Information Campaigns for a Healthy Online Ad Ecosystem

How to cite

Sen, Ananya; Zhang, Xilan; DeCelles, Katherine; Eesley, Charles. 2025. "Information Campaigns for a Healthy Online Ad Ecosystem." AEA RCT Registry. AEARCTR-0015051. https://doi.org/10.1257/rct.15051-1.0

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

Ananya Sen* , Xilan Zhang , Katherine DeCelles , Charles Eesley

* Primary investigator

Registry ID
AEARCTR-0015051
DOI
https://doi.org/10.1257/rct.15051-1.0
First registered
Last updated
Status
In development
Trial dates
2025-02-12 – 2026-02-01
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/15051

Abstract

The purpose of this study is to understand which types of informational messages are most effective in encouraging website visitors to engage with a stakeholder group, which works to create a healthier digital advertising ecosystem. Specifically, the study tests how different messages highlighting factors like community size, protection of public interest, and support by institutional entities impact visitor behavior, such as signing up for newsletters or exploring the website further. By analyzing these responses, the study aims to identify the most persuasive strategies for mobilizing support and building community engagement. This research will help organizations better communicate their mission and connect with the public.

Intervention

This study involves displaying randomized pop-up messages to visitors of a stakeholder group’s website. The messages highlight factors such as community size, protection of public interests, and endorsements from well-known institutions. The aim is to explore how these elements influence visitor behavior, including signing up for newsletters or engaging further with the website.

Experimental design

Participants in this study are visitors to a stakeholder group’s website. Upon arriving at the homepage, visitors will be randomly assigned to one of four groups: three treatment groups and one control group. Each group will see a different pop-up message highlighting community size, protection of public interests, or endorsements from well-known institutions; the control group will receive a neutral message.

Design details

Randomization method
Each individual will be randomized into one of the three treatment groups or the control group using computer software.
Randomization unit
Individual
Clusters
1 website (the stakeholder group’s homepage) will serve as the cluster for the intervention.
Observations
Since such an experiment cannot have pre-testing, it is unclear how many observations we will need to detect effects. We aim for at least 1,000 website visitors per arm. Each visitor represents one observation. The number of observation may vary on the website traffic.
Arms
The sample will be divided equally across the four groups.
Primary outcomes
Newsletter signup
Secondary outcomes
Browsing behavior and return visits to the website

Cited by

No citation data available for this registration yet.

Keywords