That is, 1 or 2 deficits out or 9 or 10 subareas is an easier criterion to meet than 1 or 2 deficits out of 4 or 5 subareas.

Although the committee searched, no further studies of the classification agreement between recently published adaptive behavior measures and IQ were found. The authors of other adaptive behavior measures have not provided data of this nature, even though the current and previous editions of the Joint Association Test Standards (Joint Committee on Standards for Educational and Psychological Testing of the American Educational Research Association, the American Psychological Association, and the National Council on Measurement in Education, 1985, 1999) require that test publishers provide validity data supporting the recommended uses of a test.

Simulations of IQ and Adaptive Behavior Classification Agreement

Because of the dearth of data examining the relationship between various levels of IQ and scores on adaptive behavior measures, particularly for individuals in the mild mental retardation range, the committee used Monte Carlo models to project the proportion of people who would be expected to have IQ and adaptive behavior domain scores in specific ranges. These Monte Carlo models used correlations of adaptive behavior domain scores with one another and with IQ (Thompson, 2001). The simulations used the best of the currently available adaptive behavior measures, applying assumptions that enhanced the likelihood of classification agreement between IQ and adaptive behavior. For example, high-end estimates of the relationship between adaptive and intellectual functioning were applied as well as internal consistency reliability estimates, rather than stability coefficients, making classification agreement slightly more likely. Projections of the proportions of adult cases meeting IQ and adaptive behavior cutoff scores were developed from the Monte Carlo analyses and are summarized in Tables 5-3 and 5-4. These projections were based on test manual data on the correlations of domain scores with one another and with IQ. The Monte Carlo results answer the question: Given an IQ in the range of mental retardation, how many cases will



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement