Market Outcomes under Different Recommender Systems: Experimental Evidence from an Online Dating Service

Last registered on May 20, 2019
View Trial History

Pre-Trial

Trial Information

General Information

Title
Market Outcomes under Different Recommender Systems: Experimental Evidence from an Online Dating Service
RCT ID
AEARCTR-0004211
Initial registration date
May 16, 2019

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
May 20, 2019, 1:41 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country
Taiwan, Province of China
Region

Primary Investigator

Name
Yuwei Hsieh
Affiliation
USC
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Ming-Jen Lin
PI Affiliation
National Taiwan University
Contact Investigator
PI Name
Kuan-Ming Chen
PI Affiliation
University of Chicago
Contact Investigator

Additional Trial Information

Status
In development
Start date
2019-05-27
End date
2019-07-31
Keywords
Firms & Productivity
Additional Keywords
recommender system, two-sided market, online dating, e-commerce, assortative matching
JEL code(s)
Secondary IDs
Abstract
We partner with an online dating service to study the market outcomes and consumer behaviors under different recommender systems. There are two treatment groups: The first treatment group receives the recommendation based on the mate preference estimated from the existing database. The second treatment group receives the recommendation based on the same preference estimation, adjusting for the matching cost implied by a structural matching model. The control group receives a randomized recommendation. We measure several outcomes: consumer satisfactory (from a survey after the experimental period), click-through rate, response rate, time to find a dating partner, cross-sectional distribution of the number of likes, degree, and centrality of the dating bipartite graph.
External Link(s)

Registration Citation

Citation
Chen, Kuan-Ming, Yuwei Hsieh and Ming-Jen Lin. 2019. "Market Outcomes under Different Recommender Systems: Experimental Evidence from an Online Dating Service." AEA RCT Registry. May 20. https://doi.org/10.1257/rct.4211-1.0
Former Citation
Chen, Kuan-Ming, Yuwei Hsieh and Ming-Jen Lin. 2019. "Market Outcomes under Different Recommender Systems: Experimental Evidence from an Online Dating Service." AEA RCT Registry. May 20. https://www.socialscienceregistry.org/trials/4211/history/46837
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2019-06-01
Intervention End Date
2019-06-30

Primary Outcomes

Primary Outcomes (end points)
click-through rate or hit rate
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
time to find a dating partner
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
We divide our sample into different groups. Control group implements the default algorithm, while the treatment group implements our recommendation system.
Experimental Design Details
Randomization Method
Randomization done in office by a computer
Randomization Unit
Individual
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
6,000 individuals
Sample size: planned number of observations
6,000 individuals
Sample size (or number of clusters) by treatment arms
2000 control, 2000 machine-learning treatment, 2000 machine-learning and the matching model treatment
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number
Analysis Plan

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials