Minimizing gender differences in children's interest assessment: Development of the Inventory of Children's Activities-3 (ICA-3)

The focus of this study was on revising the Inventory of Children's Activities-Revised (ICA-R; Tracey & Ward, 1998) to enhance its psychometric properties while minimizing gender differences in scale scores. The original 30 ICA-R items and an additional 30 items were administered to 70,280 fifth-eighth grades students. The original scoring was compared to a revised scoring method based solely on the empirically best items and a scoring method balancing empirical scoring with minimizing gender differences. All three item sets (original, empirical, and combined empirical/gender balancing) resulted in strong internal consistency estimates and adequate fit to the circular structure, yet the combined empirical/gender method had much lower gender differences especially for the scales measuring Investigative and Social interests. The implications of using the revised scale with children is discussed.