Application Essays and Characters in Higher Education Admissions

Session Type: Coordinated Paper Session

Abstract

The recent rejection of race-based affirmative action from the Supreme Court presented new challenges to U.S. higher education, besides challenges related to the use of quantitative (e.g. high school grades) and qualitative (e.g., application essays and recommendation letters) information. These challenges have a complex impact on the diversity and success rate of college students. In this session, we will explore possible answers to these challenges from multiple angles. The first paper examines the application essays’ content in its relationship with demographic and socio-economic background variables, admissions status, and post-admissions success indicators (such as first semester GPA), based on a large pool of applications for a public university. The second paper investigates AI’s role in application essay preparation and its evaluation in admissions, exploring possible biases associated with the use of large language models in both processes. The third paper shares findings in terms of using behavioral skills assessment to assist with college and graduate admissions, and its implications for diversity, equity, and post-admissions success. Together, these three papers would provide new ideas and information to further the discussion and exploration of a valid and fair admissions process for U.S. higher education.

Sub Unit

• Individual Papers, Innovation Demonstrations, Coordinated Sessions, or Organized Discussions

Session Organizer

• Guangming Ling, EDUCATIONAL TESTING SERVICE

Chair

• Ou Lydia Liu, ETS

Individual Presentations

• 1. Evaluating Application Essays’ Content in Shaping College Admissions Decisions and College Success - Guangming Ling, EDUCATIONAL TESTING SERVICE; Michael Flor, EDUCATIONAL TESTING SERVICE; sugene Cho-Baker, ETS

• 2. The Evolution of AI Bias: Large Language Models and Holistic Review - jinsook Lee, Cornell University; Aaron Jordan Alvero; Thorsten Joachims, Cornell University; Rene Kizilecec, Cornell University

• 3. Using a behavioral assessment to assist graduate admissions: Some initial evidence - Patrick Charles Kyllonen, ETS; Daniel Fishtein, Educational Testing Service

Discussant

• Li Cai