The Effect of NetzDG on Social Media in Germany

How to cite

Slivko, Olga; Andres, Raphaela. 2024. "The Effect of NetzDG on Social Media in Germany." AEA RCT Registry. AEARCTR-0014385. https://doi.org/10.1257/rct.14385-1.0

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

Olga Slivko* , Raphaela Andres

* Primary investigator

Registry ID
AEARCTR-0014385
DOI
https://doi.org/10.1257/rct.14385-1.0
First registered
Last updated
Status
On going
Trial dates
2024-08-26 – 2025-03-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/14385

Abstract

This experiment is a part of a larger study which uses observational data from Twitter to measure the effect of German regulation of hate speech (the Network Enforcement Act) on the prevalence of hate in tweets. In our observational study, we measure a decrease in toxicity in tweets tackling sensitive migration and religion related topics due to the implementation of NetzDG in Germany in 2018. In the subsequent experiment, we would like to study if users of Twitter would really perceive the difference in toxicity on Twitter after the law was implemented. We randomly assign users to seeing the Twitter timelines simulated using tweets from before and after the regulation and measure the difference in toxicity as perceived by the experiment participants.

Experimental design

We create eight simulated timelines, each containing six tweets. We randomly allocate our experiment participants from Prolific platform to see 4 of these timelines and we ask to evaluate how likely is the presence of toxicity in these timelines.

Design details

Randomization method
Randomisation is performed on Qualtrics.
Randomization unit
Each participant receives four random timelines out of eight in total.
Clusters
100 participants
Observations
app. 400 obs.
Arms
100 participants are offered to evaluate fours timelines each (from before and after the implementation of NetzDG).
Primary outcomes
The key outcome is the evaluation of each participant for the presence of toxicity on each of fours simulated Twitter timelines that the participant is exposed to. It is measured on a Likert scale from 1 (not likely at all) to 7 (very likely).

Cited by

No citation data available for this registration yet.

Keywords