Gender Differences in Negotiation – An Exploration in Virtual Reality

How to cite

Sajons, Gwendolin B.; Tinsley, Catherine H.. 2019. "Gender Differences in Negotiation – An Exploration in Virtual Reality." AEA RCT Registry. AEARCTR-0003996. https://doi.org/10.1257/rct.3996-1.0

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

Gwendolin B. Sajons* , Catherine H. Tinsley

* Primary investigator

Registry ID
AEARCTR-0003996
DOI
https://doi.org/10.1257/rct.3996-1.0
First registered
Last updated
Status
In development
Trial dates
2019-03-18 – 2019-04-05
Countries
Private
JEL codes
J16, J24, J71, C81, C72, C91
Registry record
http://www.socialscienceregistry.org/trials/3996

Abstract

Prior research has suggested gender differences in negotiation outcomes, particularly when parties are advocating for themselves in a distributive negotiation context. On average, male negotiators walk away with significantly more value than female negotiators. Two possible explanations for this phenomenon have been suggested. On the one hand, women may behave differently (e.g., less assertively) in negotiations and thus obtain inferior outcomes. On the other hand, negotiators may behave differently (e.g., more assertively) towards women, leading to inferior outcomes for them. To parse out these explanations we conduct an experiment in virtual reality. Participants are assigned avatars of their true or the other gender, while assuming that they and their negotiation partner are in fact represented by an avatar of their true gender. This allows exploring gender differences in own behavior and in the other party’s behavior towards each gender. Behavioral differences will be tracked all along the negotiation process (initial offers, concession rate, negotiation outcomes).

Experimental design

We will conduct a laboratory experiment with about 400 student participants at Georgetown University from March 18th to April 5th 2019. Students participate for course credit and will additionally earn money as a function of their negotiation performance.

In the experiment, participants will be randomly assigned to take the role of a seller (group A) or a buyer (group B) of an antique sugar bowl. Each participant negotiates through a virtual character with another student whom they meet at a negotiation table in a virtual room. Participants communicate via a chat function. Before entering the virtual reality, we ask participants to “indicate” which of a set of ten virtual characters resembles them most. While participants will likely assume that they are represented by this avatar in the negotiation, this will not be the case – an obfuscation necessary to be able to disentangle outcomes related to “true” gender (own behavior) and “displayed” gender (others’ behavior). Participants do not see their own avatar as they operate in a “1st person mode” (looking out of the eyes of their avatar).

In the negotiation, participants will be represented by an avatar we assign them out of a set of three male and three female avatars. In particular, everyone in the role of the seller (group A) will randomly be assigned an avatar of their true gender. Participants in the role of the buyer (group B) will be randomly assigned an avatar of their true or the other gender. As a result, within group B there will be four conditions: 1) true and avatar gender male, 2) true and avatar gender female, 3) true gender male and avatar gender female, and 4) true gender female and avatar gender male.

The experimental set-up will allow us to explore gender differences in own behavior (i.e., behavior of male versus female participants) and in the other party’s behavior towards each gender (i.e., behavior towards participants represented by male versus female avatars). Additionally, the negotiation outcomes of the buyers (group B) allow investigating which share of the difference between men and women in negotiation outcomes stems from each gender’s own behavior versus the other party’s behavior towards each gender.

Finally, in a pre-screen survey and a post-experimental questionnaire we will collect data on possible underlying factors that may lead to gender differences in negotiation such as inequality aversion or participant perceptions of their negotiation counterpart in terms of gender prescriptions. We will use these for exploratory analysis.

Design details

Randomization method
Randomized using Microsoft Excel.
Randomization unit
Participants within experimental session
Clusters
-
Observations
about 400 participants (dependent on student show-up)
Arms
Participants will be evenly divided into group A (sellers) and group B (buyers). Within group A, we aim at two groups of approximately 100 observations each: A1: true gender female, avatar gender female A2: true gender male, avatar gender male Within group B, we aim at four groups of approximately 50 observations each: B1: true gender female, avatar gender female B2: true gender male, avatar gender male B3: true gender female, avatar gender male B4: true gender male, avatar gender female
Primary outcomes
Initial offer, concession speed, emotional expressions used, negotiation outcome
Secondary outcomes
Perceptions of own and other’s negotiation style and satisfaction with negotiation outcome, willingness to negotiate with counterpart again, perceptions of negotiation partner following the gender prescriptions in Amanatullah & Tinsley (2013)

Cited by

No citation data available for this registration yet.

Keywords