The purpose of this study was to test the reliability and validity of an instrument designed to measure the dramatic and sociodramatic play activity of young children. A total of 31 children were selected from early education programs on the basis of sex and socioeconomic status. The spontaneous classroom dramatic play behavior of each child was videotaped and subsequently rated using a play measurement instrument developed for the study. Demographic correlates and intercorrelations among 11 behavior items in the play rating instrument indicated that the instrument was highly reliable. Its construct validity in relation to external variables was only partially confirmed. As hypothesized, most items correlated significantly and positively with socioeconomic status, overall dramatic play quantity, and the ratio of sociodramatic play quantity to overall dramatic play quantity. Only two of the items correlated positively with age. Sex differences were found on only one item. Findings on sex differences were contrary to predictions. Furthermore, items of the instrument did not cluster precisely as predicted. Two basic factors were identified and labelled 'sociodramatic play items' and 'individual dramatic play items.' Further research aimed at identifying relationships between play scores and various cognitive variables is recommended for the purpose of further validating the rating system. (Auhtor/RH)