Self-Assessments in Subjective Performance Evaluation

Last registered on January 27, 2025
View Trial History

Pre-Trial

Trial Information

General Information

Title
Self-Assessments in Subjective Performance Evaluation
RCT ID
AEARCTR-0015223
Initial registration date
January 27, 2025

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
January 27, 2025, 10:38 AM EST

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

Locations

Country
United States of America
Region

Primary Investigator

Name
Marina Schröder
Affiliation
Leibniz University Hannover
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
David Kusterer
PI Affiliation
Erasmus University Rotterdam
Contact Investigator
PI Name
Dirk Sliwka
PI Affiliation
University of Cologne
Contact Investigator

Additional Trial Information

Status
In development
Start date
2025-01-29
End date
2025-06-29
Keywords
Firms & Productivity, Labor
Additional Keywords
Self-assessment, Ratings, Performance Evaluation, Honesty, Trust
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Many organizations ask their employees to provide a self-assessment of their own performance to supervisors before these supervisors evaluate the employee's performance. We study the effect of using self-assessments for the accuracy of overall performance evaluations in an online experiment where subjects work on a real effort task and other subjects have to rate their performance. In a 2x2 experimental design we vary (i) whether agents provide a self-assessment to supervisors and (ii) whether agents receive a bonus based on the rating provided by the supervisor.
External Link(s)

Registration Citation

Citation
Kusterer, David, Marina Schröder and Dirk Sliwka. 2025. "Self-Assessments in Subjective Performance Evaluation." AEA RCT Registry. January 27. https://doi.org/10.1257/rct.15223-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
In a 2x2 experimental design we vary (i) whether agents provide a self-assessment to supervisors and (ii) whether agents receive a bonus based on the rating provided by the supervisor.
Intervention Start Date
2025-01-29
Intervention End Date
2025-06-29

Primary Outcomes

Primary Outcomes (end points)
- Informativeness of ratings
- Rating
- Self-Assessment
Primary Outcomes (explanation)
We measure the informativeness of ratings by (i) the squared deviation between rating and actual performance and (ii) the profit achieved from a fictitious job assignment decision.

Secondary Outcomes

Secondary Outcomes (end points)
Prediction quality
Secondary Outcomes (explanation)
We regress performance on the rating and compare the respective coefficients of determination (R²) between the treatments.

Experimental Design

Experimental Design
We study the effect of using self-assessments for the accuracy of overall performance evaluations in an online experiment where subjects work on a real effort task and other subjects have to rate their performance. In a 2x2 experimental design we vary (i) whether agents provide a self-assessment to supervisors and (ii) whether agents receive a bonus based on the rating provided by the supervisor.
Experimental Design Details
Not available
Randomization Method
Randomization by computer
Randomization Unit
Individual
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
4000 agents
4000 supervisors
Sample size: planned number of observations
4000 agents 4000 supervisors
Sample size (or number of clusters) by treatment arms
1000 agents per treatment
1000 supervisors per treatment
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
We conducted a power analysis based on data from Kusterer and Sliwka (forthcoming) which studied subjective performance evaluations in a related setting without self-assessments. To achieve 80% power for a sensible effect size of an increase in the accuracy of ratings (i.e. squared rating error), we will have about 1.000 supervisors per treatment. Or, respectively the MDE for the squared rating error is estimated at about 108. As, in contrast to Kusterer and Sliwka, now each supervisor evaluates 5 agents, we expect to have larger power/a lower MDE
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Ethical Review Board of the Faculty of Management, Economics and Social Sciences of the University of Cologne
IRB Approval Date
2024-07-15
IRB Approval Number
240036DS
Analysis Plan

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

Request Information