Celebrity Apologies

Last registered on August 18, 2022
View Trial History

Pre-Trial

Trial Information

General Information

Title
Celebrity Apologies
RCT ID
AEARCTR-0009812
Initial registration date
July 27, 2022

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
August 18, 2022, 3:24 PM EDT

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
Benjamin Ho
Affiliation
Vassar College
Contact Primary Investigator

Other Primary Investigator(s)

Additional Trial Information

Status
Completed
Start date
2022-07-28
End date
2022-08-06
Keywords
Behavior
Additional Keywords
Apologies, Trust
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
A growing literature in economics and other disciplines have studied the role of apologies in restoring the bonds of trust. Except for a few exceptions, most of these studies have been done in lab settings. In this study, we look to well publicized celebrity apologies. We collect a sample of celebrity apologies based on two online databases: a New York Times dataset of celebrities who were publicly reprimanded during the #metoo movement and a dataset compiled by the online apology blog SorryWatch who noted and analyzed a number of celebrity apologies in the recent decades. We use online raters to categorize and assess the effectiveness of these apologies. (We also will train a machine learning model based on the rating data to automate future apology classification.) We then match the categorized apology data to twitter data where we use sentiment analysis to analyze the sentiment (either positive or negative) of tweets that mention the celebrity in the time periods before the violation, after the violation but before the apology, after the apology, and in the years after the apology.
This research will examine real world impacts of apologies, and importantly tie the content of the apology to the outcomes. We classify apologies based on a game theory model from Ho (Mgmt Sci 2012). Recent related work examining real world impacts of the apology using that classification can be found in field experiments involving Uber customers (Halperin, Ho, List, Muir EJ 2022) and in stock market event studies (Fan, Ge, Ho, Mal Ssrn working paper, 2021).
Our research also explores new methods to use machine learning to classify apologies, and to assess their real world impacts.
External Link(s)

Registration Citation

Citation
Ho, Benjamin. 2022. "Celebrity Apologies." AEA RCT Registry. August 18. https://doi.org/10.1257/rct.9812-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
Using the Mturk sample we will ask workers to classify the apologies into the categories described in Ho (2012), perform an attention check, and then ask questions about the perceived warmth and competence of the apologizer at random hypothetical time intervals after the apology was tendered.
Intervention Start Date
2022-07-28
Intervention End Date
2022-08-06

Primary Outcomes

Primary Outcomes (end points)
Mturk worker assessment of apologies, and their effects.
Primary Outcomes (explanation)
The main questions we are interested in is whether an apology has a positive or negative effect on attitudes toward the apologizer, as measured by mturk responses and twitter sentiment.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Using the Mturk sample we will ask workers to classify the apologies into the categories described in Ho (2012), perform an attention check, and then ask questions about the perceived warmth and competence of the apologizer at random hypothetical time intervals after the apology was tendered.

The primary purpose of this HIT is to classify the apology.. But the secondary purpose is the experimental manipulation where we randomize the hypothetical time period that has passed since the apology was tendered and ask subjects to report how their attitudes toward the apologizer has changed as a function of the time that has passed and the type of apology.
Experimental Design Details
Randomization Method
computer. using mturk.

Randomization Unit
individual
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
220 apologies, each will be shown to 12 different workers divided among 4 treatment dates
Sample size: planned number of observations
2700 HITs
Sample size (or number of clusters) by treatment arms
660 by each date
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
using a dichotomous variable with alpha .05 and beta .2 we should be able to detect differences of 9%
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Vassar College
IRB Approval Date
2022-06-08
IRB Approval Number
n/a
Analysis Plan

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

Request Information

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