Studies of racial and ethnic differentials in birth outcomes require accurate identification of the race and ethnicity of infants in the study samples. This research was part of a study that uses California Birth Cohort data to investigate racial and ethnic differences in mortality and health care costs for infants weighing less than 1,500 grams when born. The research evaluates five alternative methods for coding infants’ race and ethnicity using birth certificate information on the parents’ backgrounds. The coding rules differ in how they code infants with parents of different racial or ethnic backgrounds. The rules were found to yield similar measures of infant’s race and ethnicity and infant mortality because all but a relatively small fraction of parents were of the same background. Changes over time in parents’ self-reports of ethnicity and in coding rules were found to be important analytic issues.