Inspiration
I recently watch news coverage that indicated a significantly disproportionate death rate among African Americans due to Coronavirus. In addition, this program pointed to socio economic factors being behind this inequality. I was shocked to see this impact and this made me want to explore the reason behind this.
What it does
It does automatic analysis of datasets to find a correlating relationship across educational factors and economic disparity by race over the past ten years.
How I built it
First, I found datasets on census.gov to identify which ones I could use for the analysis. Next I converted into consistent formats for programmatic processing. I installed the environment necessary for creating a python HTTP API and web services. Finally, I implemented the code and tested using sample data, then ran it on the full datasets. For easy usability, I then modified and polished the website.
Challenges I ran into
A challenge I faced with creating this program was cleaning up the format of data sets. Each dataset used was formatted differently and in order to make the code most efficient and easiest to use, when converting any files I had to clean them up. Another challenge was trying to find the right statistics in the same time period.
Accomplishments that I'm proud of
I was able to analyze large amounts of data and have it support an original hypothesis. The hypothesis I formed was also supported by research papers from the National Bureau of Economic Research. This was also my first hackathon and it was great to see how much I could get working in a limited time.
What I learned
Coding-wise I learned how to format an API using python and HTML to make it usable and readable. I additionally learned a lot about income inequality and the wide ranging social impact education can have.
What's next for Income Trend Analysis Across the US
In the future, I hope to make my code more generic so it’s easier to use and potentially connect my API to those of census.gov. I also want to look at more datasets involving other factors to see if there are any other ways we can help mitigate the crisis, such as access to healthcare. Finally, I would like to look at what countries have done that have brought down income inequality to see if any of the actions are applicable to the US, do predictive modeling with ML API such as tensorflow to see what the impact could be on improving this in the US. I hope to find more data to research and continue to find trends and possible solutions.
Log in or sign up for Devpost to join the conversation.