'Rich Dad, Poor Lad': How do Large Language Models Contextualize Socioeconomic Factors in College Admission ?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nghiem, Huy, Nguyen-Le, Phuong-Anh, Prindle, John, Rudinger, Rachel, Daumé III, Hal
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909798404980736
author Nghiem, Huy
Nguyen-Le, Phuong-Anh
Prindle, John
Rudinger, Rachel
Daumé III, Hal
author_facet Nghiem, Huy
Nguyen-Le, Phuong-Anh
Prindle, John
Rudinger, Rachel
Daumé III, Hal
contents Large Language Models (LLMs) are increasingly involved in high-stakes domains, yet how they reason about socially sensitive decisions remains underexplored. We present a large-scale audit of LLMs' treatment of socioeconomic status (SES) in college admissions decisions using a novel dual-process framework inspired by cognitive science. Leveraging a synthetic dataset of 30,000 applicant profiles grounded in real-world correlations, we prompt 4 open-source LLMs (Qwen 2, Mistral v0.3, Gemma 2, Llama 3.1) under 2 modes: a fast, decision-only setup (System 1) and a slower, explanation-based setup (System 2). Results from 5 million prompts reveal that LLMs consistently favor low-SES applicants -- even when controlling for academic performance -- and that System 2 amplifies this tendency by explicitly invoking SES as compensatory justification, highlighting both their potential and volatility as decision-makers. We then propose DPAF, a dual-process audit framework to probe LLMs' reasoning behaviors in sensitive applications.
format Preprint
id arxiv_https___arxiv_org_abs_2509_16400
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle 'Rich Dad, Poor Lad': How do Large Language Models Contextualize Socioeconomic Factors in College Admission ?
Nghiem, Huy
Nguyen-Le, Phuong-Anh
Prindle, John
Rudinger, Rachel
Daumé III, Hal
Computation and Language
Computers and Society
Large Language Models (LLMs) are increasingly involved in high-stakes domains, yet how they reason about socially sensitive decisions remains underexplored. We present a large-scale audit of LLMs' treatment of socioeconomic status (SES) in college admissions decisions using a novel dual-process framework inspired by cognitive science. Leveraging a synthetic dataset of 30,000 applicant profiles grounded in real-world correlations, we prompt 4 open-source LLMs (Qwen 2, Mistral v0.3, Gemma 2, Llama 3.1) under 2 modes: a fast, decision-only setup (System 1) and a slower, explanation-based setup (System 2). Results from 5 million prompts reveal that LLMs consistently favor low-SES applicants -- even when controlling for academic performance -- and that System 2 amplifies this tendency by explicitly invoking SES as compensatory justification, highlighting both their potential and volatility as decision-makers. We then propose DPAF, a dual-process audit framework to probe LLMs' reasoning behaviors in sensitive applications.
title 'Rich Dad, Poor Lad': How do Large Language Models Contextualize Socioeconomic Factors in College Admission ?
topic Computation and Language
Computers and Society
url https://arxiv.org/abs/2509.16400