One of the most important considerations in designing clinical infant research studies is the selection of reliable and valid measurement procedures. Few measures of caregiver-child interactions have been studied with newborns, particularly premature infants. The main objective of this study was to examine psychometric properties of the National Institute of Child Health and Human Development (NICHD) Mother-Child Interaction Qualitative Ratings in a sample of premature infants and their mothers to evaluate its use in the neonatal intensive care unit. Mother-baby dyads (N = 24) were videotaped in a 10-min interaction in the NICU. Nine raters independently assessed dyadic interactions using the NICHD Mother-Child Interaction Qualitative Ratings in a fully crossed research design. Rater reliability was strong for mother and infant ratings (0.76-0.94). Scores yielded normal distributions for maternal sensitivity, positive regard, and flatness of affect and skewed distributions for maternal intrusiveness, detachment, negative regard, and all child ratings. Positive maternal behaviors correlated positively with one another and negatively with negative maternal behaviors. Thus, preliminary analyses suggest that scores obtained using the NICHD Mother-Child Interaction Qualitative Ratings with premature babies and their mothers in the neonatal intensive care unit demonstrate adequate interrater reliability, and distributional properties provide preliminary evidence of face validity.