' REPRESENTATIONAL BEHAVIOR SCALE"

This study was designed to investigate the reliability and validity of the Play, Language and Drawing Representational Behavior Scale (Workman and Owings, 1983). This investigator researched the test-retest reliability, content validity and construct validity of the Play, Language and Drawing Scale. Children from nine months to five years of age were assessed using the scale. Sixteen to eighteen children per stage were observed; the investigator, therefore, was 95 % confident that the estimate of the proportions (p) was within 15 % of the true proportions (P). Four children at each stage were then randomly chosen for a two-week retest. To analyze the data the following formulas or strategies were used for each particular type of reliability or validity: 1) Content valid- ity was established by citing a research reference for each item, by conducting an item analysis for the scale, and by conducting a "goodness of fit " analysis. 2) Construct validity was established by citing a theoretical or research based rationale for each item of the scale. 3) Test-retest reliability was analyzed using the Pearson product-moment correlation and by calculating a reliability percentage (number of matches/number of items). The results indicated that the Play, Language and Drawing Scale is a valid and reliable screening instrument that can be used to assess children from nine months to five years of age in the areas of play, language and drawing. Suggestions for further research with the scale are given