ABSTRACT
La Teoría del Sexismo Ambivalente propone que el sexismo es un constructo multidimensional compuesto por dos tipos de creencias sexistas, unas de carácter hostil y otras de tipo benevolente. El objetivo principal de este estudio fue analizar la estructura factorial del instrumento derivado de esta teoría, su invarianza factorial entre géneros y su relación con la violencia de pareja. Mediante un formulario en internet, 745 participantes de ambos sexos de población general de Argentina, con edades comprendidas entre 18 y 45 años, respondieron el Inventario de Sexismo Ambivalente (ISA) (Glick & Fiske, 1996). El ISA demostró puntuaciones y una estructura factorial invariantes entre hombres y mujeres. La relación entre las puntuaciones en el ISA y la violencia de pareja varió en función del tipo de violencia considerado (amenazas vs. violencia física) y el género del miembro de la pareja. Las puntuaciones en Sexismo Hostil fueron mayores en las mujeres que ejercieron violencia física comparadas con otras mujeres, pero no en los hombres. También fueron superiores en quienes sufrieron esta conducta, tanto hombres como mujeres. Se discuten las implicancias de estos resultados a la luz de estudios previos que emplean el ISA, principalmente en población de habla hispana.
The Ambivalent Sexism Theory proposes that sexism is a multidimensional construct composed of two types of sexist beliefs, on the one hand, hostile and, on the other hand, benevolent. The first refers to negative attitudes towards women as weak and inferior to men, and the second refers to the set of sexist attitudes towards women considering them stereotyped and limited to certain roles (i.e. mother, wife, and housewife). The main objective of this study was to analyse the factor structure and factorial invariance by gender of the instrument derived from this theory and its concurrent validity with a measure of intimate partner physical violence. Another objective of the study was to establish comparisons in ISA scores by gender and cross-culturally. The Ambivalent Sexism Inventory (ASI) (Glick & Fiske, 1996) was answered through an online form by 745 participants of both sexes from an Argentinean general population sample aged between 18 and 45 years old. By means of a confirmatory maximum likelihood factor analysis the data in this sample showed a good adjustment to the ASI structure proposed by its authors, this is a one-dimensional factor (SH) and three subfactors (protective paternalism, complementary gender differentiation and heterosexual intimacy) that make up the second-order factor SB. This factorial structure remained invariant when discriminating between men and women. The correlation between both factors was robust and significant (r = .68) both in the general sample, and differentiating by gender. The reliability of the factors and subfactors was between a moderate and high level (α = .62 to .91). No significant differences were found between men and women in the mean scores of the ASI factors and subfactors. Approximately 1 in 3 subjects reported having perpetrated or suffered intimate partner physical violence. In general terms, those who have perpetrated or have suffered violence had higher scores in SH and SB that those who did not, but this effect was only significant in the case of women. The relationships found between ASI scores and intimate partner physical violence varied by gender and violence directionality (suffered or perpetrated). The ASI scores were lower in countries close to Argentina probably due to a sample mainly composed by young women with a high education level. Our results support the idea that ambivalent sexism is an invariable transcultural construct between men and women, and that the ISA is avalid and reliable measure in the Argentine population. The limitations of results are discussed and the implications are analyses in light of previous studies using the ASI, mainly in Spanish-speaking population.
ABSTRACT
La teoría de los cinco grandes factores (FFM) constituye actualmente uno de los modelos teóricos de mayor aceptación en el estudio de la personalidad. En argentina se cuenta con la versión española de un instrumento que permite medir los 5 grandes factores, el NEO-PIR y su versión abreviada, el NEO-FFI. El presente trabajo se propuso realizar un análisis factorial confirmatorio en una muestra de 396 estudiantes universitarios argentinos a los fines de evaluar la estructura factorial del NEO-FFI. Asimismo se complementó el análisis con métodos de parcelización de ítems y con bootstrap. Se observó que el modelo construido con parcelas estadísticas y que contempla a los factores como oblicuos, si bien presentaron los mejores índices de ajuste en comparación al resto de los modelos evaluados, no fueron óptimas. Finalmente se sugieren estudios adicionales para mejorar el funcionamiento de algunos ítems, especialmente los del factor Amabilidad.
A teoria dos cinco grandes fatores (FFM) constitui atualmente um dos modelos teóricos de maior aceitação no estudo da personalidade. Na Argentina, se usa a versão espanhola de um instrumento que permite medir os 5 grandes fatores, o NEOPIR e sua versão abreviada, o NEO-FFI. O presente trabalho se propôs a realizar uma análise fatorial confirmatória em uma amostra de 396 estudantes universitários argentinos para avaliar a estrutura fatorial do NEO-FFI. A análise foi complementada com métodos de parcelização de itens e com bootstrap. Observou-se que o modelo construído com parcelas estatísticas e que contempla os fatores como oblíquos, embora tenha apresentado os melhores índices de ajuste em comparação ao resto dos modelos avaliados, não foram ótimas. Finalmente, são sugeridos estudos adicionais para melhorar o funcionamento de alguns itens, especialmente os do fator Amabilidade.
The Five Factor Model represents currently one of the biggest developments in the study of the personality. In Argentina, there is the spanish version of an instrument which allows assess the five big factors, the NEO-PIR and his abbreviated version, the NEO-FFI. The present study has proposed to carry out a confirmatory factor analysis on a sample of 396 argentine university students to evaluate the factor structure of the NEO-FFI. Also the analysis was complemented with item parceling methods and bootstrap. It was observed that the model built with statistics parcels and that considers the factors as oblique, although having higher rates of adjustment, these are not optimal. Finally, further studies are suggested to improve the performance of some items, especially those of the Amiability factor.