Making Changes MenuMind & Consciousness MenuTalk Therapy MenuEducation & Learning MenuHealthy Relationships MenuAutism Spectrum MenuAddictions, Risk, and Recovery MenuWeight & Fitness MenuHuman Personality MenuScientific Method Menu

the Myth of Grades and Grading - "Statistical Error"

On Education and Learning



Education and Learning 1Education and Learning Week 2Education and Learning Week 3Education and Learning Week 4Education and Learning Week 5Education and Learning Week 6Education and Learning Week 7Education and Learning Week 8Education and Learning Week 9Education and Learning Week 10
Learning and Education Series - Week 11Learning and Education Series - Week 12Learning and Education Series - Week 13Learning and Education Series - Week 14Learning and Education Series - Week 15Learning and Education Series - Week 16Learning and Education Series - Week 17Learning and Education Series - Week 18Learning and Education Series - Week 19Learning and Education Series - Week 20
education and learning week 21education and learning Week 22education and learning week 23education and learning week 24education and learning week 25education and learning week 26education and learning week 27education and learning week 28education and learning week 29education and learning week 30

Have you even received an unfair grade? Have you ever noticed how no one asks people for their grades out in the real world? Why not? This week's column on Education and Learning explores The Myth of Grades and Grading; what makes grades useless, and why we believe them anyway.

The Four States of Learning

Chapter Six

the Myth of Grades - Statistical Error

This week, I'm hoping to find the words to discuss a topic I've been wanting to write about for some time now; "What Makes Statistics A Lie."  Or stated in more vernacular terms, "Why Statistics are bulls~*t." So are all statistics, "bulls~*t?" Yes, all statistics are bulls~*t. Why? Because they base their validity on a false assumption; that describing an average moment will describe all similar moments. At least those claiming to represent human behavior. And certainly, those claiming to test for the human behavior we call, "learning." Are you beginning to see why this topic interests me? Making the world better for children as learners is my thing. Remember?

OK. So I'm not exactly feeling neutral right now. I admit it. And I don't exactly place a high value on statistics in general. I admit this too. Before I launch into my tirade though, allow me to present what I see as the basic nature of statistics as related to human beings; the three directions if it were; [1] case studies, [2] correlational studies, and [3] experimental studies. Know my choice to divide statistics this way is not my work, albeit, I agree with it. In fact, this choice of division, at least the titles, comes from the work of personality theorist, Dr. Lawrence Pervin, whom I had hoped would at least read and comment on my work. He didn't. And he said it was because I had no "statistics." No "research foundation," to use his exact words.

No "Research"

So is this true? Do I have no "research foundation?" No. This is not true. Not unless you consider having some several thousand longitudinal studies of the fractal nature of personality, "no research."

What he is right about though is that I have no statistics. I, in fact, choose to do this. Why? Because I see statistics as a wholly spurious endeavor, this despite the ridicule supposed scientifically minded people heap on me for my not using statistics. Hopefully, by the end of this column, I'll have sufficiently explained why I choose not to use statistics. And why I choose to use something else with which to test my work; fractal assessments. Or more simply stated, pattern based, visual repetition tests for personal validity. More on what these tests are in a moment.

As I was saying, then, there are three forms of "statistical research." Realize, the word "research" is, in fact, just a fancy way to refer to how we "learn the truth about things." Here I go with that word again; learning. Anyway, there are three ways in which we try to learn the truth about things: [1] case studies, [2] correlational studies, and [3] experimental studies. Each, according to Dr. Pervin, has it's own penchant.

According to me, the first type, cases studies, explore things in depth, one case at a time for significant amounts of time. Because these studies happen over significant amounts of time, they end up being more personal than the other two methods. Sadly, this "more personal" part is what makes most scientists blow off case studies as invalid. Too hard to control the variables. Too many things to consider. Anyway, this kind of research is the "micro level" of learning the truth about things. It is also the best view point for a therapy to take. And therapies like Freudian and Jungian evolved most of their ideas from this type of questioning.

Then there is the second method; correlational studies. Correlational studies look for differences in similar groups of things. Or in similar situations. Or in similar groups of people. Thus, people use these kinds of studies to explore stuff like the educational level of all the black males between age twenty and twenty five, or the relapse rate of all the recovering alcoholics with over five years of recovery. Obviously, this research is quite a bit less personal. Which, perhaps, is why more scientific folks believe in it. No human vagueness to muck up their numbers. Anyway, this is what I call, "the macro level" of figuring out the truth about things. It is also the best level of detail for a theory of personality to take. And it is also the most used leverage to get money out of folks; scary numbers make for easy pickings.

Finally, there is the third exploratory view point; experimental studies. Experimental studies look to learn how altering small details will effect change in everyone, and in doing so, reveal the nature beneath everything. Heads-with-feet types tend to trust this kind of stuff very, very much. Way too much, in fact. No surprise, research at this level is very impersonal. Oddly though, many theories of personality take this approach. Theories like trait theories and such. Imagine? Theories of personality which contain not one thing which is personal? Anyway, this is the meta level of research. It is also the best view point for a theory of the mind to take. And it is also the most used method of placing people in everything from school classrooms to corporate managerial advancements.

