|
LeTendre, D., Spiker, D., & Constantine, N. A. (1992). Establishing a ceiling on the Bayley Scales of Infant Development at 24 months. In L. Lipsitt & C. K. Rovee-Collier (Eds.), Advances in Infancy Research: Vol. 7 (pp. 187-198). Norwood, NJ: Ablex. The Bayley Scales of Infant Development (BSID) are used widely in clinical and research contexts to assess the developmental functioning of infants. Norms are provided by the publisher for infants and toddlers from 2 months through 30 months of age (Bayley, 1969). Longitudinal and comparative research studies of full-term and preterm infants often include the BSID as the primary assessment instrument through 24 months. As with most individually administered developmental and intelligence tests, the BSID employs "discontinue criteria" for the examiner to use in determining the examinee's ceiling level, that is, determining when to stop administering items. A ceiling is defined for the BSID as the item representing the most difficult success for that examinee. The ceiling item is identified by means of a discontinue criterion (BSID Manual, 1969, p. 29) of the lowest 10 consecutive failed items. That is, the item just below this series of 10 consecutive failures is, by definition, the ceiling item. For the Motor Scale, the recommended discontinue criterion is six consecutive failed items. According to the manual, "these levels must be carefully established to ensure that the infant's full range of successful functioning has been tested" (1969, p. 29). The use of this type of discontinue criterion is based on two assumptions: first, that the items are sequentially ordered by difficulty level; and second, that after the specified number of failures the infant would be unlikely to achieve any more correct responses. Clinical experience suggests that many 24-month-olds are able to pass one or more of the last 10 items on the Mental Scale - thus effectively precluding the establishment of an empirical ceiling according to Bayley's recommended criterion for discontinuation. For at least some of these infants, it is possible that the test has not adequately sampled their behavior (that is, the infant's "full range of successful functioning" has not been tested). The extent of this problem and the functioning range of the infants most likely to be affected are not known. This study examines the extent to which the full range of successful functioning of 24-month-olds is adequately sampled by the BSID. More specifically, the degree to which the discontinuation criteria are not met can be seen as the degree to which the BSID does not adequately sample the full range of successful functioning of 24-month-olds. This poses the possibility that the BSID many not fully delineate the full range of individual differences among 24-month-olds. To investigate this issue, the current study employs a highly standardized research administration of the BSID to a large multi-site sample of 24-month corrected age low-birthweight (LBW) preterm infants. Because preterm and LBW infants typically function below the level of their full-term counterparts (see Pape, Bunice, Ashby, & Fitzhardinge, 1978; Field, Dempsey, & Shuman, 1979; Teberg et. Al., 1982; Eckerman, Sturm, & Gross, 1985; Rose & Wallace, 1985; Ford, et al., 1985), the sample assessed here should provide a conservative overall estimate of problems in meeting the ten-failure discontinue criterion on the Mental Scale. |