Single versus repeated assessments of infant abilities using the Bayley scales of infant development

An analysis of the score differences and characteristics achieved by 9- and 15-month-old infants tested twice with the Bayley Scales of Infant Development, one week apart, shows marked variations between what could be considered optimal estimates of ability and estimates of ability based on actual scores. There is approximately a 50% chance that an infant's actual Developmental Quotient from a single examination will differ from its optimal score by more than one standard deviation of the BSID. The fact that single test scores do not reliably estimate “true” capabilities reconfirms that a caveat be maintained concerning the use of single Bayley scores as baselines of longitudinal research or as outcome measures, but holds out the possibility that the usefulness of such scores as predictive of outcome measures can be improved over its present negligible status through repeated assessments.