Sampling, over-representation of social categories and the sexual division of labour: formulating a hypothesis about the double work (in Spanish)

Published in Empiria, 2010, volume 20: 135-161.

Objective: The main interest of this article is to explain one of the most common incidences that occur during the drawing of random samples: over-representation of social categories. Method: In that sense, the population sample obtained as a result of the accomplishment of the ERP’s study has been compared with EPA’s population sample. An important women’s over-representation has been localized in ERP’s population sample, as well other socio-demographic differences. In that sense, an explanatory hypothesis of the differences was suggested. Results: Taking into account the developed analysis we could conclude that the root of women’s over-representation in ERP’s population sample was 1) caused, from a methodological point of view, by definition of units of analysis and units of sampling and by interviewing techniques used to interview and capture these units; 2) and made possible, from a societal point of view, by gendered division of labour.