Motivated essentialism: Group ego in talent attribution

How to cite

Bonnes-Valkyser, Felix; Mechtenberg, Lydia; Zudenkova, Galina. 2025. "Motivated essentialism: Group ego in talent attribution." AEA RCT Registry. AEARCTR-0016802. https://doi.org/10.1257/rct.16802-1.0

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

Felix Bonnes-Valkyser* , Lydia Mechtenberg , Galina Zudenkova

* Primary investigator

Registry ID
AEARCTR-0016802
DOI
https://doi.org/10.1257/rct.16802-1.0
First registered
Last updated
Status
In development
Trial dates
2025-11-25 – 2026-11-25
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/16802

Abstract

We conduct an online experiment to analyze how individuals adopt worldviews on the relation between talent and salient group identities under model uncertainty. Both, talent and group identities are implemented experimentally and hence fully controlled. Our subjects observe group identities as well as individual results in a talent-related task. We investigate whether they adopt a non-essentialist worldview according to which talent is an individual characteristic rather than a group characteristic or an essentialist worldview of group talent according to which talent is a group characteristic rather than an individual one. Therefore, our experiment is designed such that the maximum-likelihood winner among worldviews is the non-essentialist belief that talent is an individual characteristic and not a group characteristic. We investigate two potential driving forces to explain why there are also individuals deviating to the essentialist worldview that considers talent as a group characteristic: group-ego utility and individual-ego utility. Both imply essentialist beliefs on talent to be more prevalent among individuals from groups in which more members solved the talent-related task successfully, compared to other groups. Furthermore, individual-ego utility implies that individuals who belong to such groups but have themselves not solved the talent-related task successfully exhibit particularly strong essentialist beliefs on talent. To separate group ego from individual ego, we reduce the salience of the group identity in a control treatment (no-color treatment). If group ego motivates essentialist beliefs about talent, then these beliefs should become less prevalent when group identities are no longer salient. This is because group-ego utility depends on group identity. In addition, we conduct spectator treatments to isolate unmotivated essentialism. Findings of our study will contribute to understanding the emergence and persistence of essentialist worldviews about group talents as well as group segregation in education, politics or the labor market.

Experimental design

In our experiment, we investigate whether group ego motivates individuals belonging to a specific group to believe that their group characteristic is linked to talent, while the group characteristic of another group is not. The group characteristics are implemented as colors yellow or blue in the color treatment, which is the treatment where the group characteristics are most salient. In the color-free treatment, these colors and hence group characteristics are removed. Being talented refers to having been assigned a 'good urn', i.e., an urn with more red balls. Vice versa, being untalented refers to having a 'bad urn', i.e., an urn with fewer red balls. There are two potential methods of urns assignment, one based on the group characteristic and the other independent of it, and only one is implemented. Which one is implemented is not observable. The two groups of subjects are shown a realization of draws from their respective individual urns. We elicit participants' beliefs on which method of urns assignment has been implemented.
In addition, we conduct two spectator treatments, one related to the color treatment and one related to the color-free treatment. Beliefs on urn assignment are incentivized. They form our main dependent variable.

Design details

Randomization method
Randomization is done by a computer, and through oTree.
Randomization unit
Individual
Clusters
6720 participants (no planned clusters)
Observations
6720 participants
Arms
ca. 2520 each for treatments 1 and 2 (players) ca. 840 each for treatments 3 and 4 (spectators)
Primary outcomes
Adopted worldview: "Group talent" (represented by 'urns assignment conditional on color' in the experiment) or "individual talent" (represented by 'urns assignment independent of color' in the experiment).
Secondary outcomes
- Belief confidence - Beliefs about participants with 'good' urns (see the experimental design section)

Cited by

No citation data available for this registration yet.

Keywords