The Yale notes were written a long time ago and were hardly exemplary expositions. I have no desire at the moment to recall the details or to improve their presentation --- the reader is encouraged to consult the writings of Shahidi, for example the book Eisenstein series and automorphic L-functions --- but there are a number of points to which I would like to draw attention, and it is more convenient to refer to my own notes. I repeat, first of all, that (1) refers to rank-one parabolic subgroups, thus to Eisenstein series arising from maximal proper parabolic subgroups, so that it does not require the second or the third steps and is analytically at the level of Selberg's original arguments.

Thursday, September 4, 2014

Limits vs. Going over the Top: Beating the Odds by Grit and Perseverance

http://cultivatingthought.com/author/sheri-fink/


 On a Chipotle bag is Sheri Fink (PhD, MD) on the case against limits. She wrote the book on the New Orleans Hospital during Katrina.  Her point here is that much of our inquiry about rationing ignores the fact that people want to do it all, and some are willing to try.


"Then I think about that innovative, can-do attitude. That refusal to accept limits, quintessentially American. A student nurse says: on my ward we won’t give up on anyone. It is unrealistic. It is exactly what might save us."

Similarly, it may well matter that you not believe the odds and try to beat them, say as an entrepreneur. 

Saturday, August 30, 2014

Lectures

There was an article in The Atlantic about Minerva, the new high-tech institution of higher education. One of its main points is that conveying information or teaching the basics of many fields is likely to be shifted to online courses. Minerva's faculty makes fun of the lecture, and for good reasons.

BUT, when I lecture I am thinking in front of a class and responding to questions and to implicit questions conveyed by the students' attitudes. I have little to convey in the way of material or techniques, nothing that is well tested by means of examinations. If I can convey to them how to think by my example and if they figure out how to do the projects I assign--that's good. Of course, you have to suspect that your way of thinking in front of a class is authentic and perceived as such by the students and it becomes part of their minds  You have to give them a sense of what it means to think and solve problems and be confused and understand. Since it is the only strength I have, and it seems my students do appreciate it, I am OK. But I am always unsure, for I have no crutch of conventional content.

Perhaps it would be good to conduct the class as a seminar, lots of back and forth, maybe like a law school class with its "Socratic" method.

Wednesday, August 27, 2014

Why apply for grants and fellowships?

I recently applied for a grant, and did not receive it. My hit rate is maybe 50%, so it is not surprising. 

I wrote:

I appreciated the chance to think about this. Making a proposal has been for me always a way to think ahead. 

I will still do the work, since it is part of my ongoing project, albeit it will have to be fit in with other stuff. But, again, thank you  for the chance and instigation to begin this work. 

I always think of a proposal as a chance for an institution to be attached to work I will do. (Similarly, I think of tenure as a way that an institution keeps its people out of the job market.) This reverses the usual analysis, but I think is helpful."

My point here is that in applying for support, what you are doing is projecting yourself forward, putting your best foot forward. So you are pushed into what comes next.

The problem funders have is finding work they want attached to their brand. From what I have heard, there are many cases of rejected proposals that have led to very significant work, work that would have enhanced the funder. This may be urban legend or academic legend; and it may be that the funded projects are really very well chosen. No one expects funders to make perfect decisions--and there will always be missed opportunities. But to reject a proposal is more of a problem for the funder than for the researcher, who is likely to pursue the work one way or another. 

 
It's not about sour grapes and the like on the part of the researcher. Rather, it's taking the grant process and viewing it from the funder's perspective. To apply for a grant is to give the funder a chance to participate in the work.

Tuesday, August 5, 2014

Laying Golden Eggs

J. C. Ward was a distinguished physicist, who had published not very many articles in his career. But the name is ubiquitous because some of the work is essential in quantum field theory, and in other fields. But he felt that he could not lay such golden eggs with any regularity, and he did not see how he could produce regularly however the quality. Ideas he considered worthy were rare for him.

(I should note that few scientists can lay golden eggs with any regularity, and most who are  productive mostly produce just the normal white ones, brittle as they are.)

Ward preferred to be at a third rate place or nth rate place where he did not have such pressure to being at a first-rate place with the pressure and the sense that he was inadequate to the task.

Many of us might say much the same as Ward, but  what we do produce is much less significant than Ward's work. In other words, most of us are not too good to lay regular eggs. And as far as I can tell, Ward himself had no such high opinion of himself, but an acute sense of what he could and could not do.

Your productive colleagues are ordinary hens, maybe once in a while laying a double yolked egg. You very productive colleagues are likely to hit one out of the ballpark (a bad analogy for actual eggs, to be sure) once in a while, and more likely contribute to a subject area.

