The objective of this study is to examine preschool-age correlates of the maternal version of the Attachment Q-Set (AQS) (Waters & Deane, 1985) in order to provide validity data. Concurrent associations between the Attachment Q-Set and measures of separation-reunion attachment classifications (Cassidy & Marvin, 1992), quality of mother-child interaction, maternal stress, and child externalizing behavior problems were examined for a low-risk sample of 152 preschool children. Results showed that, even after controlling for maternal stress, ambivalent and disorganized children had lower scores on the AQS when compared with the secure group. However, avoidant and controlling children did not differ from secure children in maternal sorts. Significant associations between the maternal Q-Set and affective quality of mother-child interaction were also maintained, when controlling for maternal stress. The AQS was significantly associated with both mother and teacher reported externalizing behavior problem reports, but the association with teacher reports was not maintained when maternal stress was entered in the equation as a covariate. The discussion focuses on the validity of the maternal Q-Set as a measure of child attachment behavior in preschoolers. (Contains 3 tables.)