This study compared two methods for classifying preschool-age children's behavior in the Strange Situation procedure, the MacArthur (MAC) and the Preschool Assessment of Attachment (PAA), to determine whether they operationalized converging or diverging approaches to attachment theory. Strange Situations of 306 randomly selected 3-year-old children and their mothers in the NICHD Study of Early Child Care and Youth Development were classified with the MAC and PAA. The methods showed 50% agreement. A block of seven demographic, child and family predictors was unrelated to MAC classifications, but accounted for 19% of the variance in PAA classifications. The MAC and PAA each had associations with some child outcomes in grades 1-5 (ages 6-10) totaling 5% and 12% of the variance respectively, but some of the MAC associations were counter to the hypothesis. The MAC and PAA were sufficiently different to reflect both different classificatory methods and different theoretical understandings of attachment. Results are discussed in terms of limitations of the sample and measures available to compare the two methods, and clinical implications.