ABSTRACT
This study examines how perceptions of residential reputation and stigma vary according to differences in social class and spatial structures. The focus of our inquiry is the city of Santiago, Chile – a highly segregated city in Latin America. We suggest that residential stigma is the focus of a larger assemblage of material and non-material marginalization; it is a point of convergence of multiple forms of social stratification. We combine a representative survey of 600 residents with geo-referenced census data, building composite measures to represent concrete class and spatial structures using clustering techniques. Results reveal that living in more affluent segregated areas is associated with a better perceived residential reputation, whereas perceived residential stigma is greater in geographical areas with a high concentration of poverty, land devaluation, population density, and overcrowding, regardless of whether these areas present high levels of criminality. Interestingly, when we include social class along with the spatial structure, the former loses significance as a predictor of both perceived residential reputation and stigma. The spatial structure therefore goes some way toward mediating the relationship between class and perceived residential reputation. This suggests the need for public policies designed to reduce the current symbolic domination that exists in Santiago.
Acknowledgments
We are grateful for support from the Centre for Social Conflict and Cohesion Studies – COES.
Disclosure statement
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Notes
1. El Mercurio, a right-wing newspaper owned by one of the wealthiest families in the country.
2. Some studies report the names of the neighborhoods in question (Cornejo, Citation2012; Lunecke, Citation2016) but others prefer to protect their anonymity in order to avoid reproduction of stigma in academic productions (Risør & Arteaga, Citation2018).
3. In our sample, all variations in individuals’ perceptions of residential reputation and stigma at census tract level were significant, meaning that all likelihood ratio tests comparing the null model with the logistic regression were statistically significant (p < .000). All represent substantial proportions; thus, the use of multilevel logistic regression is appropriate for the analysis and highly recommended in our case.
4. Land values in the Chilean real estate sector are generally expressed as multiples of a measure known as the Unidad de Fomento (UF). Its value is equivalent to approximately 40 USD at time of publication.
5. Theil’s entropy index is often used to measure income distribution. Values closer to 1 represent segregation/concentration. In our case, values were multiplied by −1 for zones predominantly populated by low and poor SES groups.