Appletcollection Vertical Menu java applet, Copyright 2003 GD

For a complete review, see:

Dumont, R. & Hagberg, C. (1994) Kaufman Adolescent and Adult Intelligence Test (KAIT): Test Review Journal of Psychoeducational Assessment, 12, 2, 190-196

Alan and Nadeen Kaufman, through American Guidance Service (AGS), have released the "new kid on the block" in intelligence testing: the Kaufman Adolescent and Adult Intelligence Test (KAIT) ($495). This individually administered intelligence test is designed to assess persons 11 to 85+ and is composed of separate Crystallized and Fluid scales. The theoretical model for the scale was derived from Horn and Cattell's (1966, 1967) conceptualization of Fluid vs. Crystallized Intelligence while models for the 8 subtests included Luria and Golden's (1981, 1980) planning ability and Piaget's (1972) formal operations. Three subtests (Definitions, Auditory Comprehension, and Double Meanings) form the Crystallized scale while three more (Rebus Learning, Logical Steps, and Mystery Code) form the Fluid scale. An additional subtest for each scale (Famous Faces for the Crystallized and Memory for Block Designs for Fluid) are available as a substitute to one of the core subtests. Two measures of delayed recall (Rebus recall and Auditory Comprehension recall) permit the comparison of performance across time. Also included with the KAIT is a normed 10-item Mental Status subtest.

The test material comes in an attractive, locking briefcase which contains 2 easels; wooden blocks for the Memory for Block Designs subtest; an audio cassette for the Auditory Comprehension subtest (the user supplies tape deck); protocols for the test including separate Mystery Code booklets; and the test manual.

Standardization

The KAIT was standardized on a sample of 2000 people selected as representative of the US population on the basis of the 1990 population estimates. 13 age groups, ranging from 11 to 75+, are divided by geographic area, race and ethnic group, and examinee or parental education levels. The size of the sample varies by age. For the children's sample, age 11 through 16, 500 children were tested, for the adult sample, age 17 through 75+, 1500 were tested.

Deviation IQs and Scaled Scores

The IQ tables in the KAIT manual are based on only 6 of the 8 subtests. The two alternative subtests, Famous Faces and Memory for Block Designs, are excluded from the calculation of the IQ except if they replace a subtest not administered. Prorating of either scale IQs or Composite IQ is discouraged.

Reliability

Reliabilities for the KAIT are generally outstanding. Each of the 3 IQ scales has an internal consistency reliability that averages .95. The six core subtests have an average reliability of .90. Test-retest reliability for the Crystallized, Fluid, and Composite IQs were .94, .87, and .94, respectively. The subtest reliabilities are not as robust. The average test retest reliability for 5 of the eight subtests was found to be in the .70s. Because of these 'low' reliabilities, interpretation of the KAIT should focus on subtest groupings and not on individual subtests.

Validity

Construct validity for the KAIT's Fluid-Crystallized make-up was examined in a number of ways. Exploratory factor analysis found support for only these two meaningful factors. Each subtest conformed to its' hypothesized Crystallized vs. Fluid assignment.

Construct validity was also determined by comparing the scores obtained on the KAIT with those obtained on various other measures of intelligence, including the WISC-R (Wechsler, 1974), WAIS-R (Wechsler, 1981), SB-IV (Thorndike, Hagen, & Sattler, 1986), and the KABC (Kaufman & Kaufman, 1983). In general the KAIT was found to have high correlations, and thus, substantial variance overlap with the Wechsler tests and the Stanford Binet, while having lower correlations with the KABC mental processing composite.

Diagnostic validity studies were carried out with various clinical samples. These included Neurologically Impaired, Clinically Depressed, Alzheimer's Type Dementia, and Reading disabled. For each clinical sample a Control group was created from the standardization group by matching the clinical samples' age, gender, race, and years of education. These studies offer the examiner and researcher interesting interpretive hypotheses.

There are no comparisons for either a mentally retarded or gifted population. This might have been especially useful for the MR group since the studies by Zimmerman, Covin, and Woo-Sam (1986) and Rubin, Goldman, and Rosenfeld (1985) found the IQ classifications based on the administration of a WAIS-R to frequently change from those classifications found from the administration of a WISC-R. Another sample not included that might have been helpful was a learning disabled group. The manual did include a Reading Disabled sample, but this included only 14 children. A sample for the construct validity study also included a Learning Disabled sample, but this group was comprised of only 8 children. School district personnel using the KAIT will have to wait for research to determine its usefulness in identification of learning disabilities. One other concern with the clinical sampling was the inclusion of such a small number of children in the neurologically impaired group without clearly delineating the age breakdown in the manual. The total group contained 44 people. Of these, only 3 (two 11-year-olds, and one 17-year-old) were below the age of 20 (Colin Elliott, personal communication, March 1993). This may lead some to over interpret the results. The KAIT may be found to be very useful in differentiating various problems in neuropsychological, and psychoeducational samples. Clearly, more research is needed and expected.

