Sharing fake news

How to cite

HENRY, EMERIC; Zhuravskaya, Ekaterina; Guriev, Sergei. 2019. "Sharing fake news." AEA RCT Registry. AEARCTR-0004060. https://doi.org/10.1257/rct.4060-1.0

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

EMERIC HENRY* , Ekaterina Zhuravskaya , Sergei Guriev

* Primary investigator

Registry ID
AEARCTR-0004060
DOI
https://doi.org/10.1257/rct.4060-1.0
First registered
Last updated
Status
In development
Trial dates
2019-05-20 – 2019-05-26
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/4060

Abstract

In this trial we plan to investigate the sharing of fake news. We address the following questions. (1) does fact-checking (i.e., providing corrective information) limit the sharing of fake news? (2) if so, what drives the decrease in sharing, image concerns or moral costs? (3) do characteristics of the sharer affect the sharing decision of the recipient? This will be done using a randomized survey experiment where participants will be randomly exposed to fake news with and without fact checking. The experiment will be conducted in the context of the European elections in May 2019.

Intervention

We detail here the intervention (see pre analysis plan for more details)

The participants will be divided into 2 groups:
Group 1 -- "the sharers"
Group 2 -- "the receivers".

The sharers will be randomly allocated to one of 3 groups:
• Alternative facts sharers (Alt Fact Share) group will be exposed to a quote from the campaign that includes some untrue information (but with full and correct attribution to the source), what we call "fake news."
• Exogenous fact-checking group (Fact Check) will be presented with the same quote (fake news) followed by fact-checking (again with full attribution to where the correct information comes from).
• Endogenous fact-checking group (Fact Check) will be presented with fake news.They will be asked whether they want to obtain the fact checking information.

Following exposure, the participants will be asked whether they want to share the news first on their facebook account, then with future participants in the experiment. Then they will be asked about their voting intentions and their knowledge of facts

The receivers will have exactly the same task as the first group. There are two differences:
1. They will be told that the content they read was shared by a previous participant (and give some characteristics of the participant, such as the reported voting intentions)
2. The proportions in each treatment will be determined by the choices made by the sharers in the first part.

We will have 3 groups:
- Control group: see nothing
- Fake from supporters: get fake information, source of sharing is a supporter
- Fake from non supporters: get fake information, source of sharing is a non supporter

Size of the groups?
The size of each group will be determined by the sharing decisions of the first group

Experimental design

see pre analysis plan attached

Design details

Randomization method
The participants to the survey will be randomly allocated to the different groups by the survey platform we use called Qualtrics
Randomization unit
Individual
Clusters
na
Observations
5000 survey respondents divided in 6 treatment groups
Arms
832 individuals per treatment
Primary outcomes
Key outcomes of interest: - proportion of sharers on Facebook across treatments - proportion of sharers within our experiment - voting intentions in the European elections

Cited by

No citation data available for this registration yet.

Keywords