Test Developer Profiles

Chapter 10

Alan S. Kaufman, Ph.D., and Nadeen L. Kaufman, Ed.D.

Tests Developed:

Kaufman Assessment Battery for Children (K-ABC)

Kaufman Test of Educational Achievement (K-TEA)

Kaufman Brief Intelligence Test (K-BIT)

Kaufman Adolescent and Adult Intelligence Test (KAIT)

Alan S. Kaufman, Ph.D., is Professor of Psychology at the Yale University School of Medicine, Child Study Center, and Nadeen L. Kaufman, Ed.D., is Lecturer at the Yale University School of Medicine, Child Study Center. Alan was born in Brooklyn, New York, in April 1944, and Nadeen was born in Brooklyn, New York, in January 1945; both were raised on Long Island. They met as college freshmen and have been married since 1964.

Alan earned the A.B. degree from the University of Pennsylvania in 1965; M.A. in Educational Psychology from Columbia University in 1967; and Ph.D. from Columbia University in 1970 (under Robert L. Thorndike in Psychology--Measurement, Research, & Evaluation). Nadeen earned the B.S. in Education from Hofstra University in 1965; M.A. in Educational Psychology from Columbia University in 1972; Ed.M. in Learning and Reading Disabilities from Columbia University in 1975; and Ed.D. in Special Education--Neurosciences from Columbia University in 1978 (under Margaret Jo Shepherd). While Assistant Director at The Psychological Corporation from 1968 to 1974, Alan worked closely with David Wechsler on the revision of the Wechsler Intelligence Scale for Children (WISC) and supervised the standardization of the revised version--the WISC-R. He also collaborated with Dorothea McCarthy in the development and standardization of the McCarthy Scales of Children's Abilities. From the mid-1970s to the present Alan has had several university affiliations prior to his current Yale position, most notably the University of Georgia (1974--1979) and University of Alabama (1984--1995). Nadeen is a researcher, clinician, and university trainer of clinical and school psychologists who has founded several clinics for the psychoeducational assessment of children and adults (including clinics at Mesa Vista Hospital and the California School of Professional Psychology, both in San Diego). Immediately prior to Yale, she was Professor at California School of Professional Psychology, both in San Diego). Immediately prior to Yale, she was Professor at California School of Professional Psychology. During her career, she has worked as an elementary school teacher, teacher of learning disabled children, school psychologist, psychoeducational diagnostician, clinic director, and university professor. Alan's texts (filled with Nadeen's case reports) have been widely used for the interpretation of Wechsler's scales for children, adolescents, and adults: Intelligent Testing with the WISC-R (1979), Assessing Adolescent and Adult Intelligence (1990), and Intelligent Testing with the WISC-III (1994).

Alan and Nadeen live in Stamford, Connecticut, about midway between Yale and Lincoln Center. They live with their daughter (Jennie, a clinical psychologist), granddaughter (Nicole, born in 1987), and four toy poodles, and share a passion for the Metropolitan Opera and New York Philharmonic Orchestra. Alan also has a passion for baseball; he and son James (a Yale doctoral student of Robert Sternberg's) have co-authored many baseball research articles as well as the 1995 book for Citadel Press, The Worst Baseball Pitchers of All Time. Alan and Nadeen also have three grandchildren who live in California, as does their son, David.

The influences on Alan and Nadeen's career in test construction have been many and varied. Alan considers his primary mentors to be David Wechsler and Alexander Wesman. Dr. Wechsler provided Alan with an informal apprenticeship in the art of clinical assessment and Dr. Wesman introduced him to the concept of "intelligent testing;" the teachings of both mentors formed the basis of Alan's psychometric/clinical approach to the interpretation of intelligence tests. Nadeen traces much of her approach to assessment to the training she received at Teachers College, Columbia University, during the 1970s. Drs. Margaret Jo Shepherd and Jeannette Fleischner (whose recent, tragic death leaves a void in the field of learning disabilities assessment and remediation) were instrumental in training her to treat each child referred for learning disabilities assessment with respect, and to apply a humanistic approach to solving each individual's particular set of learning problems. The goal was to identify intra-individual differences and pinpoint strengths and weaknesses within the child's ability spectrum, and to interpret these profiles in the context of developmental psychology. The clinician's task was to be a detective and, ingeniously, to use the integrities to help ameliorate the weaknesses. The burden was on the diagnostician. The child was perceived as ready and able to learn so long as the clinician was able to identify the most appropriate learning environment. This approach was unique to Teachers College at the time, and made a big impression on Nadeen. These vicarious experiences for Alan were also vital in his development of interpretive approaches for the Wechsler scales, and both Nadeen and Alan used this humanistic, intra-individual, developmental psychology approach in their 1977 book Clinical Evaluation of Young Children with the McCarthy Scales.

