Sharing and following on Twitter

Last registered on October 25, 2022
View Trial History

Pre-Trial

Trial Information

General Information

Title
Sharing and following on Twitter
RCT ID
AEARCTR-0010219
Initial registration date
October 17, 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
October 17, 2022, 5:35 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
October 25, 2022, 5:33 AM EDT

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country
United States of America
Region

Primary Investigator

Name
EMERIC HENRY
Affiliation
Sciences Po
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Ekaterina Zhuravskaya
PI Affiliation
PSE
Contact Investigator
PI Name
Sergei Guriev
PI Affiliation
Sciences Po
Contact Investigator
PI Name
Theo Marquis
PI Affiliation
Sciences Po
Contact Investigator

Additional Trial Information

Status
In development
Start date
2022-10-20
End date
2022-11-11
Keywords
Electoral
Additional Keywords
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
This study aims at understanding news consumption and news sharing on Twitter. Specifically the purpose is to:
* Estimate parameters of the utility function that can explain the decision to share news (true or false). In particular how do individuals trade off veracity of the news / ideological bias of the news and potential virality of this news?
* Compare different policies that have been proposed to fight the circulation of fake news, in particular (a) making veracity salient (b) asking for confirmation clicks (c) proposing fact checking.
* Evaluate the impact on beliefs and voting intentions of exposing individuals to balanced content.
Two experiments will be conducted during the campaign of the 2022 midterm elections.
External Link(s)

Registration Citation

Citation
Guriev, Sergei et al. 2022. "Sharing and following on Twitter." AEA RCT Registry. October 25. https://doi.org/10.1257/rct.10219-1.1
Sponsors & Partners

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

Request Information
Experimental Details

Interventions

Intervention(s)
The study will be composed of two experiments.

In the first experiment the participants will be offered 2$ to sign up to follow an account for a month. In the first treatment the account will stop being active, in the second treatment, the account will retweet the most popular news from both liberal and conservative media (balanced account) while in the third treatment the participant will be offered to follow either an account with only retweets from conservative sources or an account with only retweets from liberal sources.

In the second experiment, the participants will be offered a choice to retweet one of 4 tweets (2 true, 2 false). The treatments will reproduce different policies to reduce the circulation of fake news
Intervention Start Date
2022-10-20
Intervention End Date
2022-11-11

Primary Outcomes

Primary Outcomes (end points)
In the first part of the experiment there are two key variable of interest: (1) voting and opinions in the second survey answered 3 weeks after the first (2) behavior on Twitter (observed at the treatment and not the individual level).

In the second experiment, the key outcome is the choice of what tweet to retweet.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design

Experiment 1

The survey will ask questions on (a) Twitter usage (2) socio-demographic information (3) beliefs on key policy issues (4) voting behavior. The individuals will be contacted 3 weeks before the elections. All participants will be recontacted one week after the elections and will answer questions on (1) beliefs (2) who they voted for in the elections.

Treatments: participants will be offered 2 dollars if they choose to follow a Twitter account we created (see details below) that offers a selection of information. We have written an algorithm that selects the most popular tweet (as in most retweeted) from left leaning mainstream media and the most popular tweet from right leaning mainstream media related to politics. The list of media is taken from Alcott and Gentzkow.
Treatment 1: account stops being active
Treatment 2: account retweets balanced combination of conservative and liberal news
Treatment 3: either a liberal or a conservative account.

Experiment 2

The second experiment aims at studying the drivers of sharing behavior of news. We will conduct an online survey of roughly 3500 American voting-age individuals on the Qualtrics online platform, restricting our sample to Twitter users.

The survey will have the following phases
(a) Socio-demographic characteristics
(b) Twitter usage
(c) Political preferences

Then, at the end of the survey, the participants will be asked whether they want to share one of 4 tweets or not to share (we refer to this as the retweet screen). The question will be:
Do you want to retweet? (randomized order)
• Tweet 1
• Tweet 2
• Tweet 3
• Tweet 4
• Not retweet

Tweet 1 to 4 will be selected in the following way. There will be 2 tweets from the mainstream media (one tweet left leaning, one right leaning) described above and 2 tweets based on information that was fact checked and shown to be false, also one from the left and one from the right (see methodology below).

After the choice of whether to retweet or not has been made, we will have two additional questions

Q1: How many times do you expect this tweet to be retweeted? (asked for each tweet)

Q2: How likely do you think the statement in this tweet is true? (asked for each tweet)

Q3: Do you think the information in this tweet favors the republicans or the democrats? (asked for each tweet)



Treatments

T1 (300 participants): the participants do not have the retweet screen. The goal of this treatment is to compare the answers to Q1, Q2 and Q3 with participants who don’t share.

T2 (700 participants): standard treatment with the retweet screen

T3 (extra click) (700 participants): Before seeing the retweet screen we will inform them that they will need to confirm their intention to retweet in the next screen.

T4 (make veracity salient) (700 participants): Ask question Q2 (how likely is the treatment true) before the retweet screen

T5 (fact checking) (700 participants): On the retweet screen give them the option to see fact checks of the tweets.


Experimental 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
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
na
Sample size: planned number of observations
3000 in the first experiment (with 4 treatments) 3500 for the second experiment (with 5 treatments)
Sample size (or number of clusters) by treatment arms
approximatively 600 per treatment
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
PSE IRB
IRB Approval Date
2022-10-17
IRB Approval Number
2022-024
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