As I said, almost no one who I have met who disdains producing ordinary eggs has ever laid a golden one or even a double-yolked one.


Monday, July 7, 2014

The Standard Model

I spend my time in a school of public policy, a field much dominated by social scientists such as economists. It is striking how their best theories have very weak mechanisms to account for what happens in the world, how often it is a matter of A being likely related to B, but with no idea of the size of that relationship. They do have an idea of what to do if there is a problem, and it would seem that their intuitions are helpful. They can model an economy in equilibrium, but it would be helpful to know how sensitive their models are to assumptions and to less than equilibrium, and how much they would bet on a claim they are right. There is remarkable disagreement among, them, only some of which is scientific.

I believe I understand their basic microeconomic ideas and some of the Nobel-Prize improvements. I'm less sure about their macroeconomic ideas, but the basics seem clear. I am not sure if there are any quantitative tests that are decisive.

It is almost surely the case that when we measure or theorize about society, if we know the order of magnitude of a relationship, is it say 1% or 10%, or positive or negative, we are making progress, and it is likely that we know such numbers with at most two significant figures, usually just one--0.1 is different from 0.2, but unlikely that 0.16 is different than 0.17 once we are honest about how well we can measure something taking into account confounding effects. This is not an indictment of economics or social science, for to learn that A is actually related to B, and to get that relationship roughly right with an honest sense of how good is our measurement is a major achievement. It is also a major achievement to identify significant phenomena, such as the role of information asymmetry in markets, and indicate how they work. But the next question is how much and when.

I was brought up in a different field, one that was not so precise fifty years ago.  Over the last forty years, elementary particle physicists--experimentalists and theorists--have developed a remarkably good account of the basic constituents of matter. The Standard Model, as it is called, accounts for forces between nuclei, between charged particles, and for radioactive decay--the strong, electromagnetic, and weak forces, as they are called. Gravity has not yet been incorporated in that model, but general relativity (Einstein's theory) is powerful in explaining the largest phenomena we see at the everyday to cosmic scale. Those theories have proven valuable for those studying the origin of our universe, for it tells them just what kinds of particles existed in the early hottest state of the universe, in its first few minutes, and they have used that knowledge to figure out much of what happened in those first few minutes. And what has happened after.

The experimental physicists and the observational astronomers have made unbelievably accurate measurements of the world we live in, unbelievable to anyone who was in graduate school when I was in the 1960s. Practically, we now have an origin story about our universe that is quite rich and accurate. It is surely the greatest gift from physics to religion and metaphysics. But the physicists do not pretend to have an account of our moral and personal origins, and a wide variety of explanations, from religious to anthropological are offered. Moreover, there is no conflict between the physicists' account and the moral accounts. They occupy different realms.

What is most remarkable is the accuracy of the Standard Model. It is not unusual for the theory to predict not only the rough size the effects, but to predict them to an accuracy of three decimal places. Enormous effort has been required to make that theory useful to such accuracy, but it works and it works quite well.

Any such theory will have stuff it does not explain or explains inaccurately. Sometimes we have to learn to calculate more accurately, sometimes we will need a more comprehensive theory, so-called "beyond" the Standard Model. That theory might explain some of the facts the Model does not pretend to explain, such as the mass of the electron or the values of the interaction constants.

I do not expect anything like this for public policy and social science. I am not sure there could be a standard model of society and economy. The social scientists' attempts have been very smart, but too smart by a half. Physicists do not need to account for many features of the material world, as long as they focus on the main ones as they understand them. Social scientists seem to make what they take as progress by ignoring features that most people consider important. Perhaps they should not be so imitative of the physicists. The model of economic man or political man (or woman) are notably productive of theory and Nobel Prizes, but it's hard to find out how important are parts that they leave out.

Physics is not the model science or theory. But it does suggest that there is a need for more careful consideration by social scientists of what they leave out of consideration, and the sensitivity of their theorizing to those external factors. I would think social scientists and finance theorists would be excited to incorporate the experience of the Great Recession into their theorizing, what I imagine would cause of substantial revision of their ideas. Maybe their ideas incorporated the Great Recession already. Perhaps it is not possible. To defend your theories and blame others for doing something wrong is to find yourself condemned to irrelevance.

Sunday, July 6, 2014

On Hiatus Until the End of August

This blog will be on hiatus until the end of August--unless something sufficiently provocative comes my way. Most of these blog postings are the result of an event, usually the n+1-th of similar events, and so I am driven to discuss. For the moment it will have to be the 2n+1 -th before I think of posting, and that should be rare.

MK
6Jy14