2

I'm new using ArcMap and I have a GDB of schools (and its enrollment data) at block level. What I need to do is to find the places with the most concentration of enrollment in a radius of 5km. Since I'm new using this tool I'm completely lost.

  • 1
    Do you have an instructor that can help you? – Vince Feb 3 '16 at 15:39
  • 1
    what defines the "places"? and what are the "blocks"? Are these census blocks? This question is missing some background information. Please also see the tour gis.stackexchange.com/tour – jbchurchill Feb 3 '16 at 16:19
  • Thank you for your response, indeed, the "places" I want to find are at block level (it's a census of all the schools built in Perú). Therefore each dot in the map is a school, provided with enrollment information. So the main purpose of this study is to assess where are zones or hot spots with high density of students, using as a main criteria a radius of 5km. – jcrespo83 Feb 4 '16 at 16:10
3

I would suggest reading up on spatial statistics. In particular, I would look at spatial autocorrelation and clusters. The ArcGIS Resource page on the Spatial Statistics toolbox might be a good place to start.

1

As you suggested in a comment to the question, convert each enrollment value in the GDB to an enrollment density by dividing it by the block's area. Convert this layer to a grid. Compute its focal sum using a 5 km radius. Multiply that, if you like, by the area of each grid cell (equal to the square of the cellsize). Notice that this cancels out the preliminary division-by-area: you have now produced a grid whose cell values are total enrollment within a 5 km radius. Highlight the cells with the largest values: these are your "hot spots."

You will need access to a raster-based GIS, such as Spatial Analyst, to carry this out. Open-source solutions exist in GRASS, R, Python, and other platforms.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.