A Brief Look at the Three Styles of Research

OK. So now let's look at how these three styles of research deliver mostly bulls~*t, at least statistically. Let's start with the experimental approach. And instead of simply finding fault with other peoples' work, which I don't want to do, I want you to go out and do a bit of your own experimental research. Something interesting. Like asking everyone at work if they like you. Or asking all your relatives if they think anal sex is good.

Not too keen on the idea? I don't blame you. Which is why we tend to accept so much of researchers' bulls~*t as true. Why? Because most of us poor slubs do not want to expose our personal lives in order to try to learn these kinds of things for ourselves. Some people even shame other folks for trying to learn these kinds of things for themselves. These morons tell us things like, "Why reinvent the wheel?"

"Why reinvent the wheel? the moron asks. To become a wheel maker, that's why. Which is how I came to write a theory of the mind, a theory of personality, and a therapy. As well as how I've come to have twenty tons of shaming words thrown in my direction for having my own ideas. Ah, well. What doesn't kill you makes you stronger. Isn't that what "they" say?

Let me cut to the chase. You don't want to hear these explanations anymore than I want to write them. And my whole point here is that statistics lie. Why? Because they assume linearity in non linear things. Things like peoples' motivations and the movements of clouds. Do you believe the weather man? This is statistics at its finest. Can't even predict the weather marginally well for more than a few days in advance without using the words, "partly," and "chance of." Why can't they? Because the weather is never going to be even remotely linear. Nor is the stock market. Nor is human behavior. So how can we rely so much on these totally fabricated numbers?

The truth? Because we so fear the future that we pretend we can predict it. Read Heraclitus. He's an old Greek dude who was famous for saying, "You can't walk into the same river twice." And he is still right, more than two and a half millennia later.

So now to the point. If we cannot statistically predict that a child will be prepared for life based on passing grades, then why do we keep lying to ourselves that passing grades mean this? And if I am right, then what could we be using that is better?

The answer? Pattern based, visual repetition tests for personal validity. In other words, tests based on fractal measurements of learning rather than measures of parroting skills. What would these tests be like?

Well, say today you have to test a kid on his knowledge of Pythagoras's Theorem. Rather than having the kid figure out the length of a side of some stupid triangle, what about if you had the kid tell you what amazes him about Pythagoras. No amazement. No passing grade. Not for the student, nor for the teacher.

How about if you were testing a kid in English class on Gibran's, "The Prophet." How about if you were to ask him, "What does Almustafa look like?" Almustafa? Yes, "Almustafa." After all, Al is the prophet. Here again, both the teacher and the student would have to pass the exam. No amazement. No passing grade.

How about Social Studies. How about if you had to test a kid on what the Middle Ages was like. So you had her write a story from the point of view of the only thirteen year old left in her village after the plague? Amazing? God awfully amazing. In one year, some say thirty percent of Europe died. Of course, like all statistics, people fight over who is the king or queen of the true numbers. Which means, some say twenty percent, and some say forty. Statistics. Arrgh!

Regardless though, imagine a passing grade coming only after the teacher and student had both written a story from the point of view of the only teenage survivor in a village? Imagine how that would prepare a kid for life?

Now the technical part. So what is fractal about what I'm saying? What is fractal is that in each case, the statistics mean nothing. Nor do they get in the way of the kid learning. Life is the learning. And learning is fractal. All of it.

Can You Use Fractals to Predict Things?

So can you use fractals to predict things? Actually, yes you can. Given you know the right fractal, that is. For instance, given a knowledge of the right fractal, you can predict how each kid will react to each teacher, at least enough to match them to each other reasonably well. No statistics. Just fractals. Measurable repeatable patterns in human nature.

How about healing? Like to know what your injury is? Well, there is a fractal for injury which resides in all our brains, a visual pattern easily recognized by anyone with a modicum of training. Useful? Amazingly. Predicable. Very much so. So much so in fact that were there to be a statistic for how often this fractal works, the number would be uselessly high. No one, in fact, would believe it. They would simply dismiss it as bulls~*t. Which is, after all, what everything is if it does not work.

Why have I been so intense this week?

These are not cold hard statistics we're talking about. These are our children.

When are we going to wake up.

Closing Comments

OK. So there. I've done it now; I've risked offending some of you and made coffee go up the noses of others. I warned you about this in last week's column though.

So how did I do? Was I at least a bit more honest in how I've written this? I hope so. Too honest? I hope not.

In truth though, too often, I feel like I tone down much of what I'd like to say. I also find myself wishing, after the fact, that I had been more open. More truthful. And more to the point. You see, I, too, fear being shamed and ridiculed. So I, too, hide my beliefs beneath levels of b.s. at times.

I'm trying to be better about it. I really am.

Until next week then. I hope you're all well,

Steven


Education and Learning 1Education and Learning Week 2Education and Learning Week 3Education and Learning Week 4Education and Learning Week 5Education and Learning Week 6Education and Learning Week 7Education and Learning Week 8Education and Learning Week 9Education and Learning Week 10
Learning and Education Series - Week 11Learning and Education Series - Week 12Learning and Education Series - Week 13Learning and Education Series - Week 14Learning and Education Series - Week 15Learning and Education Series - Week 16Learning and Education Series - Week 17Learning and Education Series - Week 18Learning and Education Series - Week 19Learning and Education Series - Week 20
education and learning week 21education and learning Week 22education and learning week 23education and learning week 24education and learning week 25education and learning week 26education and learning week 27education and learning week 28education and learning week 29education and learning week 30


.