Education and Intelligence-A Response to Charles Murray

Editor’s Note.  This is the first of a four-part series by Tom Wood that will we will post this week.  Part 2 will be posted tomorrow, March 3.

By Tom Wood

In one of my recent postings, I discussed the Collegiate Learning Assessment (CLA) in connection with Charles Murray’s claim that the B.A. degree is meaningless. This paper continues that discussion, and sets out in greater detail the implications and importance of the CLA for educational psychology and educational practice.

In his recent book, Real Education, Murray contends that programs like No Child Left Behind (NCLB) cannot succeed in attaining their major goals. According to Murray, NCLB and programs like it will inevitably fail because they are strongly and inherently constrained by the limited academic ability of many of the students enrolled in school. Since this academic ability cannot be improved, except within very narrow limits, all such programs are destined to fail. As he puts it: “The most we know how to do with outside interventions is to make children who are well below average a little less below average.”

There are, according to Murray, severe limits to how much academic ability—e.g., the ability to reason, make inferences, employ logic and analysis—can be developed. That is because, on his view, any test, whether an IQ test or a test of academic aptitude or competence that implicates such academic abilities, will be strongly g-loaded, and because g—the general intelligence factor—is largely (though not entirely) inherited. For Murray, like many other intelligence researchers, the general intelligence factor g is pretty much set by early adolescence. As he puts it (p. 51): “All the data about the trajectory of IQ scores over the life span indicate that they stabilize around ages six to ten and typically remain unchanged until old age.”

In fact, the view that schools cannot develop academic abilities and fundamental logical skills is mistaken. To be sure, their ability to do so is modest, particularly as compared with factors like family income and other aspects of socioeconomic background. However, no one really doubts that schools do make a contribution to intelligence, IQ, academic ability (aptitude), and cognitive skills. (I mention four things here because there are important studies focusing on each of these, although researchers are often in disagreement about how they are related to each other.) This is a well established finding, and has been demonstrated using the very national exams in K-12 that Murray discusses in Real Education.

As Murray insists, the national tests he discusses do engage fundamental cognitive skills and academic abilities. (He often refers to this as IQ.) Yet these very tests get progressively harder from grade level to grade level, even though failure rates do not increase from level to level. Furthermore, studies have shown that these two facts cannot be accounted for by the dropouts that occur from grade to grade.

It must follow that cognitive skills and academic abilities do improve in school from grade to grade, as the studies have shown. Murray may have overlooked this because he is usually focused (as is typical of much IQ research) on comparing students with each other, rather than on what is happening to the average raw scores of the entire cohort. Although the relative standing of students on these national tests remains fairly stable over time from grade level to grade level, this does not mean that schools do not improve the cognitive skills and academic abilities of all students in the cohort. Clearly, they do.

Learning transfer and cognitive skills

The ability to use existing knowledge in new settings to solve problems and to gain new knowledge more rapidly is called learning transfer. Academic abilities like the ability to reason, make inferences, employ deductive and inductive logic, and to think critically are the kinds of abilities that are critical to learning transfer, which has been called the Holy Grail of education. Individuals have been found to vary much less in their ability to learn facts and benefit from rote learning than in the foregoing academic skills. Furthermore, skills and knowledge that are memorized or developed by rote learning do not get transferred from one domain or another, or from the classroom setting to practical problems in the world outside the classroom, in the absence of basic cognitive and academic skills. To a very large extent, the value of education largely depends on the claim that the cognitive skills necessary for learning transfer can be developed or learned in the classroom, so it is easy to see why educational psychologists have been obsessed about learning transfer ever since the foundation of the field in the early 20^th century.

Hence the fundamental importance for educational psychology of the finding from the first, two-year longitudinal study of the CLA by the Social Science Research Council (SSRC) that what might be called “high transfer” abilities are developed in college. The CLA includes performance tasks that are typical of management positions in the business world. The performance tasks of the CLA, in effect, test for learning transfer. They test how well the critical thinking, analytic reasoning, problem solving, and written communication skills that are engaged by the student in courses in business, math, engineering, political science, philosophy, history, or English literature, get transferred to real world tasks.

Murray in his Real Education does not discuss the CLA, but he does discuss a similar national test, begun in the 1970s, called the National Assessment of Educational Progress (NAEP), which is often referred to as the “nation’s report card.” The NAEP contains reading abilities sections for various grade levels that are very similar to the performance tasks in the CLA. (I would not be surprised to learn that these sections actually provided the inspiration and model for the CLA.)