Administration and Scoring

Administration time for the Core Battery (6 subtests) is approximately one hour, with the Expanded battery (6 core subtests, 2 alternative subtests, and 2 recall subtests) taking approximately 90 minutes. Directions for each subtest are contained on the easel pages, so no additional manuals are needed during administration. Subtests are given in the order presented in the easels.

All subtests, with the exception of the delayed recall subtests and Famous Faces, provide teaching tasks to ensure that the examinee fully understands the nature of what is expected. Teaching items allows the examiner flexibility in choosing alternative wording or method to ensure that the person tested understands each task.

Although a stop watch is necessary during the administration of the KAIT, no items are given 'bonus' points for successful completion within time limits. Several subtests require a response within a certain amount of time, and the designs for the Memory for Block Designs are shown for only 5 seconds.

Everyone administered the KAIT starts with the teaching items and begins formal testing with item number 1. No different age differentiated starting points are used. Discontinuation rules vary from subtest to subtest, and are clearly noted on the individual record form.

Because certain subtests require skills (reading, adequate hearing, etc.) that may prevent a person from adequately demonstrating ability on the KAIT, the easels provide cautions to examiners about which subtests might be omitted from the core battery because of such problems. The decision to omit a subtest needs to be made carefully and with good reason.

Administration flows smoothly from subtest to subtest. Recording responses on the protocol is simple and straight forward. Most verbal items require a simple word or number response. Scoring items as they are presented is necessary for discontinuation decisions.

Interpretation

Since the use of the KAIT is recommended for those legally and professionally competent to give existing intellectual assessments, some level of interpretive skill is expected of examiners. Four chapters of the manual provide background information about the subtests and theoretical information to aid interpretation.

The protocol provides tables used to determine, by age categories, the significance of the difference found between the Crystallized and Fluid scales. If the separate scales do differ significantly, base rate information is provided in the manual to determine frequency and unusualness of such scatter.

The last page of the protocol provides tables for comparing each subtest to the mean of the relevant scale. Three tables in the manual provide the actual difference needed for significance carried to 1 decimal point.

Because of the low test-retest reliabilities of 5 of the 8 subtests, examiners are instructed to focus interpretations on subtest groupings as opposed to individual subtests. (Each subtest does contain adequate levels of subtest specificity to be interpretable, but groupings are thought to be more stable.) Tables in the manual provide hypothesized groupings for both "Shared abilities" and "Influences affecting Performance."

Conclusion

The KAIT appears to offer an exciting new addition in psychologists' choice of intellectual assessment tools. Given the KAIT's age range (11 to 85+), it can easily substitute for a WISC-III at the adolescent level, and the WAIS-R at the adult level.

In the process of reviewing the test, the reviewer administered over thirty KAIT's. Test procedures were easily learned. The KAIT manual was very helpful. Both children and adults found the test to be non threatening, and for some, enjoyable and challenging.

Overall the test seems to be well thought-out and validated. This new test will probably become a test of choice for some researchers, especially those interested in assessing the geriatric population. The Kaufman's have gone to great lengths to standardize a reliable, and valid measure for older Americans. Valuable research should be conducted using the KAIT with school aged populations so that its merit for school psychologists can be determined.

Golden, C. J. (1981). The Luria-Nebraska Children's Battery: Theory and Formulation. In G. W. Hynd & J. E. Obrzut (Eds.), Neuropsychological assessment and the school-age child: issues and procedures (pp. 277-302). New York: Grune & Stratton.

Hansen, J. C., & Campbell, D. P. (1985). Manual for the SVIB-SCII (4th ed.). Stanford, CA: Stanford University Press.

Horn, J. L., & Cattell, R.B. (1966). Refinement and test of the theory of fluid and crystallized general intelligences. Journal of Educational Psychology, 57, 253-270.

Horn, J. L., & Cattell, R. B. (1967). Age differences in fluid and crystallized intelligence. Acta Psychologia, 26, 107-129.

Kaufman, A. S., & Kaufman, N. L. (1983). K-ABC Interpretive Manual. Circle Pines, MN: American Guidance Service.

Luria, A. R., (1980). Higher cortical functions in man (2nd ed.). New York: Basic Books.

Piaget, J. (1972). Intellectual evolution from adolescence to adulthood. Human Development, 15, 1-12.

Rubin, H. H., Goldman, J. J., & Rosenfeld, J. G. (1985). A comparison of WISC-R and WAIS-R IQs in a mentally retarded residential population. Psychology in the Schools, 22, 392-397.

Thorndike, R. L., Hagen, E. P., & Sattler, J. M. (1986). Stanford-Binet Intelligence Scale: Fourth Edition. Chicago: Riverside.

Wechsler, D. (1974). Manual for the Wechsler Intelligence Scale for Children-Revised. San Antonio: The Psychological Corporation.

Wechsler, D. (1981). Manual for the Wechsler Adult Intelligence Scale-Revised. San Antonio: The Psychological Corporation.

Zimmerman, I. L., Covin, T. M., & Woo-Sam, J. M. (1986). A longitudinal comparison of the WISC-R and WAIS-R. Psychology in the Schools, 23, 148-151.

Content on these pages is copyrighted  by Dumont/Willis © (2001) unless otherwise noted.