Thinking fast and slow 17. Regression to the mean

Following our reading calendar

It is almost as if Kahneman had read Carlos comments on the former chapter and what do we have here? A story!

After my first reading of this book, one of the things I do remember more deeply is the story that opens this chapter and brought Kahneman to propose his “regression to the mean” bias.

So the story goes, while presenting some of his ideas to flight instructors one of them challenged him saying that rewarding good behavior was a bad strategy.

“On many ocasions I have praised flight cadets for clean execution of some acrobatic maneuver. The next time they try the same maneuver they usually do worse. On the other hand, I have often screamed into a cadet’s earphone for bad execution, and in general he does better […] So please don’t tell us that reward works and punishment does not, because the opposite is the case”

There is another explanation of course. The flight instructor was just observing a regression to the mean. Performances are not steady, they fluctuate, so after some good maneuver is not strange to expect a less than perfect one, and viceversa.

They way Kahneman teaches their mistake to the instructors is also cool. He made them throw two coins without looking and see if they can hit a target. Those that did it better the first time tend to perform less well in the second try, and viceversa

After the not so convincing biases of the former two chapters, here we return to something that makes sense. Luck, randomness is a very relevant force in our daily lives, but we are not designed to accept randomness, we want causal explanations for everything. So if an cadet performs poorly with his plane last time and now he is doing better, it is because the instructor yelled to him, not because there was a regression to the mean. Very Talebian again.

With a little bit technical analysis, as he explains Francis Galton’s discovery of regression to the mean, and the main teaching of this chapter”Whenever the correlation between two scores is imperfect, there will be regression to the mean”

One of the examples he uses to explain is not ok, though. He says:

“Highly intelligent women tend to marry men who are less intelligent than they are” and invites readers to find an explanation.

If you use the word “tend” most English speakers would understand it as “there is a connection” no matter whether in Statisticianish -thank you Carlos for the word- it means “there is a correlation between two magnitudes but not necessarily a causal link”. And if you ask for an “explanation” then people will look for a “causal explanation”.  And no, for most people in Planet Earth “The correlation between the intelligence scores of spouses is less than perfect” is not an explanation.

However, in this case, this example doesn’t jeopardize the main conclusions. When judging regression to the mean we find it very difficult, and we should keep this possibility in mind all times instead of trying to find a causal explanation every time there is a flux in a distribution.

I think the key is that this time Kahneman is not revising how we use terms like “probable” but he is talking about how things, animals and people behave in real life. The world is a lot more random than we think it is.

 

 

Advertisements

Thinking, Fast and Slow. 16. Causes trump Statistics

Following the schedule of our Reading Calendar

The lessons from this chapter:

1. We are lousy Bayesian Infererers. We just forget once and again about prior probabilities. I have to admit that I gave the wrong answer to the question opening the chapter. So, lets admit it, if I make such an error while reading this book, there is not human way to avoid making such errors in normal life.

2. We are much more likely to incorporate prior knowledge into our inferences if it takes the form of causative rules. As we have commented before, we are causality searching machines. And when we cannot find it we create it.

3. System one cannot work with ambiguous traits with assigned probabilities. He needs to create stereotypes with very defined characteristics and is going to base its judgments on that defined traits.

4. The most important conclusion of the chapter is something about what Nassim Taleb has told us a lot of times: the only way of entering our minds is through stories where particular things happen to particular people. Our mind is unaffected by abstractions, conceptualizations, means, medians and briefings. We need to learn the story. And that is the only way to change our minds.

And two comments from my part:

1. This book seems to be giving us ideas of how to manipulate people and how to avoid being manipulated. The problem is that until know I am finding a lot of the first but I don’t see any real proposal that has the appearance of being effective in achieving the latter. The feeling that one gets when reading the book is of helplessness.

2. The book is quite boring. Kahneman uses abstraction to teach us that are stories and not abstraction what get into our minds. Taleb uses stories to tell us that stories is what matter. And if at the end it is going to happen that no knowledge can be trusted, lest have at least a bit of fun.

Thinking fast and slow chapter 15. Linda: less is more

Following the schedule of our reading calendar

The first time I read about Linda experiment I felt like Carlos when evaluating Tom W. problem. It was like: Come on! Everybody knows that being a bank teller is more probable than being a bank teller AND blah, blah, blah. People participating in the experiment have misunderstood “more probable” with “more plausible” or something alike. “Probable” is not a well defined word and people tend to guess what the problem is really about by taking clues on what information is given.

One has to be very critical with experiments like those. Piaget the psychologist thought that children under the age of four didn’t possess the concept of number, and had this elegant experiment to prove it. He showed four years old kids two set of marbles, distributed like this:

1) o o o o o o

2) o    o      o         o
When asked in which row there were more marbles most children choose the second row. So, according to Piaget, four year old children confuse quantity with occupying space. The more space you cover the more elements you have

Decades later, other psychologists replicated the experiment but with an interesting twist. Instead of marbles they used sweets. So when asked in which row there were more sweets, they still choose row 2, but, when invited to choose one row and get all the sweets in that row, guess what? They all choose row 1.

So the kids did know the concept of number. They just didn’t understood the question.

Nevertheless after reareading this chapter I found the experiment on people being older than 55 and having heart attacks, and the numbers of errors decreased highly if instead of talking about percentages, one was asked about 100 people, I’m not sure what to say. If such a small change transforms results so greatly, maybe this is system one getting in the way of system 2.

Thinking, Fast and Slow. 14. Tom W’s Specialty

Following the schedule of our Reading Calendar

The idea behind this chapter is really simple. It explains an experiment and two biases that can be inferred by the way subjects behave in the experiment.

When we humans are trying to assign a possibility of something (the guy that lives across the corridor) being part of an abstract class (a law student) we forget about the a priory probabilities of that abstract class (how frequent are law students) and take into account mostly the resemblance of our something (the neighbour) to the stereotype of our abstract class (how law students look like). That means that we are not processing correctly Bayesian statistics. And that we humans are lousy innate statisticians is something that has been talked about extensively along the book.

The second bias that the experiment illustrates is the fact that we do not take into account the level of reliability of information when using it.

I am beginning to feel uneasy about the book. This is just an intuition, but I don’t feel happy with deducing that much from this kind of experiments. They are so far away from real life. I feel (and again only feel) that a more robust and empirical approach can be provided by marketing research. There must be (I don’t know the field) dozens of books talking about price setting and how to take advantage of anchor effects.