Lucy Jane Miller

Age-Related Effects on the Predictive Validity of the Miller Assessment for Preschoolers:

This study examined the predictive validity of the newly developed Miller Assessment for Preschoolers (MAP) for age-graded groups. A 4-year prediction interval was used, and criterion measures were the WISC-R and the Woodcock-Johnson Psycho-Educational Battery. Stepwise multiple regressions were used to assess the value of individual MAP subscales as predictor variables. In addition, correlations between the MAP Total Score and criterion measures were calculated for all six age categories specified for the MAP. Results are discussed in terms of the instability of developmental status among young children and the need to assess skill domains that are relatively stable at an early age.

A Summary of Three Predictive Studies with the MAP

Recently, threestudies exploring the predictive validity of the Miller Assessment for Preschoolers (MAP) (Miller, 1982) have been completed. Cohn (1986) studied 134Colorado childrenafter an interval of 1.25 to 2.5 years. Lemerand (1985) studied 273Michigan childrenafteran interval ofone year. Miller (1986) studied a national sample of 338 children after an interval of four years. All three studies examined the MAPs recommended 5% and 25% cutpoints. Cohn (1986) employed t tests to analyze the relationship between MAP scores and later criterion measures; Lemerand (1985) compared MAP predictions to the base rate of reponed occurrence in the population; and Miller (1986) examined correlations of the MAP to later criterion measures.

Interpreting Evidence of Predictive Validity for Developmental Screening Tests

Evidence of predictive validity is an essential basis for the selection and use of a screening instrument. In the area of developmental screening such information is typically assessed by two methods: correlational analysis and classificational analysis. This paper describes each methodology and reviews their relative advantages and disadvantages in relation to the early identification of developmental handicaps and risks.

Neuromaturational Variables within the Miller Assessment for Preschoolers

The need for preschool screening to identify at-risk children is generally well accepted; however, the determination of which developmental or behavioral domains are the best predictors of later school performance remains a controversy in the professional literature and clinicalpractice. Althoughthere is fairly uniform professionalagreement regardingseveral of the behavioral domainssuch as verbalskills and cognitive abilities, there is controversyregarding the usefulness of neuromaturational variables (sensory integration, neurodevelopmental abilities, and spatialand sequential organization). Someprofessionalsview neuromaturational factors as having little or nothing to do with learning (Engelmann, 1969;Wong, 1979), whereas others contend that the samefactors provide the buildingblocks for successful learning in school (deQuiros, 1978; Knickerbocker, 1980). The Miller Assessment for Preschoolers (MAP) (Miller, 1982) is a standardized screeninginstrument in which neuromaturationalvariables compose one of five developmental indices (subtests), the Foundations Index. The purpose of this study was to analyze the statisticalcontribution