Infant Mortality in Mexico in 2020: A Spatial Analysis of Multiple Causes

Abstract

The objective of this paper is to analyze the spatial distribution of infant mortality in small geographic areas and to quantify the effect of a set of cofactors that are theoretically associated with the behavior of infant mortality, such as access to health services, lack of access to water in private housing, female schooling, and population size. The study began with the application of an exploratory spatial data analysis whose main purpose is to identify a series of spatial grouping patterns. The methodology used involved fitting seven regression models, six spatial regression models that configure three types of endogenous, exogenous, and error term spatial interaction, and one simple linear regression model (used as a basis for comparison). The selection criterion used to choose the model that best fits the infant mortality data was the Akaike Information Criterion (AIC), which provides a measure of the quality of the model fit. The results indicate that the most important direct effect on the spatial behavior of infant mortality comes from the average years of female schooling since the average increase of one year of education would imply a reduction of 0.076 percentage points in the municipal infant mortality rate. Similar results are observed for the variables of lack of access to water and percentage of municipal population.