Career Maturity in Adolescence: Reliability and Validity of Self-Ratings of Abilities by Gender and Ethnicity

Abstract

To obtain reliable and valid self-ratings of abilities of adolescents, a self-rating instrument was designed to incorporate six of Maybe and Wests' (1982) measurement conditions. The instrument was administered to 210 ninth-graders. Test-retest reliability coefficients are mostly in the .70s and .80s for self-ratings and accuracy of self- ratings. Coefficient alphas and test-retest correlations for composite scores range from .74 to .90. For the total sample of students, the major findings are: (a) Correlations between self-rated abilities and measured abilities range from the .30s to the .60s, with medians in the .50s; (b) Students overrated all of their abilities except scholastic aptitude and Space Relations; (c) Students were most accurate in rating their scholastic aptitude and least accurate in rating their special aptitudes; (d) and self-ratings of some abilities were inaccurate, although they were reliable and substantially correlated (.63, .61) with measured ability. Ethnic differences were found on all ability scores, all self-ratings, and all accuracy of self-rating scores. Gender differences were found on some special abilities, some self-ratings, and some accuracy of self-rating scores. The self-rating instrument does not show evidence of differential validity across different ethnic and gender groups.