To be or not to be anonymous guides discrimination in online reciprocal feedback

SND-ID: snd1061-1. Version: 1.0. DOI: https://doi.org/10.5878/0w40-2q34

Citation

Creator/Principal investigator(s)

Emma von Essen - Stockholm University, Department of Economics

Research principal

Stockholm University rorId

Description

Theoretical and empirical work highlight the importance of trust and reciprocity in economic exchange. Online reputational feedback systems are crucial in generating trust and this is essential for economic success. Discrimination in feedback may, therefore, create inequality in the long run. We expect competition and selection to eradicate price discrimination and we, therefore, focus on discrimination in feedback. In addition, we look at discrimination and anonymity. Anonymity is a common feature online and is sometimes used as a strategy to circumvent discrimination. We construct a field experiment on eBay, where half of the sellers disclose their names in their usernames while the other half do not. eBay, however, automatically communicates the seller’s names to the buyer after the auction has ended. We find discrimination in feedback, but it only occurs when sellers had anonymous usernames, suggesting that anonymity as a fairness strategy might backfire.

Purpose:

Explore discrimination by gender and foreignness in buyer feedback online and how user anonymity can affect this possible dis

... Show more..
Theoretical and empirical work highlight the importance of trust and reciprocity in economic exchange. Online reputational feedback systems are crucial in generating trust and this is essential for economic success. Discrimination in feedback may, therefore, create inequality in the long run. We expect competition and selection to eradicate price discrimination and we, therefore, focus on discrimination in feedback. In addition, we look at discrimination and anonymity. Anonymity is a common feature online and is sometimes used as a strategy to circumvent discrimination. We construct a field experiment on eBay, where half of the sellers disclose their names in their usernames while the other half do not. eBay, however, automatically communicates the seller’s names to the buyer after the auction has ended. We find discrimination in feedback, but it only occurs when sellers had anonymous usernames, suggesting that anonymity as a fairness strategy might backfire.

Purpose:

Explore discrimination by gender and foreignness in buyer feedback online and how user anonymity can affect this possible discrimination.

We construct a field experiment on eBay, where half of the sellers disclose their names in their usernames while the other half do not. eBay, however, automatically communicates the seller’s names to the buyer after the auction has ended. The seller names signal a combination of gender and foreignness (from a stereotypical Swedish perspective). We manually collected the data. Show less..

Language

Method and outcome

Unit of analysis

Population

Buyers on Swedish eBay

Time Method

Study design

Experimental study

Sampling procedure

Other
The first part of the study used an audit method and the second part of the study used a randomized experiment.

Time period(s) investigated

2012-04-30 – 2012-08-30

Variables

40

Data format / data structure

Data collection
  • Time period(s) for data collection: 2012-04-30 – 2012-08-30
  • Source of the data: Communications, Population group
Geographic coverage

Geographic spread

Geographic location: Sweden

Administrative information

Ethics Review

Stockholm - Ref. 2011/1328-31

Topic and keywords

Research area

Social sciences (Standard för svensk indelning av forskningsämnen 2011)

Economics (CESSDA Topic Classification)

Publications

A matter of transient anonymity: Discrimination by gender
and foreignness in online auctions, in von Essen, E. (2013). Understanding unequal outcomes: Studies on gender, social status and foreignness (Doctoral dissertation, Department of Economics, Stockholm University).

If you have published anything based on these data, please notify us with a reference to your publication(s). If you are responsible for the catalogue entry, you can update the metadata/data description in DORIS.

Versions

Version 1.0. 2020-02-28

Version 1.0: 2020-02-28

DOI: https://doi.org/10.5878/0w40-2q34

Contact for questions about the data

Published: 2020-02-28
Last updated: 2020-02-28