Álvaro Chirino2026-03-222026-03-222017https://dialnet.unirioja.es/servlet/oaiart?codigo=6230470https://andeanlibrary.org/handle/123456789/66447The national census of population and housing in Bolivia of 2012 proposed a change in the type of question regarding indigenous self-identificaion with respect to the 2001 census, official results showed a reduction in self-identification by 20% between the two censuses. Social analysts argue that this reduction is due to a cultural component and a media phenomenon that was experienced during the development of the last census regarding the concept of mestizo. However, almost nobody links this reduction to the change in question. This paper presents the results of an experimental design applied to 218 college students, which conducted a survey of study habits, that served as a placebo to introduce the treatment. The sample was randomly separated into two groups, the treatment group received a survey containing the question posed in the 2012 census, and the control group received the question from the 2001 census. Logit and probit models were estimated to find the causal effect of the change in question, the results show that there is a reduction of 12% in the indicator of self-identification because of the change in the question. This results implies that the questions asked in the Census 2001 and 2012 are not comparable, at least in groups with similar characteristics to the experimental group.esHumanitiesGeographyIndigenousarticle