“It’s better to know nothing than to know what ain’t so,” the 19th-century humorist Josh Billings once observed.
Speaking recently to a parent-teacher group, I argued that the so-called “basic skills” assessed by standardized tests were not basic in any meaningful sense. I allowed that most people think these skills are prerequisites for learning higher-order thinking skills, but that people who think this know what ain’t so.
A parent interrupted, saying that there had to be a certain order to learning. A child couldn’t learn to multiply, the parent said, until he’d learned how to add.
I smiled and shook my head no. His look suggested that he thought either that I was crazy or that I might be right and the ground underneath his reality had been revealed as quicksand.
Fortunately, teachers were nodding affirmatively--yes, you can learn to multiply without knowing how to add. We teach the various mathematics operations as if addition has to come first, but it ain’t so.
It would be easy to dismiss this man as one parent operating from a common but untenable theory of how learning progresses. It would be easy, but wrong.
In fact, not only the general public but also many people in education and test publishing believe that there are “basic skills,” that they are measured by the standardized achievement tests currently on the market, and that they must be mastered before children can proceed to higher-order skills. This view is nonsense.
The director of research and measurement for the publisher ctb/McGraw-Hill, Paul L. Williams, was recently quoted as saying, “If children don’t have basic skills, there is no way that they can do higher-order skills.” (See Education Week, Nov. 16, 1988.) For Mr. Williams, the “basic skills” were defined by the California Achievement Test and the Comprehensive Test of Basic Skills--both published by ctb/McGraw-Hill.
Mr. Williams should know that the cat and the ctbs can’t measure whether children “have” basic skills because the items on these and other norm-referenced tests are selected so that 50 percent of the test takers miss them. Measures of mastery cannot be obtained from such tests.
Again, it would be easy, but wrong, to dismiss Mr. Williams’s comment as the self-serving propaganda of a man whose livelihood depends on selling more of the cat and the ctbs in 1989 than in 1988.
In February, the most recent report of the National Assessment of Educational Progress implied that the way we teach low-level skills is inimical to the teaching of higher-order skills. A Denver newspaper quoted the principal of a local high school: “I don’t see the relevance of the 1st to 4th graders worrying about high-level skills if they don’t have the basic skills.”
And last May, I visited the Key School, a magnet school in the Indianapolis public-school system. The principal and teachers there have tried to develop a curriculum derived from the Harvard University psychologist Howard Gardner’s notions of multiple intelligences. While the school has its supporters, it also has detractors. The latter, I learned, were waiting for results from the Indiana Statewide Test for Educational Progress--a modified version of the ctbs--to arrive. Poor scores would prove that the Key School’s approach was based on a fuzzy concept without educational merit.
Mr. Williams and those who share his outlook hold a naive, behavioristic view of the learner. Their position reflects no theory of how children grow and develop--largely because most people in the field of testing know nothing about children. They are untrained in either cognitive psychology or child development and spend no time in schools. Many believe in what has been called the “bean-jar curriculum.”
According to this perspective, large skills such as reading develop from the simple accretion of small skills--the “beans.” One learns to read by filling the jar labeled “reading” with the digraph bean, the consonant-blend bean, the comma bean, and so forth. No one in the field of reading would argue this position, but it is implicit in most tests and in many texts.
Many people who hold this view of learning believe, for example, that the literacy skills assessed on standardized tests are necessary for thinking. This particular nonsense, if true, would mean that no blind person can think. It would mean that Socrates, an illiterate who argued against literacy on the grounds that it would destroy memory, could not think.
To the best of my knowledge, no anthropologist has ever returned from studying a preliterate society and said, “Those folks can’t think.”
In a 1982 analysis that went well beyond its latest report, naep wrote:
“Children can and do learn large chunks of very difficult material very early. Lower-order so-called basic skills are not necessarily the ‘building blocks’ essential to acquiring higher-order skills such as problem solving, analyzing, synthesizing. ... Learning is not the linear process as popularly perceived by the public.”
What are the skills that are considered “basic” by the test makers? In fact, they vary from test to test--a circumstance that raises in a different way the question of what is basic. One test will ask a child to recognize which word of four is spelled incorrectly; another will ask which is spelled correctly. Neither will ask the child to spell the word.
In the early grades, tests assess word-analysis skills, vocabulary (''find the picture of an insect”), visual recognition of letters, visual recognition of sounds, reading comprehension of sorts. Whether or not these particular skills are useful, the tests in their current form are oriented toward a skills approach to reading, not to the current theories of reading as a meaning-making--as opposed to meaning-extracting--activity. But if “basic skills” are basic to only one approach to teaching, in what way are they truly basic?
