Census Data


Using Census Data to identify Uninsured Minority Population in the Mecklenburg County


Problem

Mecklenburg County is the largest county in North Carolina. Approximately 30% of the county population is African American. North Carolina did not expand Medicaid, which could have allowed substantial subsidy to residents earning up to 138% of the Federal Poverty Level. The Mecklenburg County Commissioners is considering providing health care to uninsured African Americans in the county and would like to know where the hot spots of uninsured African American residents are located in the county.

Analysis Procedure

My overall strategy was to use American FactFinder to select data on health insurance coverage and race using the 2017 ACS estimates (5-year) from Mecklenburg County, NC. The attribute of interest is the percentage of Black or African American Uninsured. I will join the attribute data to the Mecklenburg County, NC census tracts shapefile and conduct hot spot analysis (Getis-Ord Gi*) using ArcMap.

From the American FactFinder website, I limited the search to the Mecklenburg County, NC by selecting all census tracts from Geography. I searched and downloaded table S2701. After importing the CSV file into excel, I removed decimals from the first three columns. The attribute of interest was HC05_EST_VC25 – Percent Uninsured; Estimate; RACE AND HISPANIC OR LATINO ORIGIN – Black or African American alone. Next, I downloaded the 2017 NC census tracts shapefile and imported it into ArcMap. Using Select by Attribute – NC Mecklenburg county FIPS code = 37119 (“COUNTYFP” = ‘119’) I extracted and saved the Mecklenburg county shapefile. I added Table S2701 to ArcMap. Since the linking attribute field (GEOID) in the Mecklenburg county shapefile was a text field I created an integer field before joining table S2701. There were three census tracts (9801, 9802, 9803) in the Mecklenburg county that contained null values. I excluded them leaving a total of 230 census tracts. I developed a thematic map of the percentage of uninsured African American and conducted hot spot analysis (Getis-Ord Gi*) in ArcMap. I used NAD_1983_StatePlane_North_Carolina_FIPS_3200 Projected Coordinate System for the maps.

Flowchart


Results

Map 1. Percentage of uninsured African Americans in Mecklenburg County, NC – American Community Survey, 2014


Map 2. Hotspot analysis of uninsured African Americans in Mecklenburg County, NC – American Community Survey, 2014


Application and Reflection

In recent years, homicide rates have surged in Mecklenburg County. The Charlotte-Mecklenburg County police department (CMPD) wants to know the areas where the homicides are concentrated. It is hypothesized that homicides are mostly occurring in low-income neighborhoods. A list of homicides occurring in the past three years is available. The addresses will be geocoded. The table (S1701) of the percentage of residents living under the poverty level will be downloaded from the American Find Finder and after making necessary edits will be joined to the county shapefile. Choropleth map of the attribute percent below the poverty level (HC03_EST_VC01) will be generated and the homicide point data will be displayed using graduated symbols. Finally, Hot spot analysis (Getis-Ord Gi*) will be performed in ArcMap.