Smiles in Profiles: Improving Fairness and Efficiency Using Estimates of User Preferences in Online Marketplaces

How to cite

Palikot, Emil; Athey, Susan; Karlan, Dean; Yuan, Yuan. 2022. "Smiles in Profiles: Improving Fairness and Efficiency Using Estimates of User Preferences in Online Marketplaces." AEA RCT Registry. AEARCTR-0010030. https://doi.org/10.1257/rct.10030-1.0

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

Emil Palikot* , Susan Athey , Dean Karlan , Yuan Yuan

* Primary investigator

Registry ID
AEARCTR-0010030
DOI
https://doi.org/10.1257/rct.10030-1.0
First registered
Last updated
Status
Completed
Trial dates
2021-12-05 – 2021-12-25
Countries
United States of America ()
JEL codes
J710, L1, C9
Registry record
http://www.socialscienceregistry.org/trials/10030

Abstract

Online platforms often face challenges being both fair (i.e., non-discriminatory) and efficient (i.e., maximizing revenue). Using computer vision algorithms and observational data from a micro-lending marketplace, we find that choices made by borrowers creating online profiles impact both of these objectives. We further support this conclusion with a web-based randomized survey experiment. In the experiment, we create profile images using Generative Adversarial Networks that differ in a specific feature and estimate it's impact on lender demand. We then counterfactually evaluate alternative platform policies and identify particular approaches to influencing the changeable profile photo features that can ameliorate the fairness-efficiency tension.

Intervention

In the intervention, subjects are shown six pairs of images generated using a computer vision algorithm and are asked to pick the preferred image in the pair. Images randomly vary by the gender of the person shown in the image, and whether the person smiles or not and the image is a body-shot or not.

Experimental design

Subjects recruited on Proflic.co receive one of fifteen predefined protocols. Protocols consists of six pairs of images in which gender of the person in the image and whether the person smiles or not and whether the image is a body-shot or not, is randomly assigned. All images are generated using a computer vision algorithm. Subjects are asked to choose one of the profiles in each pair. Thus, a treatment is a pair of images a subject is choosing from. Additionally, we include three attention checks; first, we ask for the reason people participate in microfinance; second, for the explanation of the choice decision, and third, for the occupation of the person shown in the previous slide.

Design details

Randomization method
Randomization is made by computer.
Randomization unit
The unit of the randomization is an experimental session. Experimental sessions are choice instances in which a subject chooses between two profiles.
Clusters
2460 experimental sessions
Observations
2460 experimental sessions
Arms
A treatment arm is defined as a choice instance in which each profile varies in three binary dimensions. In result, we have 64 treatments each of the size 37.
Primary outcomes
The primary outcome is whether the subject choose a profile or not.

Cited by

No citation data available for this registration yet.

Keywords