Resumo

LAMFRE, Laura; PERREN, Joaquín  e  BRAMARDI, Sergio. Conditioned multiple correspondence analysis. An application to the study of the quality of life according to social class in Argentina at the beginning of the millennium. SaberEs [online]. 2023, vol.15, n.2, pp.157-175. ISSN 1852-4418.

Abstract. The Conditional Multiple Correspondence Analysis (cMCA) is a descriptive method that makes possible to jointly analyze the stable trends that appear in several categorical data tables, based on a qualitative variable that works as a conditioning form. Its empirical application in the construction of a composite quality of life indicator (CQLI) is presented as a challenge. This paper studies, from a theoretical and methodological point of view, the Conditional Multiple Correspondence Analysis (cMCA) technique and implement it in the construction of a CQLI conditioned by social stratum. We worked with the microdata base of the Permanent Household Survey for the 31 agglomerates of Argentina in the 2nd quarter of 2010. The results were analyzed comparatively between the classical Multiple Correspondence Analysis (MCA) to the original data set, the MCA to each of the seven social stratum and the cMCA that incorporates the social stratum as a conditioning variable. This let us find similarities and differences between these three methodological alternatives. Finally, the construction of a CQLI and its comparative analysis by social stratum enable to deepen the study of the Argentine social structure in relation to the relevant dimensions of quality of life.

Palavras-chave : Multivariate Analysis; Social Structure; Quality of Life Indicator.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )