Inspiration
It is no secret that insulin prices have been increasing with big pharma for years now. It has been a daily struggle for years for people with type I diabetes to access their medications affordably. These factors may have changed now. Due to this issue, our team wanted to analyze the correlation of diabetes with different influences such as race and gender. Our team also wanted to see the correlations between type I diabetes treatment cost and dosages depending on the type of insurance the patient has throughout the years. Studying these trends can help us understand the continuing issues from 2000-2008 to now.
What it does
In Python, we spent an ample amount of time cleaning, filtering, and sorting the data and variables of importance. Once we were able to drop unnecessary columns we grouped the variables to generate a contingency table and then a Chi-Squared test, as we used two categorical variables. Additionally, we made a bar chart in Excel. Considering that we had little background in coding and experienced technical issues, these processes took a while but we each learned valuable lessons from loading a csv file into github to appending columns in a dataset in Python.
Challenges we ran into
This was our first hackathon, and we all had minimal experience with data science, so we struggled with finding a direction to take our data analysis in, and settling on a tech stack to use. We were also challenged to learn new technologies in a very short period of time, so it was difficult creating functional code and visualizations.
What we learned
Everyone on the teams came from a different background with different levels of strength. We all had different experiences in coding. Some were more comfortable with Python, and others were more comfortable with SAS. We were struggling with using statistics to help us break down the information. Our computers were giving us many issues as well. We were having trouble setting up and updating everything necessary. Looking past all that, we worked together to solve these issues. It was a phenomenal learning method. We learned a lot about coding and using it in real-life situations. It allowed us to bring our skills out. We all worked with each other to gain experience. While learning from each other, we all learned a lot about type I diabetes and its different correlations. We had a little idea, but this project allowed us to expand our knowledge.
Log in or sign up for Devpost to join the conversation.