In 2020, Mexico’s National Institute of Statistics and Geography (INEGI) began working with the IDEAMAPS Network to apply our frameworks and approaches to deprived area mapping. INEGI has developed a data pipeline to integrate dozens of recent datasets with hundreds of indicators – from the national government, local municipalities, and outside sources - into a standard 600x600m grid system which enables very localized estimates of household poverty and area-level deprivation. Each indicator is classified according to the IDEAMAPS Domains of Deprivation framework which characterizes nine general themes that define deprivation in any given neighbourhood and city.
INEGI’s data pipeline integrates diverse spatial datasets into one harmonized dataset in one of three ways. (1) Point and line data, such as schools or road locations, are aggregated (e.g., as counts, sums, averages, etc.) to 600x600m grid cells. (2) Similarly, data in small polygons (areas), such as city blocks, are aggregated in the same way based on the polygon centroid (geographic centre). For data associated with larger polygons such as municipalities, each 600x600m cell is assigned the value of the polygon in which it is located based on the cell’s centroid.
The data pipeline includes four main stages. Data are stored in their native formats in a Data Lake. Kedro is used to process data with one of the above three methods into a single multilayer geographic dataset with metadata which enables speedy queries of the data on a highly responsive web platform. A web tool was created to explore the data, where D3.js is used to visualise the 600x600m grid cells as a map and generate interactive plots.
INEGI’s web platform is still in Alpha phase, and only available to INEGI users at the moment, though there are plans for the platform, its visuals, and the underlying gridded data layers to become publicly available. The simple maps and data plots provide insights about the relationship of indicators at fine geographic scale. For example, the graph below shows correlations between high proportion of ingenious households, fewer years of education, and higher population crowding.
Next steps are to model deprived areas and specific domains of deprivation. The INEGI IDEAMAPS outputs will support local governments to plan and prioritise development initiatives, and identify and engage vulnerable communities for more targeted upgrading.
