Summary of Bias Analysis Of Ai Models For Undergraduate Student Admissions, by Kelly Van Busum and Shiaofen Fang
Bias Analysis of AI Models for Undergraduate Student Admissions
by Kelly Van Busum, Shiaofen Fang
First submitted to arxiv on: 3 Dec 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper presents a comprehensive study on bias detection and mitigation in AI predictive models. The authors developed an AI model using six years’ worth of admissions data to analyze variables affecting student admission decisions at a large urban research university. They found that the decision to exclude standardized test scores led to changes in the demographics of admitted students. To detect biases, they evaluated the predictive models against three sensitive population variables: gender, race, and first-generation college attendees. The analysis revealed persistent biases, which were further analyzed using fairness metrics. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making sure AI decisions are fair. Researchers created a computer model to predict student admission at a big university. They used data from six years and found that when test scores aren’t required, the demographics of admitted students change. The study also looked for biases in these predictions against gender, race, and family background. The results showed that some biases persisted. The researchers also discussed ways to measure fairness. |
