Inspiration
We were inspired to create this solution because of the impact we found outside factors to have on students' education. Many of these factors are out of the students' control and we would like to help level the playing field and make it easier for students to learn and succeed no matter their background.
What it does
The goal of this project is to reduce the amount of external influence on a young student by limiting their exposure to harmful impacts. This is done by keeping the student at school doing fun activities where it is safe and there is supervision. The idea being that when at school and under supervision it's harder for anything bad to happen but when outside of school there is a lack of supervision and then anything could happen, which could not only affect the students ability to learn, but harm them aswell.
How we built it
We performed a lot of Exploratory Data Analysis to build the foundation for what we wanted to study. We were able to see clear correlation between certain ethnicities, economic disadvantage, and crime. We decided to further prove our findings of correlation through a random forest where we were able to confirm the importance of these features.
Challenges we ran into
We ran in to multiple challenges when doing this project. This data had many missing values so we aggregated the data based on all grades instead of each grade individually. With this, we had to make the assumption that all grades tests are scored the same and their distributions were equal.
Accomplishments that we're proud of
Besides confirming that there is a strong correlation between test scores and factors such as crime rates, poverty and chronic absenteeism, we implemented a random forest regression model that showed that economic status is the largest predictor of poor test scores.
What we learned
We learned many new things about New York City and its education system. Prior to this we had only heard or read about the awful segregation and gentrification in New York but it was astonishing how bad it was when we plotted the data. Another thing we learned was the impact home economic standing can have on test performance. This makes sense when you think about the lack of food or steady living arrangements could have on your ability to think.
What's next for Empowering Students in NYC
The next step would be to conduct an experiment with random low performing schools to see if we are able to find causation in our policy and better test scores. If we do find proof, we will bring this study to the City and the National Government in hopes of getting legislation passed.
Built With
- geopandas
- geoplot
- matplotlib
- pandas
- python
- scikit-learn
- seaborn
Log in or sign up for Devpost to join the conversation.