Parental interrater reliability as a function of situational specificity and familiarity of target child.

Child behavior ratings were made by 17 mothers and fathers of the same child. Agreement was computed for each parent pair under the following experimental conditions: (1) Ratings of their own child's behavior in general (CBC); (2) ratings of own child based on observations of him/her in a videotaped sample; (3) ratings of an unknown child based on observations of him/her in a videotaped sample; and (4) ratings of the videotaped sample of own child and unknown child using a one minute time-sampling procedure. Parents achieved higher rates of agreement (X = .81) than have previously been reported. Increasing the specificity of the behavior being rated did not significantly affect agreement. Those parent pairs who agreed the most did not necessarily spend a large amount of time in the same kind of situations with their child. Agreement was significantly greater when parents rated their own children's videotaped behavior sample as opposed to that of an unknown child.