The Bell Curve

[_EAllusion]

touché. I may have fallen asleep in the movie. I usually do.

In any case, my point was that to his mother's credit, he didn't grow up stunted and malnourished. To his own credit, he was able to overcome all of his disadvantages and become an incredibly talented athlete, stay off of drugs, stay out of gangs, and stay out of jail.

To his credit, he pulled himself up by his own bootstraps and achieved an incredible amount of success as a teenager.

He doesn't seem like the kind of guy who accepts being the victim of his circumstances.

Your original point was that he lacked disadvantages of his circumstances, so that can't be used to explain relative lack of academic ability. After all, they got to go to the same school (not so much Agee, but let's ignore that) and had good families, so they are on the same footing.

The film covers extensively, you might even say is in large part about, their disadvantages in life especially as it relates to racial politics and poverty.

Now you've morphed your point into a ra-ra about picking yourself up by bootstraps. If you are one of the best high school athletes in the country, anyway.

[_Analytics]

Spearman's "g" is defined by factor analysis, which is about rotating clouds of multi-component data points around in their many-dimensional space to try to see which combinations of components may form simple patterns and which seem to show random scatter. It's not a dumb idea but it can't spin straw into gold.......

I got all that from a combination of my own background in physics and the nice pedagogical explanations by Steven J. Gould in The Mismeasure of Man, which was written largely in response to The Bell Curve. Perhaps Gould was biased but he did know his factor analysis, since he used it a lot in his own work, which got him to Harvard.......

The premise "IF "g" is actually real" becomes implicit without anyone noticing, until you're taking its reality as proven by all your ample research, and counting all that unresolved noise as just a few loose ends concerning the precise nature of "g".

From the perspective of statistics and predictive analytics, the analysis in The Bell Curve is quite dated. The way you describe Gould's work has the same basic issue. All of those models are based on the assumption that the underlying reality has a linear form with error terms that are normally distributed, independent of each other, and have identical distributions. In reality, few things if any actually conform to those assumptions.

The purpose of predictive analytics isn't to figure out what is "real." When it comes to the ultimate nature of reality, the human brain is ridiculously complex, and everybody has their own unique abilities, synapcit connections, life experiences, mental tricks to approaching complex problems, etc. Ultimately, there isn't a real, literal "g" in the human brain. There couldn't be.

What these statistical models are trying to do is come up with a model that is useful for either explaining the world around us or making predictions. That's all. Different models could be useful, insightful (or misleading) in different ways. But none of them are the literal underlying truth. As Covey would say, "The map is not the terrain."

[_Analytics]

Now you've morphed your point into a ra-ra about picking yourself up by bootstraps. If you are one of the best high school athletes in the country, anyway.

He had advantages and disadvantages in life. We all do.

Your last point here confuses me. It's as if you are suggesting that he was simply born one of the best high-school athletes in the country--like that was in his genes, and that being born into a challenging circumstance naturally couldn't repress that genetic attribute of his.

Murray says that studying for an IQ test can raise your score on the test, but it takes a lot of study to raise your score by just a little bit. The movie illustrates Murray's point on this.

[_EAllusion]

He had advantages and disadvantages in life. We all do.

That seems like a trite comment that diminishes how those advantages and disadvantages translate into outcomes that otherwise would be different if that mix was different. This seems like a dismissive "we all got problems" that completely overlooks how some problems more difficult to overcome than others and negatively affect people more than others.

Your last point here confuses me. It's as if you are suggesting that he was simply born one of the best high-school athletes in the country--like that was in his genes, and that being born into a challenging circumstance naturally couldn't repress that genetic attribute of his.

We are talking about William Gates now. He was the little brother of a local basketball legend. He almost certainly was genetically predisposed to being good at basketball, but beyond that, he lived a life that focused on grooming him for basketball talent starting at a young age that carried with him both in his private life and in his formal schooling. He also worked hard at being good at basketball until that trailed off due to injury and personal circumstances in his senior year. This is why he was positioned to be one of the top recruits in the country and thought of as a potential star.

This is not in any way analogous to his experience academically or the circumstances surrounding it. You can't one-to-one compare how life circumstances affect the ability to do trigonometry with how they affect field goal % even if he was equally coached in life at both, but there were obvious non-trivial differences in his schooling paths.

Murray says that studying for an IQ test can raise your score on the test, but it takes a lot of study to raise your score by just a little bit. The movie illustrates Murray's point on this.

Gates was a teenage parent who likely was experiencing depression during this time, living in one of the worst ghettos in the country, carrying with him all the academic history he had up to that point starting before he was ever even formally schooled. To dismiss this and say, "see, studying don't work" is obtuse.

[_Analytics]

Book Summary:

I arrived at the Darkening Forest several weeks ago with my tape measure, a pair of climbing boots, a calculator and a spade, and I have carefully studied a fair sample of the trees of this forest. Oh, how beautiful they are, every single one. I do hope we can find a solution for this growing tragedy, but I warn you, what I've discovered I'm afraid to say is quite shocking, and it's going to take real bravery to speak about these issues openly and fairly.

Critical to understanding my findings is familiarizing ourselves with the notion of HighQ....

Great read, thanks for taking the time to write this!

To get it a bit closer to Murray's point, I'd say something like this:

Every tree has a HighQ that can be evaluated when the tree is just a sapling. Common wisdom is that the eventual height of a tree depends upon where it is found in the forrest in terms of wind, terrain, water, etc., but otherwise all trees have the same potential. While these things do matter, the most important predictor of whether a tree will grow to be dominant is the HighQ that is measured when the tree is still a sapling.

Some people say that trees of the species Arboles Brunneis have distinct disadvantages in the Darkening Forrest because their ancestors were brought from another forrest and the shock from that and other enviornmental disadvantages keep reverberating across the centuries.

However, while some of those factors are real, it's also true that all Brunneis trees have a HighQ that can be measured when they are saplings, just like other trees. It's also true that individual Brunneis trees have a wide distribution of HighQ metrics, just as all trees do. It turns out that sapplings with a low HighQ have the same dismall prospects, regardless of whether or not they Brunneis, and saplings with a high HighQ are likely to dominate the forrest, regardless of whether or not they are Brunneis.

[_Analytics]

To dismiss this and say, "see, studying don't work" is obtuse.

I might be wrong about this, but the studies that Murray summarizes about how much work it takes to marginally improve your SAT score are probably based on a sample size greater than one.

[_EAllusion]

To dismiss this and say, "see, studying don't work" is obtuse.

I might be wrong about this, but the studies that Murray summarizes about how much work it takes to marginally improve your SAT score are probably based on a sample size greater than one.

You used an anecdote involving a person in a documentary to show that even when this he lacked disadvantages and had good schooling and tutors available to him, he still had a heck of a time doing better on the ACT as an illustration of Murray's arguments. That this was a completely inaccurate characterization of his circumstances, and missed central themes of the movie you were watching no less, seems relevant.

Then, looping back into the previous discussion, note what you are saying about schooling. He was trying to pass the ACT, not SAT as you said. The ACT involves a section on scientific reasoning. I'm not sure what the testing method looks like now, but I took the same exam a few years after Gates when it was similar. At that time, it involved primarily testing your ability to make inferences from presentations of scientific data. If you knew the topics the questions were on, that made it easier. If you could read graphs and charts, you were most of the way there regardless. I had a lot of science background by the time I took it and scored a perfect. I thought it was a piece of cake to the point of being a joke, actually. But it was both right in the sweet spot of what I'm good at and an easy-mode version of many classes I had up to that point. Maybe you did or would have done similarly.

Your position is that how students are at this kind of exam is prefigured by a young age and has little to do with variance in quality of science classes (among other things) they are exposed to. Maybe folks like you or me could benefit from top-end, boutique education, but for the unwashed masses lacking our golden brains, is it not a waste? Again, if that is the case, then why spend so much money on science classes that aren't getting the job done? Just adopt the practices of the cheaper science classes and people's natural ability will explain the variance in their science aptitude. Do you believe what you are arguing?

[_Analytics]

You used an anecdote involving a person in a documentary to show that even when this he lacked disadvantages and had good schooling and tutors available to him, he still had a heck of a time doing better on the ACT as an illustration of Murray's arguments. That this was a completely inaccurate characterization of his circumstances, and missed central themes of the movie you were watching no less, seems relevant.

Then, looping back into the previous discussion, note what you are saying about schooling. He was trying to pass the ACT, not SAT as you said...

Thank you for correcting me on that.

