Investigating factors of food deserts

Inspiration

Over 17 million Americans live in food deserts—low-income areas where limited access to fresh, healthy food forces residents to choose between unhealthy fast food or traveling long distances for groceries, perpetuating cycles of poor nutrition and declining health.

What it does

We investigated the correlations of several factors with food insecurity, allowing us to gain a better understanding about the reasons behind the difficult access to food sources.

How we built it

Looking over the data, we noted down several variables that we believed would have a strong correlation to food insecurity. To test if this was indeed the case, we took these columns and combined them into one dataset for easier ease of access. After cleaning this dataset, we applied a weighted least squares model on the data in order to calculate the correlation coefficients between the test variables and food insecurity.

Challenges we ran into

Having no background in data science before coming into this competition, one of the biggest challenges was having to learn how to use the tools and libraries used for data manipulation. We also discovered the importance of cleaning and combining the multiple datasets properly so that there would not be any duplicate values within the rows, otherwise we would run into further errors when creating the model.

Accomplishments that we're proud of

We are proud of being able to extract information from the dataset given to us to create a model that was able to give us information on what the large factors at play are in the presence of food deserts.

What we learned

From this project we learned the steps needed to take a large dataset, cleanse it, identify variables of interest, and then create a model for the data. From the model that we created, we learned that one of the largest factors relating to food insecurity is diabetes rates.

What's next for Investigating factors of food deserts

Our model on different factors of food deserts is promising in identifying what variable are most highly correlated to food insecurity. To improve the model, we could input more variables into it or test out other types of regression, which could yield some interesting results.