Just The Facts and Guessing

You are meant to play the ball as it lies, a fact that may help to touch on your own objective approach to life.
Grantland Rice

We ran out of Rupp Arena as fast as we could to get to the car and back on the road. We wanted to avoid the crush of traffic if we could. The Illini had lost to Kentucky in the last few seconds of the Elite Eight, the regional final game in the 1984 NCAA Tournament. Really, we were robbed. With a little less than 20 seconds left in the game and us down by a basket Hank Nichols, he of the stellar reputation, made a bad call. Bruce Douglas, our point guard and a great defensive player, had trapped Dickie Beals, who had picked up his dribble. Beals definitely dragged his foot, so the right call was traveling. Nichols, however, called a foul on Douglas and that was the ballgame. The following year the NCAA changed the rules to no longer allow any of the schools participating in the Tournament to play on their home court. This game was the reason why.

It was a bittersweet ending to a glorious few days. This was the first NCAA Tournament I attended. We played the second game on Thursday night. Our opponent was Maryland. Their star was Len Bias. In the first game, Kentucky played Louisville. Rupp Arena is an interesting place. There is an indoor mall immediately adjacent. You could enter the Arena from the mall without going back outside. Larry and I got to the La Quinta where we were staying probably around 2 PM and after dumping our stuff we headed over to the mall. Everyone was in a jovial mood in anticipation of the evening games. Somehow we connected with some diehard Kentucky fans and had a really good time with them. They speculated that Maryland would take us. Uh-uh.

We watched both games. Our seats were really high up, but it was a thrill just to be in the arena. After the first game, many of the fans cleared out. That was ok. This way the Illini fans could be heard. We were leading Maryland by eight or ten points in the second half when Efrem Winters came down wrong and sprained his ankle. The Illini held on to win that game. Afterward Larry and I went out for a Prime Rib dinner, relishing in the victory. Then some drinks back at the hotel room. We had an off day Friday so why not? One of us woke up kind of early, around 5 AM, and discovered that the game was being replayed on ESPN. So we got to watch it again, this time with a much better view. What a delight.

There was a lot of anticipation going into the game with Kentucky. We had played them in Champaign the prior December in a game that was odd because it was so cold and blustery outside that most of the fans stayed away and the regular refs couldn’t make it. So they got some High Schools refs to do the game. They were there to watch from the stands. They did a great job. We lost in the last minute or so. Kentucky had a lot more depth and eventually wore down Illinois. But we led most of the game. The best play happened in the first half when Efrem stuffed Mel Turpin, all six foot eleven inches of him, just as he was trying to dunk. Totally awesome.

In this rematch on their court Efrem was not at full speed and couldn’t cut well. So it is amazing that we stayed right with Kentucky through the entire game. But we did. Bruce Douglas was fantastic. With Turpin and seven foot one inch Sam Bowie, they were much taller than us. But our floor play was better. It should have been Illinois playing Georgetown in the Final Four. Not fair.

That year was the acme for NCAA College Basketball. Five years earlier, the Tournament featured Magic Johnson and Larry Bird, which brought it into prime time. The talent pool was a lot deeper in 1984 – Michael Jordan, Pat Ewing, Hakeem Olajuwon, Charles Barkley, and many others. And with that, the Illini were right there in the Tournament. Wow.

* * * * *

Mostly I’ve taught economic theory, so adopting a dispassionate tone goes with the territory. Economic theory is apt to put students to sleep, not get their juices flowing. So I really didn’t think much about the issue till a friend who teaches English and Film, Ramona, had me meet with her and a colleague to talk about some learning technology project. This was long ago and I can’t remember the nature of the project. But I do recall Ramona saying something like, “She really gets into it with the students, doesn’t she? When I teach I try to take a neutral stance rather than take sides.” The thought has been with me ever since. Students are apt to be intrinsically interested in edgy subjects. If as an instructor you deliberately bring those topics into the classroom to draw the students in, how should you play it, above the fray or one of the participants? Does the answer matter if specifically you are trying to teach the students how to guess?