The sample 8^th grade reading item from the NAEP that Murray discusses at some length in Real Education can be found on the National Center for Education Statistics (NCES) site here; the reading assessment begins on p. 16 of that online document. The sample reading assessment consists of a short essay with pictures entitled “The Lost People of Mesa Verde."

I will not give the exam questions for this reading assessment here, as I prefer to give the details about the 12^th grade exam later, but I do want to discuss what Murray says about the test. In his discussion of the 8^th grade reading assessment test (he would make identical statements about the 12^th grade test), Murray stresses the cognitive skills that the assignment engages. He points out that the task goes far beyond decoding the syntax and semantics of the words. Simply knowing the language will not be enough for the student: she must use academic and cognitive skills like the ability to make logical inferences and search strategies to answer the questions. As he points out, that is why it is difficult if not impossible to teach to the test in state competency exams. Teachers can improve students’ scores if the tests stick to a particular pattern, but if the context is new, the student cannot do well on the test unless she can transfer what is learned from the context that was learned or drilled in the classroom to the new one.

Interestingly, Murray cites Edward L. Thorndike—the founder of educational psychology, one of the founders of the IQ testing movement, a prominent behavioral psychologist, and one of the earliest and greatest skeptics about learning transfer—in this connection. The quote comes from a 1917 paper by Thorndike entitled “Reading as Reasoning.” In discussing the reading assignment about the Anasazi at Mesa Verde at the 8^th grade level in the NAEP exam, Murray says:
 

“…[Thorndike] put the intimate interconnection between reading comprehension and academic ability as well as anyone since: ‘The mind is assailed as it were by every word in the paragraph. It must select, repress, soften, emphasize, correlate, and organize, all under the influence of the right mental set or purpose or demand.’”

Murray insists that turning in a satisfactory performance on the NAEP requires the ability to make inferential leaps: putting two and two together in novel settings. To the readers of his book, he says, doing this seems easy and obvious. We do it all the time, without even thinking about it, so it takes a moment of reflection to realize how difficult it would be to teach such inferential skills. Murray insists that this is not just hard: it is actually impossible unless the student brings to the task the requisite—and mostly inherited—academic ability, i.e., g or IQ. Murray makes the very same point about the logical-mathematical ability that is required in the algebra test items in the NAEP.

 The U.S. Department of Education is in agreement with Murray about the kinds of demands its tests place on students. About its reading tasks (at grades 4, 8, and 12) the NAEP says:

 Across the three contexts for reading, students are asked to demonstrate
their understanding by responding to comprehension questions that reflect four
different approaches to understanding text. The NAEP framework accounts for these
different approaches by specifying four aspects of reading that represent the types
of comprehension questions asked of students. Forming a general understanding
questions ask students to consider the text as a whole. Developing an interpretation
questions ask students to discern connections and relationships within the text. Making
reader/text connections questions ask students to connect information from the text
with prior knowledge and experience. Examining content and structure questions ask
students to critically evaluate the content, organization, and form of the text. All four
aspects of reading are assessed at all three grades within the contexts for reading
described above.

Thus, the U.S. Department of Education itself insists, like Murray, that the NAEP reading tests involve tests of reasoning ability and not just grammatical skills. Doing well on the tests requires the student to use analytical and cognitive skills that transcend simply knowing English and the meaning of the sentences. As in the case of the CLA considered above, critical thinking, analytic reasoning, problem solving, and written communication skills are invoked, however modestly.

Forty-one percent of students got the wrong answer to the question in the 8^th grade level reading assignment about the Anasazi at Mesa Verde that Murray discusses at some length in Real Education. The central point of Real Education is that in most cases schools cannot be held responsible for this high failure rate. Schools cannot be held responsible for the fact that almost half of the NAEP test takers cannot successfully negotiate simple items involving straightforward cognitive tasks, because they aren’t responsible for the levels of academic ability (IQ) that are largely determined by heredity, and because academic abilities of this sort cannot be taught or developed by schools.

But this is precisely the challenge presented to Murray by the preliminary findings of the CLA, as I pointed out in my previous posting. According to Murray, the CLA should not be able to demonstrate cognitive gains and learning transfer from the classroom to its performance tasks on the reading tests. These are essentially the same kind of performance tasks as those presented by the various grade levels of the NAEP. Nevertheless, colleges, it appears, do demonstrate gains for students on these skills. The CLA findings indicate that higher education does much more than simply impart facts in academic disciplines: it actually in a broad and general and fundamental way makes college students smarter. The early findings from the CLA therefore present a challenge to all educational psychologists who are skeptical about learning transfer and the claim that fundamental cognitive skills can be learned and developed.