Unfortunately, “basic skills” on tests keep changing. Once upon a time, children coped with terms like commutativity and associativity on mathematics tests. Later, they had to handle problems in set theory. Currently, they respond to questions about quantity and operations. One would think that if these skills were “basic,” they would have more longevity.
For years, the National Council of Teachers of Mathematics has recommended de-emphasizing computation, leaving it to machines--it bores children and there is even evidence it inhibits thinking. Yet the current form of the Iowa Test of Basic Skills requires children to do computations under time pressure.
All of these “basic skills” are tested in artificial, decontextualized settings. Below 3rd grade, some of these settings can be characterized only as bizarre. They probably do stimulate higher-order thinking--as the children struggle to figure out how to deal with them. Some of the questions can be answered correctly only when children have figured out the rules of the testing game.
As soon as we look at any theory of the development of children and their intellects, the notion of “basic skills” as defined by standardized tests--and of the bean-jar curriculum--evaporates into silliness.
All cognitive theories of which I am aware stress the role of prior knowledge in acquiring new knowledge and making sense of it. The schema theorists, whole-language theorists, and cognitive scientists alike emphasize the importance of context in enabling learning. But the basic-skills approach presents fragmentary skills in isolation.
Our addiction to low-order skills has had some disastrous consequences. For example, over the last few years, scores on the basic-skills tests have been rising while those on tests of higher-order skills have been declining or at best holding steady.
This is no accident. Scores on the higher-order-skills tests have been falling precisely because we have been overteaching for tests such as the cat and the ctbs at the expense of the other skills. Teachers say they no longer give essay exams--so they can prepare children for tests requiring them to respond to decontextualized, fragmentary bits of knowledge.
The result? Is there any other nation in the world where “Trivial Pursuit” could have become a national pastime? Teaching children in this way and hoping that they will learn to think is like teaching them when and how to slide into second base and hoping that they will get the gestalt of how to play baseball. It won’t work; they’ll strike out.
We never ask “what is this skill basic to?” With most skills as taught currently, we would see that we have few answers to this question--and thus little justification for4teaching what we do the way we do except that of inertia.
Some years ago, I proposed what I called the “zero-base curriculum,’' analogous to the zero-base budget: Toss everything out and then justify everything you let back in. It still seems like a reasonable exercise.
It is difficult for us to ask “what is this skill basic to?” for two reasons. First, the cognitive theories are not yet very useful in specifying what might in fact be basic, prerequisite skills; they are better for seeing what won’t work than for planning what will.
Even given these ambiguities, however, a list of the characteristics of higher-order thinking shows how irrelevant tests of “basic skills’’ are. Higher-order thinking is nonalgorithmic; it tends to be complex; it of8ten yields multiple solutions. It involves nuanced judgment and interpretation, the application of multiple criteria, uncertainty, self-regulation. It requires imposing meaning, and it’s effortful. Were we to devise tests of the precursors of such thinking, we would end up with something quite different from the typical four-choice multiple-choice test.
Second, by concentrating on fragmented skills, we have lost any clear idea of what we want children to be good at. This problem, though terrible, is more tractable. In our curricular deliberations, we would do well to imitate the performing arts and athletics: When we suit up for football or sit down at the piano, there is never any question what shape the final performance ought to take. Almost every exercise we undertake, jumpel10ling jacks or scales, serves that final performance.
Could we not ask what we want children to be good at in academics and work back to the skills that will contribute to their succeeding in these areas? I think we could--admitting all along that some skills seem important although their connections cannot be verified and that some skills should be learned just because you never know when they might come in handy.
For example, historians say they want students to develop a sense of historical empathy, a sense of sequence, an understanding of cause and effect, a distrust of cut-and-dried explanations of events, an understanding of why things change or don’t, and an understanding of the tension between traditionalism and dynamism. Yet no basic-skills test measures those outcomes or any of the mental processes contributing to their development
But there is no reason why we couldn’t build a test of these outcomes and tests of other skills that seem important along the way.
Until we decide what children ought to know and be good at, I predict we will continue to see report after report in which our young people--compared with children of other nations--do poorly on tests of all kinds. I hasten to add that this stance does not imply creating a canon of common knowledge. The description of important knowledge and skills could be as open-ended in other subjects as it is in history.
But if we create such a list, we will find that many of these abilities develop slowly and that they are hard, if not impossible, to measure with standardized, multiple-choice tests. And we might even take some steps toward a meaningful definition of what’s basic and why.
A version of this article appeared in the April 05, 1989 edition of Education Week