Inequity and College Applications: Assessing Differences and Disparities in Letters of Recommendation from School Counselors with Natural Language Processing
Letters of recommendation are affected by inequality.
Using natural language processing techniques, we analyzed counselor letters of recommendation from 600,000 college applications submitted through the Common Application platform. We found large and noteworthy naïve differences in letter length and content across nearly all demographic groups (e.g., many more sentences about Athletics among White and higher-SES students, longer letters and more sentences on Personal Qualities for private school students). Findings reflect the importance of reading letters and overall applications in the context of structural opportunity.