Introduccion : el prejuicio sexual (homofobia) cobro mayor atencion en las ciencias biomedicas y sociales en el marco de la epidemia por VIH. La Escala para Homofobia (EHF) es uno de los instrumentos mas breves para la cuantificacion de este constructo. Objetivo : observar algunos indicadores de validez y confiabilidad de EHF en estudi­antes de medicina de una universidad de Bucaramanga, Colombia. Metodo : estudio de validacion. Participaron 366 estudiantes entre 18 y 30 anos de edad (M=20.1, DE=1.9). El grupo femenino represento 59.1% de la participacion. Se estimaron medidas de validez (analisis factorial exploratorio, correlaciones para validez convergente y divergente y funcionamiento diferencial de items segun sexo) y de con­fiabilidad (alfa de Cronbach y omega de McDonald). Resultado : se observo un factor que explico el 43.8% de la varianza, alta validez con­vergente y divergente (r=0.82 y 0.03, respectivamente), y un funcionamiento diferen­cial de items sin impacto por sexo. Alfa de Cronbach 0.78 y omega de McDonald 0.79. Conclusion : EHF presenta alta validez de constructo, convergente, divergente, sin fun­cionamiento diferencial de los items segun sexo y buena confiabilidad tipo consistencia interna. Rev.cienc.biomed. 2014;5(2):287-294. PALABRAS CLAVE Homofobia; Validez de las pruebas; Reproducibilidad de resultados; Estudiantes de me­dicina; Estudios de validacion. SUMMARY Introduction: sexual prejudice (homophobia) received major attention in the area of the biomedical and social sciences in the frame of the epidemic for HIV. The scale for homophobia (EHF) is one of the briefest instruments for the quantification of this con­struct in epidemiological researches. Objetive: to observe some indicators of validity and reliability of the EHF in students of medicine of a university of Bucaramanga, Colombia. Methods: a study of validity was carried out. 366 students participated, with ages be­tween 18 and 30 years (M=20.1 SD=1.9). The female group represented 59.1% of the participation. There were estimated measures of validity (Exploratory factor analysis, correlations for convergent/divergent validity and differential item functioning accord­ing to sex) and of reliability measures (Cronbach’s alpha and McDonald’s omega). Results: A factor that explained the 43.8% of the variance, high convergent and di­vergent validity (r=0.82 and 0.03, respectively) and a differential functioning of items without impact for sex. Cronbach’s alpha was 0.78 and McDonalds’ omega, 0.79. Conclusions: EHF presents high quality of convergent and divergent construct without differential functioning of the items according to sex and good reliability of type internal consistency. It is necessary to know the psychometric performance of this instrument in other contexts in Colombia . Rev.cienc.biomed.2014;5(2):287-294. KEYWORDS Homophobia; Validity of test; Reproducibility of results; Medical students; Validation studies.