Testing the effect of police forces' social media use

How to cite

Weber, Till; Zambiasi, Diego; Beckett, Melissa. 2025. "Testing the effect of police forces' social media use." AEA RCT Registry. AEARCTR-0017173. https://doi.org/10.1257/rct.17173-1.0

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

Till Weber* , Diego Zambiasi , Melissa Beckett

* Primary investigator

Registry ID
AEARCTR-0017173
DOI
https://doi.org/10.1257/rct.17173-1.0
First registered
Last updated
Status
In development
Trial dates
2025-11-05 – 2026-01-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/17173

Abstract

This study investigates the effect of social media content shared by English police forces on viewers' support for and attitudes towards the police. We include content that focuses on (1) an outcome of police work (e.g., a drug seizure or arrest), (2) engagement (e.g., community events or neighbourhood patrols), or (3) soliciting public assistance (e.g., crime prevention or reporting). We use an online experiment with English social media users to test if viewing the social media content increases support for police forces as measured by donations to a charity affiliated to active-duty police officers. Additionally, we test if the social media content affects self-reported confidence in the police, crime concern, crime reporting intentions, spending preference, and the estimated clearance rate.

Experimental design

First, participants complete a short background questionnaire.
Then, participants are randomly allocated to one of four treatments: In T0, participants are shown social media content that does not depict any police work. In T1-T3, participants are shown social media content by English police forces. In T1, the content shows an outcome of police work. In T2, the content depicts police engagement work. In T3, the content solicits public assistance.
This is followed by an incentivised donation task. Participants are asked to divide £100 between themselves and a charity with the mission to support active-duty police officers. Participants also make donation decisions to a second charity associated with harm reduction and a third charity linked to grassroot crime prevention.
Additionally, participants complete self-reported outcome measures of confidence in the police, spending preferences, estimated clearance rate, crime concern, and crime reporting intentions.

Design details

Randomization method
The survey platform Qualtrics will be used to randomly assign participants to treatments.
Randomization unit
Individuals will be randomly assigned to treatments.
Clusters
Approximately 6,284 participants.
Observations
Approximately 6,284 participants.
Arms
Approximately 1,571 participants in each of the 4 treatment arms.
Primary outcomes
Our primary outcome is the amount of money given to charity in an incentivised donation task.
Secondary outcomes
Our secondary outcomes are self-reported measures of confidence in the police, spending preference, estimated clearance rate, crime concern and crime reporting intentions. We will also elicit emotional responses (i.e., levels of happiness and excitement).

Cited by

No citation data available for this registration yet.

Keywords