Clearly, there are any number of subtlties to the movie and to his life that I don't know and misremembered. I was wrong for to bring him into this conversation, and for that I appologize.

[_Analytics]

EA, your last post does raise a question. If you are an advisor to an elite athlete trying to get into college, passing either the ACT or the SAT will get your man through the door. So which test do you advise him to take? If your man is weak in science, why not avoid the whole issue by taking the SAT instead?

According to one source:

Scores on the SAT correlate very highly with scores on standardized tests of intelligence, and like IQ scores, are stable across time and not easily increased through training, coaching or practice. SAT preparation courses appear to work, but the gains are small — on average, no more than about 20 points per section.

If that is true, cramming for the ACT might be the better choice.

[_Gadianton]

Analytics,

thanks for reading the book summary from Murray's cousin. A serious error was made, and it was overlooked to note that there was only one kind of tree in the forest, since as I'd mentioned earlier, my issues with the book were long before racial considerations. It actually doesn't make a difference to my argument but it is an unnecessary complication that gets in the way of communicating my point. Your recasting with saplings is certainly fair, and I should have gone that route in the beginning; I actually did not quite intend it to read the way it did when I checked in this morning. So here is the updated version of that last post with alterations colored:

Analytics wrote:
only that low IQ is at rhe root.


Analytics wrote:
His fear going forward is that even if everyone's IQ significantly rises, society is only going to have a good place for those in the top x%.

To me these statements are incompatible, and may even get to the root of what I've been trying to say in many of my posts in regards to what bugs me about the book.

the analogy..

As I understand it, in forests, trees compete for sunlight. Whether the average tree is forty feet tall or two hundred feet tall, the trees at some standard deviation(s) below mean will get so little light that they are at greater risk for disease and death. Perhaps trees are predisposed for the height they have as some believe or as others believe, we could send gardeners to fertilize some of the smaller trees to give them a better chance, and perhaps they will grow taller, but as they do, they will crowd out trees that were nearly their peers in height that we didn't fertilize. If we fertilize all trees, it doesn't matter because it's the same problem but twenty feet higher in the air. Well, nature has its way of optimizing across millions of trees and thousands of forests, and perhaps generally, the loss of trees within the lower tail isn't too bad.

In this analogy we are interested in a particular forest called the Darkening Forest, and many more trees within this forest begin to die than we expect. Note there is only one kind of tree in this forest. We're worried, and so we send Murray's cousin to the scene who is a botanist. He brings all his gear and studies the trees and and then writes a book that he summarizes below.

Book Summary:

I arrived at the Darkening Forest several weeks ago with my knife, a microscope, a calculator and a spade, and I have carefully studied a fair sample of the trees of this forest. Oh, how beautiful they are, every single one. I do hope we can find a solution for this growing tragedy, but I warn you, what I've discovered I'm afraid to say is quite shocking, and it's going to take real bravery to speak about these issues openly and fairly.

Critical to understanding my findings is familiarizing ourselves with the notion of HighQ. Every tree has a HighQ, and this is discovered by, first, cutting into a branch and observing the cells and taking a good look at a certain innate property. With several trees measured in such a fashion, we at last, find the average measure and normalize that to the value 100. The average tree, then, has a HighQ of 100. Most trees find themselves within the vicinity of a HighQ of 100, which is depicted by the middle hump of a bell curve that represents a normal distribution. HighQ is important in that it represents a trees ability to access sunlight, which is needed for it to grow. A tree with a high HighQ is a tree we expect to drink gobs of sunlight and rise toward the clouds stout and true.

Now, here's where our tale gets shocking. As I measured these trees, I found that most of the trees within the Darkening Forest that are dying have low HighQs. How bad is it? Brace yourselves, because you may not wish to accept it. I measured dozens upon dozens of trees with HighQs a full standard deviation below the mean! I wouldn't have believed it myself if I didn't read the tick from the tape with my own eyes. I recall looking at one tree in particular that really struck a chord within me. I liked this tree and wished to save it. Could it be helped? Well, whatever our plans for that tree, just bear in mind that it's HighQ was only 80. Seriously folks, how can you save a tree with an 80 HighQ?

And so my book concludes that tragically, the root cause of the deaths of trees within the Darkening Forest is low HighQ. There are just too many trees below average, unfortunately; no wonder the whole things going to pot.