Do parents and children know each other? A study about agreement on personality within families

Abstract

Se estudiaron las relaciones entre autoevaluaciones y heteroevaluaciones de cuestionarios de personalidad en contextos familiares españoles. 336 matrimonios evaluaron la personalidad de uno de sus hijos. Al mismo tiempo, cada hijo se autoevaluaba. Se aplicaron los marcadores de Goldberg. En una segunda muestra, 120 estudiantes universitarios evaluaron la personalidad de ambos padres. Al mismo tiempo, ambos padres se autoevaluaron. Se aplicaron los marcadores de Goldberg y el NEO-PI-R. Se replica el grado de acuerdo encontrado en otras culturas, siendo las correlaciones entre autoevaluaciones y heteroevaluaciones de alrededor de 0.40, y con tamaños del efecto de las diferencias de medias no grandes. Se observa que: 1) los hijos son más acertados que los padres; 2) el grado de acuerdo es más alto para el NEO-PI-R que para los marcadores de Goldberg, especialmente en el factor de Apertura a la Experiencia; y 3) la variable sexo no afectó a los resultados. Se discuten las implicaciones y las limitaciones del presente estudio.

The relationships between self-report and observer-rating versions of personality questionnaires within families in the Spanish cultural context were studied. In the first sample, 336 couples rated the personality of one of their children. The children also assessed their own personality. Goldberg’s adjectives were administered. In the second sample, 120 university students rated the personality of their parents. Both parents also assessed their own personality. In this case, Goldberg’s adjectives and the NEO-PI-R were administered. Results replicate the level of agreement found in other cultural contexts. Correlations were around .40, and the effect size of the mean differences was not large. Children were more accurate than parents, larger agreement coefficients were found for the NEO-PI-R than for Goldberg’s adjectives in Sample 2, especially for the Openness factor, and no substantial role for the sex variable was found in either sample. Implications and limitations of the present study are discussed.