INTEROBSERVER AGREEMENT IN MEASURING SOCIAL DEVELOPMENT IN YOUNG CHILDREN

Abstract

The study was intended to determine interobserver agreement for studying social interaction among young children in a natural setting. Social interaction was categorized into parallel, associative, and cooperative play. The observational setting included a sandbox with miniature toys. The sample of pre and primary school-level children was drawn from two private schools located in a medium-sized city. Forty children, ranging in age from four to seven years were selected on a voluntary basis. Three students from The University of Arizona along with the investigator served as observers. The student observers received training from the investigator prior to the commencement of the study. The observed data was scored to include both frequency and time spent on each category. Two procedures for computing interobserver agreement were utilized. These procedures yielded results indicating interobserver agreement ranging from 68% to 90% agreement. These results indicated adequate reliability of the observational procedures used in the study. In addition to reliability of observational procedures the findings also indicated that the observed frequency of parallel play was highest for subjects of all ages, whereas the observed frequency of cooperative play was found to be the lowest for all subjects. Additionally findings indicated that there were no significant differences in the difficulty of observing any category of social interaction of children of varying ages. Educational and social implications of the findings were discussed.