Why Intelligence Matters
Many socioeconomic outcomes correlate with intelligence. To understand a social phenomena, you need to consider that intelligence could be playing an important role. Frequently, it is.
The purpose of the essay is to make the case that intelligence is extremely important to the social sciences and understanding the world properly. In order to truly understand discussions of intelligence, it is necessary to have some familiarity with some statistical concepts. I will provide intuitive explanations, rather than mathematical.
First consider the concept of correlation. To understand correlation, imagine the relationship between height and weight. We know that taller people tend to weigh more. This means there is a positive correlation between height and weight. This does not mean that if a person is taller they always weigh more, only that they tend to. Now, imagine I precisely measure a group of people’s height in centimeters and in inches. Everyone who is taller in centimeters will be taller in inches. The correlation between these two height variables will be 1.
The variability of observed correlated variables can partially be explained by unobserved variables called factors. If we give someone a bunch of tasks that rely on thinking and problem solving, we will find that someone who is good at one is usually better at others. Nature did not have to be this way. It is conceivable that if someone was good at doing mathematics problems that they would be bad at verbal logic. However, it is the case that scores on cognitively demanding tests positively correlate. All mental tests specifically tap into a factor called the g factor to some degree and will correlate positively with it. Such tasks are said to be g loaded. The g factor (sometimes just called “g”), general mental ability, is a psychometric construct that underlies all intellectual activity.
The test that we use to measure g is an IQ test. IQ stands for “intelligence quotient.” There are different kinds of IQ tests but they all attempt to measure g. The IQ test has a normal distribution centered about 100 and given a standard deviation of 15 usually. A normal distribution is a bell shaped distribution and a standard deviation is a measure of the distance from the middle. Many human traits fall on bell curves. In fact, bell curves are everywhere throughout the natural world. For IQ, most people are between about 85 and 115. The vast majority are between 70 and 130. Less than 2% of the population is above 130.
For some people, intelligence research is of particular interest. There are associated controversies surrounding intelligence testing that I will not delve into. It will suffice to say that we can think cognitive ability is an interesting topic on the individual level. Arthur Jensen, one of the most influential researchers in intelligence, wrote an encyclopedic discussion cognitive ability in his book The g Factor. He explains why g is so important:
My research has led me to regard the g factor in a much broader perspective than I had envisaged at the outset. I have come to view g as one of the most central phenomena in all of behavioral science, with broad explanatory powers at least as important for understanding human affairs as E. L. Thorndike’s Law of Effect (or Skinner’s reinforcement principle). Moreover, it became apparent that the g construct extends well beyond its psychometric origin and definition. The g factor is actually a biologically based variable, which, like other biological functions in the human species, is necessarily a product of the evolutionary process. The human condition in all of its aspects cannot be adequately described or understood in a scientific sense without taking into account the powerful explanatory role of the g factor. Students in all branches of the behavioral and social sciences, as well as students of human biology and evolution, need to grasp the essential psychometric meaning of g, its basis in genetics and brain physiology, and its broad social significance.1
Everything that is cognitively demanding is tapping into the g factor. So understanding the g factor is important for understanding things that are cognitively demanding or could rely on cognitive ability to some extent. Think of tasks in our modern society that are reliant on reasoning, thinking, contemplating, problem solving and so forth. They are endless. Here is a list of correlations between intelligence and measures of success (r is the correlation):
Intelligence does not totally explain the above measures of success. But totally explaining any phenomena is exceedingly difficult. Such high correlations for the social sciences are impressive. A correlation gives us a decent reason to investigate whether or not causation is taking place between two variables. If you see some sort of sociological phenomena, it is reasonable to ask “Is IQ involved?” because you would probably be correct.
Intelligence research can give us some skepticism about other explanations. If someone believes that the number of books in a home makes a child good at school, what may actually be going on is that a parent who likes to read is more intelligent and keeps books at home and that intelligence is passed onto their children because intelligence is heritable. It will not be the case that intelligence is always the cause but it frequently is playing a role. The downside is that it gives us doubt about the possibility for intervention. We do not know how to raise IQ effectively but we know how to not lower it. Avoiding iodine deficiency is an example of a possible intervention.
The above chart displays IQ ranges and the percentage of each category that matches the description. We see that most people are in the middle and at moderate risk for the negative socioeconomic outcomes. At the far right, a small percentage of the population is unlikely to be a high school drop out, chronic welfare recipient, incarcerated, live in poverty and so forth. At the far left, the chances are significantly higher. When we think about socioeconomic outcomes, we need to understand that not all populations are equally at risk. It is reasonable to suspect that IQ is playing a causal role in these outcomes rather than just being a correlate. To discuss the reasons for thinking so would take much longer. I wanted to make this an introduction to the topic and an explanation for my interest in the topic and my belief in its importance.
Jensen, Arthur. The g Factor: The Science of Mental Ability (1998). pp. xi - xii
Interesting introduction to a fascinating topic. I think the biggest questions in intelligence research ought to center on whether or not intelligence is heritable, and if so, by how much.