Salud Pública de México (Jan 2003)
Validación de un cuestionario para el estudio del comportamiento sexual, social y corporal, de adolescentes escolares Validity and reliability of an instrument for assessing the social, and corporal sexual behaviors among school adolescents
Abstract
OBJETIVO: Validar un cuestionario autoadministrado de comportamientos sexuales, sociales y corporales para adolescentes por nivel escolar y género. MATERIAL Y MÉTODOS: Instrumento de 26 preguntas construido según DeVellis, aplicado en un estudio longitudinal prospectivo, hecho en 1994 a 1 243 alumnos y alumnas de secundaria y preparatoria en Guadalajara, Jalisco, México. Validez de constructo establecida por análisis factorial exploratorio y la confiabilidad por la prueba Alfa de Cronbach. RESULTADOS: Aparecen cinco factores en hombres usando valores propios (eigenvalue) >1 y cinco en mujeres, forzando la mejor solución. Veinte factores explican entre 59.3 y 70.6% de varianza, con promedio de pesos factoriales de 0.63- 0.75, y con 13 factores diferentes. El promedio de Alfa de Cronbach fue de 0.87-0.93 y en 16 factores hubo valores de Alpha de 0.71 o más. CONCLUSIONES: Cuestionario para poblaciones equivalentes con validez de constructo y confiabilidad interna aceptables.OBJECTIVE: To validate a self-administered questionnaire about social and corporal sexual behaviors among adolescents, by school level and gender. MATERIAL AND METHODS: The data collection instrument was constructed according to DeVellis and consisted of 26 items. It was used in a prospective longitudinal study conducted in 1994 in Guadalajara, Mexico, among 1243 secondary and high school students. The construct validity of the instrument was established using exploratory factorial analysis, and its reliability using the Cronbach's alpha test. RESULTS: Five factors were obtained in males and an equal number in women, using eigenvalues >1 and by forcing the best solution. The explained variance was 59.3-70.6%; the factorial weight average was 0.63-0.75, and 13 different factors were selected. The Cronbach's alpha mean was 0.87-0.93 and 16 factors had Cronbach's alpha over 0.70. CONCLUSIONS: The instrumen has an acceptable validity and reliability for studying the sexual, social, and corporal behaviors in adolescents similar populations.