Another Teachers College influence on Nadeen was Dr. Martha Bridge Denckla, whose stimulating course on the neurology of learning disabilities and Minimal Brain Dysfunction brought a new dimension to her understanding of children's learning problems. This aptly initialed professor (MBD) taught the anatomy of the brain during the first part of the course, and then used a dynamic case study approach to show how specific anatomic dysfunctions impact on visual perception, motor behavior, cognition, language development, auditory processing, and so forth. The blend of neurology, psychology, and education helped focus Nadeen's approach to assessment and, ultimately, to the development of new intelligence tests.

At the same time that Nadeen was benefiting from the unique approach of her professors to learning disabilities assessment, both Nadeen and Alan were gaining new experiences within the growing field of school psychology. Alan joined the Educational and School Psychology Department at the University of Georgia in 1974 and Nadeen was actively involved in the program at the university, first as a school psychology intern and later as a professor who supervised student teachers and school psychology graduate students. Dr. Paul Torrance was chairperson of the Educational and School Psychology Department, and his passion for Roger Sperry's cerebral specialization theory served as a catalyst to the deep interest that Alan and Nadeen had already felt for the left-right distinction. They saw wonderful possibilities in the application of split-brain theory to the measurement of intelligence. As a new trainer of school psychologists at the university, Alan was taken out of the ivory tower of The Psychological Corporation and put on the direct firing line of the real school and clinic cases encountered by his graduate students during their practicum and internship experiences. He quickly saw the flaws of the tests that were universally used for assessment (Wechsler and Binet). He was especially concerned about their lack of theoretical foundation and about the fact that all tasks on these instruments stretched back more than a half-century to the Binet-Simon and Binet-Henri laboratories and to the construction of nonverbal tests during World War I. Nadeen worked as a psychologist at the Rutland Psychoeducational Center in Athens, Georgia, and faced the real-world dilemmas of trying to develop suitable educational plans for the emotionally disturbed and learning disabled children and adolescents that she assessed. She found the Wechsler scales limited in their usefulness for making these educational translations, and experienced a great need for theory-based instruments designed with educational applications in mind.

The more Alan and Nadeen explored the literature, the more they realized that other theories besides Sperry's focused on distinctions between two types of problem solving, one characterized by linear, analytic, sequential thinking (attributed to the left hemisphere by Sperry and his colleagues) and the other defined by gestalt, holistic, simultaneous integrations (right hemispheric processing to Sperry). The Russian neuropsychologist, Luria, distinguished between successive and simultaneous information processing, and the cognitive psychologist, Neisser,  described serial and parallel or multiple processing. Alan and Nadeen were intrigued by the convergence among theorists from the diverse domains of psychobiology (Sperry), neuropsychology, and cognitive psychology, and came to believe that these two processes--regardless of their localization in the brain--corresponded to two fundamental types of mental processing or problem solving. The link seemed obvious: If you can measure how well (or poorly) children solve problems by sequential and simultaneous methods, then you are one step closer to making meaningful school-based recommendations for children who have learning difficulties.

There was also a practical side to Alan and Nadeen's decision of a conceptual framework for their initial intelligence test, the Kaufman Assessment Battery for Children (K-ABC). The preponderance of children who are referred for evaluation have learning problems of some sort, and many of these children are bilingual, bicultural, or both. Traditional tests of verbal intelligence are heavily weighted in school-based tasks that are really measures of achievement (oral arithmetic, vocabulary, general information). To the learning disabled and minority children who are referred for evaluation, such tasks are unfair as measures of ability and are more reasonably classified as achievement tests. How is it fair to compare IQ to achievement as part of the learning disabilities diagnostic process if a chunk of the IQ score is really achievement? That practical question, which was an outgrowth of Nadeen's and Alan's clinical and training experiences in Georgia, impelled the decision to separate intelligence (mental processing) from achievement (fact-based and language-based knowledge) on the K-ABC. The availability of an incredibly strong team of graduate students at the University of Georgia, who were eager and willing to help build a new, theory-based intelligence test that departed from the Wechsler model, completed the test construction package. That team included many of the current leaders in school psychology and assessment: Bruce Bracken, Jack Cummings, Patti Harrison, Randy Kamphaus, Jack Naglieri, and Cecil Reynolds. Thanks to the group effort, the K-ABC, begun in 1978, was published in 1983.

How the invitation came about to develop the K-ABC is another story. Alan and Nadeen often drove the two hours from Athens to Atlanta to the diversity of shopping malls in the larger city. On one particular Sunday in March of 1978 they spent the two hours to Atlanta developing the blueprint for the test that would become the K-ABC. (They were taking their three children to see Luis of Sesame Street fame to appease their guilt about working too hard; then they promptly ignored the kids while developing the test.) On that car trip, they discussed having intelligence scales corresponding to Sperry and Gazzaniga's notions of analytic sequential and gestalt-holistic brain processing; separating problem solving from acquired knowledge; and including a variety of tasks to measure many different cognitive styles. (The latter idea was dropped, which was a good thing. The battery they developed that day, if fully operationalized, would have taken about ten hours to administer. The term "experimental mortality" would have taken on a new meaning.) In essence, though, they developed the entire foundation of the K-ABC on that car trip, from the basic cerebral specialization processing model to the distinction between mental processing (intelligence) and achievement to the need to reduce mean differences between Whites and minority children. However, reality hit on the ride home that night. Alan and Nadeen realized that no major publisher would would challenge the Wechsler-Binet monopoly and that all of their planning for a theory-based intelligence test was fruitless. By the time they reached Athens, they dismissed the whole notion as fanciful. The very next day, Dr. Gary Robertson, then the Director of Test Development at American Guidance Service (AGS), telephoned to ask Alan, "Would either you, or Nadeen, or both of you want to develop a new intelligence test to challenge the Wechsler?" Alan replied, "Why we just developed one yesterday!" He immediately realized how foolish that answer sounded, and hemmed and hawed in an effort to regain credibility. Within a month, a contract and blueprint were hammered out in a meeting at an Atlanta Holiday Inn, and the K-ABC was begun. And that is a true story.

One of the most interesting occurrences during the test development phase was an ethical dilemma that Alan and Nadeen faced. One goal was to construct novel tasks (and, oh, was it difficult to think of tasks that no one had thought of before, either for a test or for an experimental research task). One new subtest developed by the research team was a nonverbal paired-associate learning task in which a stimulus picture was paired with a response picture; the child had to select the correct response picture for each stimulus picture until he or she learned all of the pairs. This task was included in a pilot study along with other new tasks. Alan and Nadeen were momentarily excited to discover that this nonverbal paired-associate learning task was the only one that failed to discriminate among three groups--learning disabled, mentally retarded, and normal controls. The three samples earned almost exactly the same mean score, even though they earned predictably different mean scores on all other tasks, including another paired-associate learning task that involved pictures paired with words. What a thrill! A cognitive task was developed that failed to discriminate against mentally retarded children, the ultimate in fairness. Then reality set in almost instantaneously. If a task did not discriminate between those with IQs of 100 and those with IQs below 70, then it had no construct validity. Alan and Nadeen had to scrap the task as a failure and have felt guilty ever since. They discovered a test that did not penalize retarded children and couldn't include it in their test.

Ultimately, Alan and Nadeen have thoroughly enjoyed their venture into test development. After the K-ABC came many other tests, such as the K-TEA, K-BIT, and KAIT. The latter test was based primarily on the Horn-Cattell distinction between fluid and crystallized intelligence. Even though the K-ABC was partly rooted in the Horn-Cattell model (with mental processing roughly equivalent to fluid problem solving and achievement analogous to crystallized knowledge), the KAIT model, applied to the 11 to 85+ age range, was different. It did not deal with sequential versus simultaneous processing, and it represented a theoretical rerouting. The Horn-Cattell model was used to explain the rise and fall of intelligence across the lifespan by researchers who conducted many aging/IQ studies, and it made more conceptual sense for a test for adolescents and adults than did the neuropsychological processing model.

This shift in theoretical frameworks was consistent with Alan and Nadeen's realization that different sets of tasks were needed on the preschool and school-age levels of the K-ABC because of developmental changes in the constructs with age. Similarly, developmental considerations influenced the need to modify the basic theoretical framework when developing a test for adults versus children. If they were to give advice to new test developers it would be to keep abreast of the vast array of theories from a variety of disciplines and keep up-to-date on research in the many areas that pertain to cognitive assessment. Be flexible in your choices because assessment is done for many reasons, and these reasons differ by setting, age group, and cultural or subcultural group; no one theory or model is of the vast array of theories from a variety of disciplines and keep up-to-date on research in the many areas that pertain to cognitive assessment. Be flexible in your choices because assessment is done for many reasons, and these reasons differ by setting, age group, and cultural or subcultural group; no one theory or model is necessarily the right one, nor is your favorite approach necessarily pertinent to the test development challenge you may be facing. And even though it is essential to be psychometrically sophisticated and develop a test that meets the highest standards regarding reliability, validity, and standardization procedures, it is always wise to remember the credo of David Wechsler: "IQ tests are, first and foremost, clinical instruments."