Part One: Context & Problem Statement
Research Background
Disclaimer: This series of articles are from the perspective of an admissions team at a “Small Pond” university, whose job is to raise awareness of the school, and persuade prospective university students to enroll. In the context of this research paper, a “Small Pond” school is defined as a safety net school for students (FAQ, 2022). At a “Small Pond” university, a student is able to become a “Big Fish” — a top student who is able to utilize many of the college’s resources.
In this paper, a “Small Pond” university offers the same curriculum as a more prestigious university. The school generally has a higher acceptance rate than its prestigious counterparts, so it attracts students with a wide disparity of GPA and test scores.
This web article explores the possibility of using machine learning technology to optimize marketing to a diverse high school student population, given the constraints of an admission team’s limited budget. The goal of each “Small Pond” university is to maximize attendance at the university, while prioritizing high-achieving students to boost the school’s academic capabilities.
However, the team operates in a world with limited resources. Schools with smaller endowments cannot compete with their peers in terms of academic popularity or capital. In fact, researchers conclude that a student who attends the most selective university in the state can expected postgraduate earnings to increase by about 20%, particularly for white men (Hoekstra, 2009). Similarly, students have heard their whole lives about universities like Harvard and Stanford, and students chase the status of these schools reinforced by parents, peers on campus, and the outside world (Binder & Abel). Universities all over the globe strategically brainstorm opportunities to target high school seniors that their wealthier, high-prestige rivals do not.
This is an ongoing five part series: Read on to learn about the different chapters creating and analyzing this project!
Previous:
Introduction: “Luring Big Fish to Your Small Pond”
Current:
Part One: Context & Problem Statement
Next:
Part Two: Reason for Models and Data Collection
Part Three: Machine Learning Solution 1: Logistic Regression
Part Four: Machine Learning Solution 2: K-Means Cluster Analysis
