Does discrimination impede access to college? A Field Experiment

How to cite

Rocha, Hugo Vaca Pereira; Melo, Vitor; Sigaud, Liam. 2024. "Does discrimination impede access to college? A Field Experiment." AEA RCT Registry. AEARCTR-0013713. https://doi.org/10.1257/rct.13713-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Hugo Vaca Pereira Rocha* , Vitor Melo , Liam Sigaud

* Primary investigator

Registry ID
AEARCTR-0013713
DOI
https://doi.org/10.1257/rct.13713-1.0
First registered
Last updated
Status
Completed
Trial dates
2023-10-01 – 2024-02-10
Countries
United States of America ()
JEL codes
I23, J71, I20
Registry record
http://www.socialscienceregistry.org/trials/13713

Abstract

The long-running debate about affirmative action in college admissions intensified in 2023 after the Supreme Court ruled that race should not affect admissions decisions. We examine the extent and mechanisms by which race affects the admissions process after the 2023 ruling. We explore evidence from a field experiment where fictitious applicants request application fee waivers from nearly all admissions counselors at US universities. We find that Black and Asian applicants are much more likely to have their requests rejected than White applicants.
Asian applicants are also 30% less likely to receive a waiver than a white applicant. We find that the a majority of the discrimination comes from very selective universities.

Intervention

The intervention in the experiment involved sending emails from fictitious applicants to nearly all admissions counselors at four-year universities in the United States. These emails requested application fee waivers and were designed to assess the impact of race on the likelihood of receiving a waiver. The fictitious applicants had racially indicative names (White, Black, Asian, Hispanic) to measure differential treatment based on perceived race.

Experimental design

The experimental design of the study involved a randomized email audit study aimed at investigating the impact of racially-indicative names on the behavior of admissions officers and U.S. colleges related to the granting of application fee waivers.

Design details

Randomization method
Email Messages: Each sampled email address received a message from a fictitious male college applicant requesting information and fee waivers due to financial hardship. Templates and Names: There were eight email templates designed to signal two levels of applicant quality (low and high), and originating names that belong to one of five racial/ethnic groups (White, Chinese, White/Chinese, Hispanic, and Black). Random Assignment: Within each college, both the racial/ethnic identity of the fictitious applicant and the email template were randomized without replacement to ensure variation both within and between institutions.
Randomization unit
The randomization unit in this experiment was the admissions counselor at each university. Each individual counselor received one email from a fictitious applicant. The randomization was done at the level of individual emails sent to these counselors, ensuring that within each university, the race/ethnicity of the applicant and the quality of the email were randomly assigned to each counselor without replacement.
Clusters
0
Observations
4,658
Arms
4,658 unique admissions email addresses at 1,294 distinct institutions
Primary outcomes
1. De Facto Rejections 2. Waiver Approval Rates 3. Free Application Disclosure 4. Requests for More Information 5. Response Rate 6. Sentiment Analysis of Responses 7. Use of Applicant’s Name in Responses

Cited by

No citation data available for this registration yet.

Keywords