Team members: Andrew Larimer, Deepak Nagaraj, Daniel Olmstead, Mike Winton

From the Kaggle Data Science for Good project statement:

PASSNYC and its partners provide outreach services that improve the chances of students taking the SHSAT and receiving placements in these specialized high schools. The current process of identifying schools is effective, but PASSNYC could have an even greater impact with a more informed, granular approach to quantifying the potential for outreach at a given school. Proxies that have been good indicators of these types of schools include data on English Language Learners, Students with Disabilities, Students on Free/Reduced Lunch, and Students with Temporary Housing.

Part of this challenge is to assess the needs of students by using publicly available data to quantify the challenges they face in taking the SHSAT. The best solutions will enable PASSNYC to identify the schools where minority and underserved students stand to gain the most from services like after school programs, test preparation, mentoring, or resources for parents.

Submissions for the Main Prize Track will be judged based on the following general criteria:

• Performance - How well does the solution match schools and the needs of students to PASSNYC services? PASSNYC will not be able to live test every submission, so a strong entry will clearly articulate why it is effective at tackling the problem.

• Influential - The PASSNYC team wants to put the winning submissions to work quickly. Therefore a good entry will be easy to understand and will enable PASSNYC to convince stakeholders where services are needed the most.

• Shareable - PASSNYC works with over 60 partner organizations to offer services such as test preparation, tutoring, mentoring, extracurricular programs, educational consultants, community and student groups, trade associations, and more. Winning submissions will be able to provide convincing insights to a wide subset of these organizations.

More on PASSNYC.

Our analysis is performed in a series of notebooks, with the primary notebook being final_project_overview.ipynb. Links to supporting notebooks for data cleaning and ML model development are all included in that notebook.