Masked Emotion: Police Officer Perception of Threatening or Nonthreatening Expressions in the era of COVID-19

How to cite

Coles, Carolyn; Owens, Emily. 2021. "Masked Emotion: Police Officer Perception of Threatening or Nonthreatening Expressions in the era of COVID-19." AEA RCT Registry. AEARCTR-0006987. https://doi.org/10.1257/rct.6987-1.1

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

Carolyn Coles* , Emily Owens

* Primary investigator

Registry ID
AEARCTR-0006987
DOI
https://doi.org/10.1257/rct.6987-1.1
First registered
Last updated
Status
Completed
Trial dates
2021-01-18 – 2021-06-30
Countries
United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/6987

Abstract

This project estimates the ability of 500 officers in the Los Angeles Police Department to assess non-verbal expressed emotion in the era of covid-19. One of the central strategies currently used to control the spread of COVID-19 is encouraging, or requiring, the public to cover their face. The impact of these coverings on policing is currently unknown. In the proposed pilot study, we will test the impact of a simple, non-medical coronavirus mask on a police officer’s ability to detect emotion, and their assessment of that emotion as threatening or nonthreatening. Eligible officers will be asked to assess expressed emotions in a series of photographs. Officers assigned to the control group will be given a set of unmasked emotive pictures to assign, and the treatment group will be given a set of otherwise identical masked emotive pictures.

This pilot will serve as a preliminary test of the non-verbal communication skills and de-escalation tactics of police officers, and how public health directives have impacted how police and the community interact. This is critical for understanding how the COVID pandemic has impacted policing, which as an institution plays a central role in the persistence of structural inequality in the US.

Experimental design

All eligible officers will be randomly assigned to a control or treatment group where they will be given a survey link to engage in the study. Each detective and patrol officer will have the opportunity to take the survey once, to prevent them from “learning” the study. The control group will view unmasked photos and the treatment group will view masked photos only. Both groups will be presented with several photos of the neutral faced pictures to assign threat or non-threat and after completion of that round, they will be viewing the sequence of photos to assess facial expression. On the back end of the survey we will keep track of the time it takes to make decisions on each of the questions and their accuracy of identifying the facial expressions.

Design details

Randomization method
randomization done by Qualtrics conditioning algorithm
Randomization unit
We will conduct a stratified random sample. The sample population will come from the Los Angeles Police Department officers. Within their ranks only patrol officers and detectives will be eligible to participate. Across the 22 districts, eligible officers will be randomly assigned to control or treatment by the online platform algorithm.
Clusters
N/A
Observations
500 police officers
Arms
21 police districts, 2 job assignments (patrol and detective), 500 officers
Primary outcomes
Speed and accuracy are the outcome measures. Speed at which an officer designates a face as threatening or nonthreatening. Accuracy of the officers assignment of emotion to the pictured expression.

Cited by

No citation data available for this registration yet.

Keywords