Effects of repeated assessment on standardized test performance by infants.

Infant-intervention programs that evaluate their effectiveness by repeatedly testing infants with standardized tests may confound intervention effects with repeated-testing effects. Further, maternal participation in testing may increase infants' test scores, either directly by giving infants practice with test-like items at home between test administrations or indirectly by helping infants adapt to the test setting. Thirty-five infants were tested at 3-month intervals between 4 and 28 months of age with the Bayley Scales of Infant Development and at 31 months of age with the Stanford-Binet; 25 infants were tested with the Bayley at 4 months and the Stanford-Binet at 31 months. Mothers of some infants in each group participated in the testing process. Between-subjects analysis on the Stanford-Binet revealed no effects for repeated testing or maternal participation. Within-subjects analyses for repeatedly tested groups demonstrated that maternal presence during testing, but not repeated testing, was associated with significantly higher Bayley scores. It was concluded that maternal presence is a potent influence on Bayley performance, that repeated experience with the Bayley biased neither performance on another standardized test of development nor subsequent Bayley performance, and that infant intervention programs are not likely to confound their evaluation designs by testing infants repeatedly.