Those who’ve been trained in the scientific method have this terrible fear about losing objectivity. (Though there is a sub discipline called experimental economics, in the main economics is not about the design of experiments but rather about looking at those data that present themselves from ordinary economic transactions, yet in all other respects it follows the scientific method.) The thought is that if a priori you are for one side of an argument, you will discount or totally ignore evidence that seemingly supports the other side of argument. That bias will have a consequence on the conclusions that are obtained, conclusions apt to be obviously wrong either on inspection or after further evidence has been accumulated. Scientists don’t want to be shown to have been obviously wrong especially when the record indicates they should have known better at the time. Some mistakes will be made in any event. Let’s try not to make those mistakes that could have been avoided.

Many of us will want to have it both ways. If we’re aware of the problems that stem from bias, can’t we adequately counter for it and still maintain our strongly held views? The question is ethical at its core. In mild circumstances it is reasonable to expect that we can offset the bias and look at all the evidence in an even fashion. When the situation is more extreme passion may win out over sound judgment. A physician who treats himself has a fool for a patient. (See especially the commentary, which suggests that the problem is fairly common.) How does one know when events will conspire to turn what at first appears under control into an extreme circumstance? This is, in effect, the argument for objectivity.

There are many who are not well versed in the scientific method, who nonetheless invoke the mantra that is the title of this chapter – just the facts. They too are aiming for objectivity though sometimes I fear they have an additional agenda, to close off further argument. Anecdotes are evidence. They may not be the best sort of evidence, especially when more systematic evidence is available, in which case relying on anecdotes exclusively is silly. But throwing them out is bias. When the systematic evidence points one way and the anecdote another, there is learning in carefully reconciling the two. Likewise, the expressed opinion of a friend, colleague, or opponent is evidence too. The vast majority of people are rational and thoughtful. When they express an opinion that appears contrary, they are apt to have access to information that you don’t have or to have related experiences that are unknown to you. Ignoring the opinion then is inconsistent with weighing all the evidence. Of course, we are awash in polemical argument in the political arena, where often the goal is to seek political advantage rather than to illuminate the truth. So there is a tendency to discount if not entirely ignore opinions of the other side. To the extent that politics is like sports and we voters are like fans, perhaps that’s ok. Outside of sports and politics, however, it’s a problem.

The best articulation of the principle I’ve seen is by Steven Sample in his book The Contrarian’s Guide to Leadership. The first chapter is on Thinking Gray, which means several things all at once. First, don’t make a decision before you have to and don’t tip your hand as to how the decision will eventually come out to encourage others to provide you with evidence that you will weigh fairly. Second, actively encourage argument and debate about the decision so different points of view can be well articulated. Third, while the first two are really external behaviors this one is truly internal to yourself. It’s not that you have a quickly formed opinion that you are not sharing because of the first two reasons. It’s that you maintain neutrality on the issues until when judgment is needed. You do this so you can make the best and therefore unbiased judgment when it’s time for that. As Sample says, this is contrary to the way most of us behave because we’ve been taught to make snap judgments.

F. Scott Fitzgerald once observed something similar to thinking gray when he observed that the test of a first-rate mind is the ability to hold two opposing thoughts at the same time while still retaining the ability to function.

Thinking gray is contrary to what I was taught in formal economic theory/statistics, Bayesian Decision Theory. This theory admits an element of subjectivity captured in the decision maker’s beliefs represented by the prior distribution over the unknown parameter. The theory then explains how beliefs get updated based on experience, generating a posterior distribution. That part is pure statistics. The economics comes in when experience is driven by choice, call it a consumption choice, and when different consumption choices have varying degrees of informativeness. For example, in consuming a drug, a low dose will have little effect simply because it is low, while a higher dose may have substantial effect if the drug actually works. So taking a high dose is more informative than taking a low dose. The economic theory prediction is that early on part of the drive of choice is experimental consumption, to encourage learning. Ultimately the choice settles down to what is optimal given beliefs. (In some cases beliefs reach the truth with certainty, but there can be instances where beliefs are stable with residual uncertainty.) This approach can rationalize the binge drinking of teenagers.

Really, the two approaches are distinct. Sample is contemplating a large decision that once made remains fixed for quite a while. The theory of experimental consumption focuses on repeated decisions of a smaller nature. The information gathering that Sample has in mind is also different from the statistical approach in Bayesian Theory. One metaphor that might help in understanding the Sample view is to imagine having to understand a three dimensional object from getting to view a finite number of two dimensional snapshots of the object, each taken from a different perspective. Another snapshot from essentially the same perspective doesn’t really help. One from a new perspective helps a lot. Sample doesn’t argue that we get to choose the perspectives from which we get to take the snapshots. He just argues that we have a better understanding with more perspectives.

