I do pro bono GIS work from time to time, and a recent contribution was towards a local school that was undergoing a capital campaign to build science and computer labs to enhance the quality of education at that elementary school. A non-profit, this school needed to raise funds that exceeded the ability of the parents giving capability. This meant reaching out to local area foundations for financial support to supplement the fundraising campaign.
The school in question is located immediately adjacent to an affluent area in Los Angeles. The academic reputation, relative low tuition rate, and the centrality of the school to many of the local employment areas of Los Angeles, draws students from many areas around Los Angeles County. This means, that despite the location of the school next to some of the most expensive neighborhoods in Los Angeles, many of the students that attend come from lower income families. The school’s development efforts ran into an obstacle when local foundations rejected the school’s proposal for funds under the impression that the school’s parents should be able to contribute more than the development plan was outlining. Local foundations, in assessing the economic need of the school, were pulling median household income data based on the ZIP code the school was located in. By calculating the supposed economic background of the school at the ZIP code level, the significantly higher than average incomes of the richer areas was pulling that income number higher than what was represented by the actual economic backgrounds of the school’s students.
Since (as is the case with most schools) income information was not requested of parents, the school needed a secondary way to provide a more meaningful analysis of the student’s economic backgrounds. When I met with the person in charge of fundraising at the school who outline this dilemma to me, I proposed analyzing the economic background of the school’s students based on their residency using GIS analysis as a way of providing a more accurate understanding of the school’s background than simply looking at the ZIP code of the school’s location.
To do this, the residential addresses for all the students were geocoded. Duplicate addresses (i.e. families with more than one student at the school) were removed so as to not skew the results. Mapping out the location of the students also had an added benefit of providing a better understanding geographically of where students commuted from. The school’s student population originated from different economic areas around the Los Angeles County area. The closest student lived a mere 900 feet from the school while the farthest student resided in the northern LA County city of Santa Clarita, commuting a distance of 31 miles. The overall average distance student’s commute to attend the school was about three miles.
The next step was to apply the median household income to each student’s location based on the census block group that student’s family resided in. To do so, I turned to Esri Data which, as part of its GIS data offerings, provides demographic estimates and projection data at the Census block group level. (Side note: Esri recently commissioned a study by Cropper GIS to assess the accuracy of its demgraphic forecasting data. In a blind study comparison with four other demographic firms, Esri’s data was evaluated by a team of demographic experts consisting of Jerome N. McKibben, PhD, McKibben Demographic Research; David A. Swanson, PhD, University of California, Riverside; Jeff Tayman, PhD, University of California, San Diego; and Matthew Cropper, GISP, Cropper GIS. Esri’s data was judged to be the most accurate. Esri also published an article about groundtruthing its demographics data in the October 2010 edition of ArcUser.)
Estimated 2011 median household income data at the Census block group level from Esri Data was used to derive all economic statistics used in the GIS analysis. The Census block group is the smallest geographic unit created by the U.S. Census Bureau for which sample data is available (the data collected and statistically calculated from a smaller subgroup of residents). The economics of the Los Angeles County region are disparate with estimated 2011 median household income ranges from a low of $8,644 to a high of $200,001. The cumulative average 2011 median household income for Los Angeles County is estimated at $56,881. This aligns with the United States Department of Agriculture’s Economic Research Service’s 2010 Los Angeles County Median Household Income estimate of $52,595 (when factoring in for an annual cost of living increase). Understanding the median household income across the county allowed for a benchmark against which to compare the relative economic backgrounds of the students.
Using the identify function, the median household income value for each of the block groups within which a student family resided was applied to student location layer. From there, some basic calculations were used to understand the economic background of the students that attended this school. The average median household income was calculated by totaling the median household income for each block group in which a student resides and dividing by the number of student households. To understand the income population for the student households, the median household income was broken into categories representing a proportion of the mean median household income (see the map below for the breakdown). This allowed the school to see the spread of income across the student population (See the graph further down in this article).
Median household incomes for census block groups where students resided ranged from $21,487 to the maximum calculated block group value of $200,001, showing a great range in the economic background of the students and that the school served a population that was very diverse in its socio-economic makeup. The average median household income for all students was $75,994. This average was significantly less than the $134,662 median household income figure being used by local foundations based on the ZIP code of the school. The calculated median household income for the student populated was 43% less than the ZIP code based assessment. Furthermore, this analysis as able to demonstrate that almost 41% of students came from neighborhoods with a median household income less than the median household income for Los Angeles County. The chart below shows that two income populations are predominate at the school; a large proportion of lower income students, a dip in population for middle income, and then a rise again for higher income students.
A report was prepared presenting updated economic data including a breakdown showing the distribution of median household income for the students. A map showing the distribution of the students overlain onto a thematic map showing the median household income as compared to the overall average for the county was also provided to further visualize the data. Through the report, the school was able to present a convincing argument to local foundations that despite the affluent location of the school, the economic background of the students merited need-based donations.
Like this article?
Sign up for GIS Lounge's weekly newsletter for more great content: