The Execution

Part One

There was an easy-to-use ArcGIS version of the census data, but it wasn’t the most recent. So, I went over to the Census Bureau website and downloaded the entire dataset myself. The dataset was quite large, so I trimmed down the data.

After downloading the data and struggling with it for a while I learned that the Census Bureau has an API. I used their API through Python to extract exactly what I needed. By leveraging the Census Bureau API with a simple Python script, I was able to retrieve CSV format data for every city in California. Using pandas, I combined the three different datasets I retrieved from the Census Bureau: income, poverty, and unemployment.

Once I created the CSV file with all three datasets, I imported it into QGIS along with the city boundaries for California. To make the data cohesive, I joined the CSV and boundary data, a process simplified through Python.

Next, I applied an income bracket from the U.S. Census Bureau to a graduated color ramp in QGIS to visualize income levels across California.

I didn’t find any base maps that were easily accessible, so I explored the ArcGIS basemaps and found the ESRI Gray map, which suited my needs. Using the QuickMapServices plugin, I easily imported the ESRI Gray basemap into QGIS.

I wanted to focus on the San Francisco Bay Area where I lived. I used the "California 2020 TIGER/Line shapefile" provided by the Metropolitan Transportation Commission to define the Greater Bay Area. After identifying all the counties within the Bay Area, I dissolved the inner county borders to create one unified outer border for the region.

The product of everything so far is shown on the image to the right.

Part Two:

After the boundaries were fleshed out and data input, I asked myself some questions.

My first question to myself was what was a great visual representation of how terrible the SF Bay Area can be? The answer is straightforward if you have commuted by train in the Bay Area: BART

I input the BART stations as well as where the tracks were. A pattern appeared to me which may come as to no surprise to natives.

It seems as though if BART had a stop in a Bay Area city, it had lower median household income. This was not across the board or definitive but assessing the map would make one believe that BART was some sort of indicator of income.

Part Three: BART Cities vs. No BART Cities

Comparing all SF Bay Area cities that have BART stations against all SF Bay Area cities that do not have BART stations paints a fairly boring outcome.

Stats for SF Bay WITHOUT BART

MEAN: $130,743.41

MEDIAN: $113,839.0

Above is a map of all cities in the SF Bay Area without BART stations.

Above is a map of all cities in SF Bay Area with BART stations.