Much as I like Sample, however, he is an engineer by training and he leaves you with the impression that after all the information is in the situation and high intelligence he brings applied to the situation more or less dictates the solution he comes up with. Mostly, I don’t think it works that way. Prior disposition and point of view matter for these decisions. Consider this episode from the West Wing called The Supremes, with Glenn Close as Judge Evelyn Baker Lang (very left of center) and William Fichtner as Judge Christopher Mulready (just as far right of center). Mulready exemplifies the F. Scott Fitzgerald conception of a first-rate mind; he is able to articulate the Liberal view better than the staffers at the White House while he comes at his opinions from an opposing vantage. We care about the politics of our Supreme Court Justices because in the way they decide on cases their politics matters. In the context of judicial opinion, that is an unremarkable assertion. In broader contexts prior disposition plays the role politics plays in the judicial case, hence there is an inherent subjectivity to the decisions. Sample conveys the idea of an optimal (and unique) solution to his decisions as the afterward of thinking gray. Optimal is the engineer’s credo. Though as an economist I was trained to think that way as well, my experience as an administrator suggests there are multiple possible approaches, none a priori optimal, with preference over a particular alternative determined by prior disposition. So I’m inherently subjective in my approach and my interest is in understanding the interplay of that subjectivity with the facts.

* * * * *

My first semester at Illinois I was an instructor rather than an assistant professor because I hadn’t yet completed my dissertation and wouldn’t do so till the following spring. I taught one class that term, intermediate microeconomics. I had been a popular TA at Northwestern, but this course was a bomb. I took what proved to be a “blow them out of the water” approach, partly as a result of my graduate training, partly because I hadn’t taken the equivalent course as an undergraduate so didn’t have a sense of what should be in it, but mostly because of my own insecurities. How else would I establish my authority than by teaching a tough and demanding course? I didn’t concern myself with what the kids could grasp or not. How could I know that? It turns out I frustrated a lot of the students that way. They had their revenge; they gave me awful teaching evaluations.

On the final exam I wrote an essay question based on Akerlof’s Market for Lemons. One kid in the class aced the question. The rest flubbed it. I gave that kid an “A” for a course grade. He had a solid “C” going into the final. My reasoning was that this kid had shown a flash of insight into an incredibly important problem in economics. Insight needs to be rewarded. I believe that still. Another student who knew this kid was offended by my approach. She had a high “B” going in and that’s exactly how she performed on the final. She got the basic stuff right, but on deeper perception, nada.

Nowadays grading according to rubric is quite popular wherein there are various criteria that count for the grade and for each criterion there are performance standards – what counts for poor performance, what counts for good performance, and what counts for superior performance. Finally, there is a way to aggregate performance rankings across criteria. Each ranking gets a certain number of points and those points are added up. I maintain some disdain for grading rubrics (I do understand that it helps to reduce arbitrariness in assessing student work) because it rewards and therefore encourages consistent mediocrity. True insight is difficult to put into the rubric because it is so rare. In a country addicted to Lake Wobegon, reserving the superior performance standard for true insight means most of the students will not be happy campers. Engineering colleges have a reputation of grading this way. The rest of us have grade inflation.

In his book, Blink, Malcolm Gladwell talks about thin slicing information and the power in doing that. In grading that final exam, I was thin slicing the student performance. Thin slicing is quite different from understanding the total view of an object. When you thin slice you look for a particular aspect and throw away a lot of other information, because that other stuff is of secondary importance. You pay attention only to the really informative stuff. If you know what to look for ahead of time, you sit in waiting ready to pounce when it shows up. If you don’t know what to look for ahead of time, but you thin slice nonetheless, you’ve got to search for it.

As the essay by Akerlof previously cited indicates, economics is a peculiar discipline in that it provides insight by isolating on particular variables of economic interest and ignoring much else, sometimes of substantial social importance. My second year in graduate school I had the good fortune to take a course from Dale Mortensen on Micro Foundations of Macroeconomics. In that course we spent a good deal of time learning about search theory, the economic variety. In the simplest possible model the unemployed worker is looking for a job characterized solely by the wage it offers. Time is scarce. Each period the worker gets a random draw from a time invariant wage distribution. The worker has to decide whether to keep the job or, like the fisherman, throw it back into the pool to hope for a better catch in the future. The optimal strategy is characterized by a reservation wage. The worker takes the first job with an offer above the reservation wage and throws back all other offers before the one that is accepted. The theory is the basis for what has come to be known as “frictional unemployment.”

Real world search is unlike how it is modeled in economics. The idea that time is scarce certainly is correct but actual search is not a random draw from a distribution. Much of it is driven by…….guessing. Smart guessing is about generating searches that are apt to come to a satisfactory solution quickly. Actual search is also not about thinking gray, at least on the point of whether the searcher can use whatever the current search has returned to address the issue at hand. That judgment is made immediately. One doesn’t wait to perform other searches and make a comparison across the set of them. Because time is scarce if the results of the current search will do the trick, that’s all she wrote.

I have never seen a discussion of information literacy, a topic of increasing importance in considering how we go about teaching our students, that distinguishes finding information for a thin slice decision from finding information for a thinking-gray seeing-the-whole-picture exercise. Google is incredibly useful for the first type of search and one of the key skills for good guessing in this instance is to know what to type in the Google text box to initiate the search.

A week or two ago we had a problem with our home computer. It was giving an error message on accessing both my College of Business email via a browser interface and in going to Google supported sites, such as Blogger. I had no idea what was wrong. I feared that there was a virus on the computer. I had read of some new virus from one of the blogs I follow. Getting a little distressed by the whole thing but with some hidden hope that it might be something else than a virus, I thought to take the text from the error message that Google produced and do a verbatim search on that. I found the source of the problem immediately. My clock had gotten out of wack. It was running 15 minutes late and the year said 2004 rather than 2009. Why the clock settings got changed is a mystery I can afford not to solve. Adjusting the settings back to the correct time fixed the problem we were having and that was enough.

One tries to reduce a seeing the whole picture kind of search into at most a few thin slicing type of exercises done first that allow you to “do your homework,” gather information to address a particular issue or question. When I was writing papers that I hoped would get published in leading economics journals, I would first find a topic that piqued my interest, then I’d look for a related paper on that subject and simultaneously look to see if there is a seminal paper in the subfield of study. If I could identify that seminal paper and see how other derivative papers departed from that original, I’d then have a fair understanding about how what I wanted to write about would fit in. Which to read first, the seminal paper or the derivative one might be determined by other factors – did I know the name of the author, did one seem easier to get through, which I would stumble upon first, etc. I would still spend a lot of time reading other related papers, but with this approach I’d have a better basis to understand the contributions of what I was reading. More or less coincidentally, I’d also be working on understanding the model I’d be building. My model would offer a separate pathway into the issues, giving me a different center of gravity to determine what was important and separate that from the chaff, and it also allowed me to indulge in something I enjoyed and felt comfortable doing. So I’d advocate for taking at least two and possibly more pathways into the subject to see the whole picture. Preferably those pathways are independent. That they converge in perspective later on should not be a driver initially. Convergence needs to be a conclusion, not a supposition.

I do something similar for my job where I constantly need to get perspective on what is going on. I will talk with different people about the same issue so as to “triangulate” the information I’m getting. I believe this is indispensible for understanding. Sometimes that is simply confirming what I already know. Other times, there is new information. And still other times it is the same information but with a different spin. All are useful.

A hard question for which the answer has to be subjectively determined is whether to keep on gathering information or stop and focus on something else. We really don’t help students learn to do this well at all, with our requirements about having so and so many sources from such and such places in their term papers. A sense of taste plays a very important role here and we teachers should be spending a lot of time cultivating students in their sense of taste on how to answer that sort of question.

If I’m gathering information to write something, I know in advance I value subtlety and nuance and I also value making some sort of contribution, a reframing of a familiar argument, connecting ideas together that have previously not been associated, or bringing in a personal experience to show how I’ve come to think about the issues. If I have enough information to do that, I can stop searching. In my earlier chapter on Writing as Guessing, I went into some detail on the point that the search might resume in the process of writing because a new twist occurs to the writer and then there is an on the spot investigation of that. How long that takes is determined by the same principle.

The information literacy issue is frequently framed about how much trust the reader should place on the source. I have always found this somewhat strange, particularly for doing economics, especially at the undergraduate level, because there are other factors that have to be brought to bear to make the source useful in advancing the student’s thinking. There is first the question of whether the student has enough background and perspective to read the paper in question. Often that background is absent. The paper might be brilliant but it can nonetheless be inappropriate for the students; the paper is opaque to them. The second issue is whether the students have the technical wherewithal to slug through the paper, where again opaqueness is the concern.

