Using mothers versus trained observers in assessing children's secure base behavior: theoretical and methodological considerations.

The Attachment Q-Set (AQS) has emerged as a psychometrically sound method for assessing young children's secure base behavior in the home. However, considerable disagreement exists about whether mothers versus trained observers should be used as AQS sorters. The present study examined associations between mothers' and trained observers' AQS sorts for preschoolers, and assessed mother-observer concordance in relation to observers' confidence about how representative the behavioral samples they witnessed were of the domain of AQS items. Mothers with careful training and supervision on the AQS system completed AQS sorts with regard to their children's current behavior, and the same children were assessed with the AQS during a 2-3 hour visit 1-2 weeks later by trained, "blind" observers. Trained observers provided a confidence rating regarding the degree to which the samples of behavior observed were representative of the universe of AQS items. Mothers' and observers' sorts were significantly intercorrelated; however, observer sorts converged with mother sorts as observers' confidence ratings increased. Results are discussed in relation to circumstances that affect mother-observer reliability with the AQS and to factors that should be weighed when considering whether to use mothers versus trained observers as sorters.