So here’s the loose theory, and you’ve got to get this exactly right to understand this properly you got to get it exactly right, and it’s really important because In so far as you guys are interested in psychology, especially in the experimental end of psychology measurement is everything and so much of what psychologists Publish and write about is incorrect and the reason it’s incorrect is because they do not have their measurements properly instantiate it It’s a massive problem, especially in Social Psychology. In fact it’s probably a fatal problem in that most of the things that social psychologists measure don’t exist and Social psychology has been rife with scandals for the last four or five years and and there’s good reason for it But a big part of the problem is is that the measurement that… People are not stringent and careful enough about their measurements, so we’re gonna walk through this very very carefully So I’m gonna set forward a set of propositions, and you have to think about them because each of them are their axiomatic So you sort of have to accept them before you go on to the next step and there’s certainly room to question them But here’s the bare bones of the psychometric a model of personality, so we’ll call it Roughly the big five model and the reason it’s called the big five model is because the psychometric Investigations have indicated that you can specify human personality along five basic dimensions you might ask well What exactly is personality and well that’s partly what we’ve been trying to wrestle with the entire course so far and I would say Or what exactly is a trait? think of a trait as an element of personality And I think the best way to think about a trait is as a sub personality So you’re you’re made up of sub personalities that are integrated into something vaguely resembling a unity but the unity is is Diverse… there are.. there are… there are.. there are describable… Stable elements that characterize you that are elements of your being so for example here’s here’s some common ones I might say well, are you social or or would you rather be alone? So here here’s a good question for you to Define decide whether you’re extroverted or introverted It’s pretty straight forward because that’s that’s the first major dimension basically if you take any set of questions about Any any set of questions that could be applied descriptively to a human being? and you subject them to a Statistical process called factor analysis, you can determine how they group together So what I would be interested in let’s say I ask you a hundred questions Let’s say I asked a hundred questions of you and a hundred other people what I would find was that reliably if some person answered question ‘A’ say on a scale of one to seven Six or seven there would be other questions in the set of questions that they also tended to answer on the upper end of the scale or Reliably if they answered one question at the top of the scale, they’d answer another question at the bottom that’s a pattern of covariation so you’re looking for how the Questions covary across large numbers of people so let’s say here’s a stupid example, but it’s really it’s really straightforward easy to understand I might say How often do you smile one to seven? How often are you happy one to seven well what you’d find obviously is that? people who tended to answer that they smiled a lot would also tend to answer that they were happy a lot and so smiling and happy are not exactly the Same thing which is why we’d have two different phrases to describe them But they’re close enough so that they seem to be reflective of some underlying structure, and so that’s what a factor Analysis is does it allows you to take a large set of questions to administer it to a large number of people and then to Statistically analyze it looking how the questions relate to one another across the entire group so that you can infer What the underlying structure is Here’s the question in some sense if I ask you a hundred questions how many questions am I really asking you? Because you might say well, are you? Do you smile a lot? Are you happy? Do you wake up eager to start the day? you say well is that one question Asked three ways or is it three separate questions and the answer is well if the answer is Reliably co-vary then it’s reflection of an underlying single dimension now Obviously those questions are slightly different Now but they’re they’ll relate to one another stabily and so you can infer out the central Stable Factors now it might be the case so here’s here’s an example Because you might ask how many stable underlying dimensions are there in any set of questions if I ask you? questions that relate to your capacity to manipulate abstractions, I’ll find that there’s one factor, so imagine you had an infinite library of Problem-solving questions doesn’t matter what they are capital of, Georgia Here’s a sequence two four six eight ten what’s the next number? Here’s five patterns? Here’s and and they Transform Predictably across the Pattern array. Here’s five alternatives that the next pattern might be pick that one Here’s ten words. Tell me what they mean anything like that Here’s a mathematical operation compute it, anything like that imagine you had a very large library of questions like that Okay an infinite library and you took random sets of a hundred questions from that library and you gave that Those sets (of questions) to a thousand people What you’d find was that? People who and and the score say you gave them a hundred questions And then you summed across all the items to see how well they did what you’d find was that people who did very well on one set of items would do very well on another set of items and Very well on another set of items and
that would be the same for people who did badly if they did badly on any one of the sets of randomly chosen items of Abstraction they do badly on the rest. That’s basically IQ that’s all there is to it, so it What IQ does is correct that for age but other than that? That’s all there is to it and the thing that’s interesting about Those random sets of abstract problem solving questions. Is there’s one Dimension that’s it intelligence has one dimension And it’s one of the most terrifying statistics that are known to social scientists And IQ is a an extraordinarily powerful predictor of long-term success Especially in complex jobs and the reason for that, it’s quite straight forward Most complex jobs throw random sets of complex problems at you That’s what that’s that’s their definition So for example if you’re working as a lawyer on complex on complex court cases you have to be able to read very quickly You have to be able to abstract you have to be able to problem solve you have to be able to formulate arguments, and you have to do that? repeatedly in Different ways across very large spans of time and so the fact that your ability to solve any set of random Problems is a really good predictor of your ability to solve any set of random legal problems It’s more or less self-evident that that would be the case But the but the thing is the thing that makes IQ so damn powerful And it’s one of the personality traits Roughly speaking the thing that makes IQ so powerful is you can basically get a decent measure of it in 20 minutes It’s very terrifying anyways, we’ll go into IQ and some depths as we progress through the course But you get one dimension out of out of a factor analysis of IQ now in the personality domain We’re using descriptive items, you don’t you get five dimensions?