The issue isn’t just for students. It applies to professional economists as well. As I’ve pointed out in my blog a few years ago, there was a paper by Roy Radner that preceded Akerlof’s Lemons paper by a couple of years that correctly modeled asymmetric information and with much greater generality than what Akerlof produced. But nobody understood the implications from Radner’s paper in practical terms. So Radner’s paper became a theoretical curiosity, nothing more. Akerlof’s paper communicates the ideas with less generality but nonetheless more powerfully; because the model is so simple everyone can understand it.

True as that is, the search for powerful but easy to understand information can be a path to the slippery slope. Wrong ideas can seem plausible, especially to the novice who doesn’t check through them carefully for their correctness. For example, due to the NCAA tracking the academic performance of student athletes, there is now publicly available information about graduation rates at all NCAA member schools. Some have inferred that those graduation rates can be taken as a measure of school performance, just as they have taken that patient success rates are a measure of hospital performance. But such measures in and of themselves don’t account for how talented the students are or how sick the patients are. More talented students have a greater likelihood of graduating, irrespective of the college they attend. Likewise, sicker patients are more likely to fair poorly, even when they receive excellent treatment from the hospital. The raw measures may be of some interest in themselves but using them to judge institutional performance conflates inputs with outputs.

It’s not just latching onto wrong ideas. Correct ideas can nonetheless be of minor consequence. It is certainly true that quality of life in Champaign-Urbana would be better if we had jet plane service to several major metropolitan areas in the U.S. But the likelihood of getting such service is nil so trying to base our approach to retain highly regarded cosmopolitan faculty members is foolhardy indeed. Finding a retention strategy that will be both successful and do-able is a hard problem. That sense of taste comes in identifying the nexus between what will work and what is actually possible to achieve.

If we think of information literacy as dealing both with accessibility of the source and its trust worthiness, that is closer to the mark in my view. But then we need to understand that these two dimensions can compete with one another. Taste serves to resolve the tension.

I fear that the Republican Anti-Intellectualism has had an enormous consequence entirely outside the political arena; it has a pernicious impact on learning by its appeal to take ideas on faith rather than to work them through based on evidence and reason. Contrary to what David Brooks argued in his belated acknowledgement of the problem, the Reagan years were the beginning of alienating youth from reasoned argument. The embodiment of the problem was the CNN show Crossfire. The critique offered by Jon Stewart when he appeared on the show (soon thereafter the show was cancelled) demonstrates that many others are aware of the problem insofar as it has affected the general discourse. The ethnographic work of Carol Trosset convincingly illustrates the problem with college students. Argument is not about learning; it is about cajoling the uninitiated. Most of us don’t want to listen to a sales spiel that we ourselves have not encouraged. Nor do we want to be bullied into our ideas. We want to retain the power to make our own judgments.

There is a history of great debate in this country, none more famous than the Lincoln-Douglas debates of 1858. Information traveled slower then. Vitriol and venom may be effective in their immediacy, but the impact does not endure well at all. The Lincoln-Douglas debates were high minded and there was a great deal of respect between the adversaries. That, in large part, is why their reputation has endured. When I was growing up in New York City, there were both the Times and the Daily News. They could peacefully coexist, with essentially distinct readerships. Cable television has encouraged a kind of Gresham’s Law of argument. Now there is only polemic and posturing. It is theater of the absurd. Our youth need to see an alternative, but there is none to be provided.

* * * * *

There is another reason to embrace objectivity in considering issues and to avoid overt demonstrations of subjectivity in analyzing information, namely self-protection. If optimism flows from good news and pessimism from the reverse then to stay on an even keel one needs to treat the news as neither good nor bad, just more information to process. Objectivity is then associated with discipline and detachment, a way to stay on an even keel, while subjectivity is identified with self-indulgence and self-absorption, producing a path full of mood swings. One wonders whether it is possible to exercise a strong point of view and yet maintain an even disposition. To me, this is the search for the Holy Grail. I’ve not yet found how to do it.

What I have to offer instead is to recognize the importance of looking inward on occasion. Sometimes there will be anger, or dullness, or alienation. Other times there will be joy, or great creativity, or intense engagement. Even if it is a bit of a roller coaster, hold on for the ride.