MEASURES OF CHILDRENS SEX TYPING IN MIDDLE CHILDHOOD

Measures of sex-typing suitable for use with 8- to 14-year-olds are proposed. They involve interests, toys, sports, household jobs, friends, similarity to peers and personality. The measures fit conceptually into Huston's (1983) framework within the categories ''Behavioural enactment or adoption'' and ''Identity or self-perception''. The sample comprised first and second born children from 191 families (first borns - 98 males, 93 females; second borns - 95 males, 96 females). They were selected using a stratified random sampling procedure involving area risk scores. The children in 161 families were assessed on a second occasion, on average 16 months after the first testing. The measures were shown to have suitable psychometric properties: sex differences in the expected directions, moderate coefficient alpha values, consistency over time in terms of moderate correlations, and few changes in means. Intercorrelations among the measures revealed little coherence either within each of the two Huston categories or among the measures in general. These results support a multidimensional view of sex-typing and draw attention to limitations of studies which use single measures of this variable.