Embedded Files
Summary
Summary
India, the second most populated country with more than 1.3 billion people, contributes a massive amount towards counts of infant and under five mortality. Prior studies have examined determinants of infant and child mortality-- mainly, socioeconomic and environmental risk factors. However, all of these studies have failed to take into account the important effects of geographic space. Prior studies have assumed that the relationship between the surveyed risk factors and infant and under-five mortality are homogenous through space, which is known to be false. For example, northern states have much higher levels of infant mortality rate than their southern counterparts. Ultimately, this paper displays intra-state and inter-regional variance in regard to infant and under-five mortality. Additionally, using multiple geospatial techniques, the authors depict that those regions with negative rates of the measured exposure variables (wealth, female literacy, child nutrition, or safe delivery) faced higher rates of infant and under-five mortality.
data
data
All the data for this study came from the National Family Health Survey (NFHS, 1992–1993 and 1998–1999) and the second round of the Reproductive and Child Health-District Level Household Survey (DLHS-II, 2002–2004). These are large demographic survey studies which aid in the creation of a comprehensive demographic and health database for India, helping city and state officials in implementation and monitoring of population and health programs throughout the country.
geospatial techniques
geospatial techniques
Moran's I
Moran's I
Spatial Autocorrelation technique
Measures the degree to which data points are either similar or dissimilar to their spatial neighbors
Values can be positive or negative, ranging from -1 to 1
Positive values indicate a positive autocorrelation (points with similar values are closely distributed) whereas negative values indicate a negative spatial correlation (neighboring points are more dissimilar)
-1 indicates perfect dispersion, 1 indicates perfect correlation, 0 is a random spatial pattern
Local Indicators of Spatial Association (LISA)
Local Indicators of Spatial Association (LISA)
Univariate: measures the correlation of neighborhood values around a selected spatial location
Bivariate: measures the correlation between a variable and a weighted average of another variable
Spatial Regression
Spatial Regression
Evaluates clustering of a variable not explained by independent variables
Error term measures influence of unmeasured independent variables
All images were taken from the study
Page updated
Report abuse