With the available census tables (county, tract, block, zcta) and shapefiles, is it possible to identify "hotspots" of population concentration within a ZCTA?

The ZCTA gazeteer file provides the population density per square meter and square mile, but what I've been asked to do is, essentially, to put a "stick-pin" wherever there are dense clusters of people within a ZCTA, using the census data.

Is that even possible with the census data that is available? I am new to this subject but it seems the census block is the smallest granule, is that right? If the block's shape were to be spatially intersected with the ZCTA's shape, the ratio of the intersection in square meters to the block's total square meters could be used to estimate the population density of the intersection, assuming an even distribution of people throughout the block. But isn't that an unwarranted assumption?

  • 1
    Yes, your thinking is correct. Check out Modifiable Area Unit Problems, gives a quick run down of exactly what you are thinking about. – ed.hank Jul 22 '17 at 14:13
  • ncbi.nlm.nih.gov/pmc/articles/PMC1762013 "ZCTAs are compiled based on census block topology. In the generation of a ZCTA, each underlying block is assigned one, and only one, ZCTA code – regardless of its location. Therefore, it is possible for blocks to straddle more than one ZCTA or ZIP code. This can be problematic when aggregating population data to both units." – Tᴚoɯɐuo Jul 22 '17 at 14:28

Your Answer

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

Browse other questions tagged or ask your own question.