Most parents know that school administrators use intelligence tests to identify gifted children. What most parents do not know is that the content of these tests can have a powerful influence on who is identified as gifted and who is not. The tests can block minority children, especially those with limited English-language skills or limited preschool experience, from admission to special programs, or they can be the keys that let these children in. It all depends on which tests are used.
First, it is important to consider the difference between a test of achievement and a test of ability or intelligence. A test of achievement measures what children have learned by being alert and aware; what they have been taught by parents, siblings, friends, and teachers; and what they have absorbed from other sources of information. It measures how well they read, spell, do math, and apply vocabulary and factual information. A test of ability or intelligence, by contrast, measures how “smart” they are. In other words, a child’s intelligence makes it possible for him or her to acquire information and skills, given the opportunity. Achievement is acquired; ability is a factor in attaining knowledge.
Can someone be smart but not know much? Yes. Many intelligent children are not good at math, do not speak English well, cannot read at grade level, and do not do well in school simply because they have not had much opportunity to learn these subjects and skills. How can we find these smart, perhaps gifted, children who do not look the part in the classroom?
Nonverbal tests offer a different way to identify children with high potential.
We will not find smart children who do not have good language skills by using traditional intelligence tests. Why not? Group-administered intelligence tests, sometimes described as tests of “school ability” or “scholastic aptitude,” are typically made up of questions divided into verbal, quantitative, and nonverbal parts. The verbal parts often include questions that assess vocabulary and the test taker’s ability to make word analogies. The quantitative parts include word problems, as well as numerical series (e.g., 1, 2, 4, 8, ?).
Understandably, then, the verbal and quantitative parts are difficult for children who have limited English-language skills and have had few chances to develop literacy skills. Even the so-called nonverbal parts of these tests can be difficult for such children, because the directions are often presented in a verbal format.
Nonverbal tests of general ability are designed specifically to measure intelligence independently of language and math skills. What do such tests look like, and how do they work? The figure on page 2 shows a typical question from the Naglieri Nonverbal Ability Test (NNAT). The top row contains the sequence oval, square, cross, with light shading, no shading, and moderate shading, respectively. The first column, by contrast, contains the sequence oval, cross, square, but with the same sequence of shading. It is clear that there are several patterns in this figure and that the test taker has to understand them to answer the question. The NNAT is given with a minimum of verbal instructions: for instance, “Look at this picture [the examiner points at the sample]. Which one of these goes here [the examiner points at the options and then at the question mark]?”
Since one’s knowledge of words and mathematics is much less important on nonverbal tests of general ability than one’s ability to determine the relationships among shapes in a design, children who do not come from enriched backgrounds are not at a disadvantage when they take these tests.
Studies that my colleagues and I have done show that the NNAT’s approach to measuring intelligence is more fair to a wide variety of children: groups of black, white, and Hispanic children earned similar scores; similar percentages of black, white, and Hispanic children earned scores that qualified them for gifted programs; Hispanic children with limited English-language skills and Hispanic children who only spoke English earned similar scores; and, finally, Native American children and non–Native American children earned similar scores. These studies support the view that a nonverbal test can help identify gifted minority children.
Has this approach been tried in the schools? George Mason University’s Center for Cognitive Development recently completed a project to help Fairfax County Public Schools, in Fairfax, Virginia, identify gifted children. Located in the Washington, D.C., metropolitan area, this is the twelfth largest public school district in the United States; it serves about 160,000 students, many of whom speak English as a second language. The NNAT helped find more gifted minority children who are bilingual—children who would not have been considered for a gifted program if their verbal and quantitative scores had been used.
Not that verbal and quantitative tests should not be used. But a nonverbal test can supplement them, because it increases the percentages of gifted minority children who are identified as such. Moreover, these children, since they are different from children with high verbal skills, will require a differentiated curriculum commensurate with their differing intellectual and instructional levels. They will need this special assistance because they deserve to be treated as gifted children, even though they do not currently demonstrate high academic achievement.
—Jack A. Naglieri, Ph.D.
Jack A. Naglieri is professor of psychology and director of the Center for Cognitive Development at George Mason University. He is trained as an educational and school psychologist specializing in the assessment of intelligence and cognitive processing.