Abstract

LOPEZ-HERRERA, Nelson Ricardo; ACEROS-BUENO, Marlon Augusto  and  LUZARDO-BRICENO, Marianela. Analysis of thefts in Colombia during 2017 using multiple linear regression models and geographically weighted regression. Rev. Crim. [online]. 2019, vol.61, n.3, pp.141-163. ISSN 1794-3108.

According to information from the crime observatory of the National Police of Colombia, thefts from people and of cell phones have shown an upward trend since 2003 (Norza, Peñalosa and Rodríguez, 2017). This trend motivated the carrying out of the current study to analyze the relationship between socioeconomic factors and theft in the different municipalities of Colombia during 2017, through the use of multiple linear regression models and geographically weighted regression using secondary information sources segregated to municipal level. It was validated that variables enrolled in higher education institutions per thousand people, budget per capita allocated by the general system of participations and the category of the municipality account for 69,5% of the variability of the logarithm of theft from individuals and of cellphones in 532 municipalities using a globally estimated multiple linear regression model and 50,16% using the geographically weighted regression model omitting in the latter the category of the municipality. In this model there were slight variations in the coefficients at the municipal level, reflecting that the social and economic heterogeneity has effects on indicators of theft nationwide.

Keywords : Factors of crime; theft; criminal statistics; crime measuring; Statistics; GWR; OLS.

        · abstract in Spanish | Portuguese     · text in Spanish     · Spanish ( pdf )