Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 44 of the 44 references it cites were found in a registry or an index.

Influence of fake news in Twitter during the 2016 US presidential election

Open access
Bovet, Alexandre, Makse, Hernán A. · 2019 · Nature Communications

Abstract

The dynamics and influence of fake news on Twitter during the 2016 US presidential election remains to be clarified. Here, we use a dataset of 171 million tweets in the five months preceding the election day to identify 30 million tweets, from 2.2 million users, which contain a link to news outlets. Based on a classification of news outlets curated by www.opensources.co , we find that 25% of these tweets spread either fake or extremely biased news. We characterize the networks of information flow to find the most influential spreaders of fake and traditional news and use causal modeling to uncover how fake news influenced the presidential election. We find that, while top influencers spreading traditional center and left leaning news largely influence the activity of Clinton supporters, this causality is reversed for the fake news: the activity of Trump supporters influences the dynamics of the top fake news spreaders.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-018-07761-2

Cites

44 of 44 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Importance of different types of news outlets in Twitter. Number of distinct tweets (a) and number of distinct users having sent tweets (b) with a URL pointing to a website belonging to one of following categories: fake or extremely biased, right, right leaning, center, left and left leaning news outlets. While the tweet volume of fake and extremely biased news is comparable to the tweet volumes of center and left volume (a), users posting fake and extremely biased news are around twice more active in average (see Table 1). Consequently, the share of users posting fake and extremely biased news (b) is smaller (12%) than the share of tweets directing toward fake and extremely biased news websites (25%)
Fig. 1 Importance of different types of news outlets in Twitter. Number of distinct tweets (a) and number of distinct users having sent tweets (b) with a URL pointing to a website belonging to one of following categories: fake or extremely biased, right, right leaning, center, left and left leaning news outlets. While the tweet volume of fake and extremely biased news is comparable to the tweet volumes of center and left volume (a), users posting fake and extremely biased news are around twice more active in average (see Table 1). Consequently, the share of users posting fake and extremely biased news (b) is smaller (12%) than the share of tweets directing toward fake and extremely biased news websites (25%)CC BY · PubMed Central
Fig. 2 Retweet networks formed by the top 100 news spreaders of different media categories. Retweet networks for fake news (a), extreme bias (right) news (b), right news (c), center news (d), left leaning news (e), and left news (f) showing only the top 100 news spreaders ranked according to their collective influence. The direction of the links represents the flow of information between users. The size of the nodes is proportional to their Collective Influence score, CIout, and the shade of the nodes’ color represents their out-degree, i.e. the number of different users that have retweeted at least one of her/his tweets with a URL directing to a news outlet, from dark (high out-degree) to light (low out-degree). The network of fake (a) and extreme bias (right) (b) are characterized by a connectivity that is larger in average and less heterogeneous than for networks of center and left leaning news (Table 2)
Fig. 2 Retweet networks formed by the top 100 news spreaders of different media categories. Retweet networks for fake news (a), extreme bias (right) news (b), right news (c), center news (d), left leaning news (e), and left news (f) showing only the top 100 news spreaders ranked according to their collective influence. The direction of the links represents the flow of information between users. The size of the nodes is proportional to their Collective Influence score, CIout, and the shade of the nodes’ color represents their out-degree, i.e. the number of different users that have retweeted at least one of her/his tweets with a URL directing to a news outlet, from dark (high out-degree) to light (low out-degree). The network of fake (a) and extreme bias (right) (b) are characterized by a connectivity that is larger in average and less heterogeneous than for networks of center and left leaning news (Table 2)CC BY · PubMed Central
Fig. 3 Types of top news spreaders accounts per media category. Proportion of verified (green), unverified (orange), and deleted (black) accounts among the top 100 news spreaders in each media category
Fig. 3 Types of top news spreaders accounts per media category. Proportion of verified (green), unverified (orange), and deleted (black) accounts among the top 100 news spreaders in each media categoryCC BY · PubMed Central
Fig. 4 Retweet network formed by the top 30 influencers of each media category. The direction of the links represents the flow of information between users. The size of the nodes is proportional to their out-degree in the complete combined network, i.e. the number of different users that have retweeted at least one of her/his tweets with a URL directing to a news outlet, and the color of the nodes indicates to which news category they belong. Nodes that belong to several news categories are represented by pie charts where the size of each slice is proportional to their CIout ranking, taking into accounts only their rank among the top 30
Fig. 4 Retweet network formed by the top 30 influencers of each media category. The direction of the links represents the flow of information between users. The size of the nodes is proportional to their out-degree in the complete combined network, i.e. the number of different users that have retweeted at least one of her/his tweets with a URL directing to a news outlet, and the color of the nodes indicates to which news category they belong. Nodes that belong to several news categories are represented by pie charts where the size of each slice is proportional to their CIout ranking, taking into accounts only their rank among the top 30CC BY · PubMed Central
Fig. 5 Activity correlation between news outlets and supporters. a Pearson cross-correlation coefficients between activity time series related to the different types of news outlets, Trump supporters and Clinton supporters. b Graph showing the correlation relations between the types of news websites and the supporters. The edges of the graph represent correlations larger than r0 = 0.49. Fake news, extreme bias (right), and right websites form a first cluster, indicated by a red square in a and shown in orange in b, while center, left leaning, and left news websites form a second cluster, indicated by a blue square in a and shown in blue in b. The activity of Trump supporters is equally correlated with all news sources and the activity of Clinton supporters, which represents the largest activity, is mainly correlated with the second media cluster and only poorly with the first one
Fig. 5 Activity correlation between news outlets and supporters. a Pearson cross-correlation coefficients between activity time series related to the different types of news outlets, Trump supporters and Clinton supporters. b Graph showing the correlation relations between the types of news websites and the supporters. The edges of the graph represent correlations larger than r0 = 0.49. Fake news, extreme bias (right), and right websites form a first cluster, indicated by a red square in a and shown in orange in b, while center, left leaning, and left news websites form a second cluster, indicated by a blue square in a and shown in blue in b. The activity of Trump supporters is equally correlated with all news sources and the activity of Clinton supporters, which represents the largest activity, is mainly correlated with the second media cluster and only poorly with the first oneCC BY · PubMed Central
Fig. 6 Granger causal network reconstruction between top news spreaders and supporters activity. a Activity time series corresponding to the top 100 left leaning news spreaders (dashed) and the Clinton supporters (continuous, right vertical axis). b Activity time series of the top 100 fake news spreaders (dashed) and the Trump supporters (continuous, right vertical axis). We show the residuals of the STL filtering after the removal of the seasonal (daily) and trend components. A causal effect seems apparent from the top 100 left leaning news spreaders to the Clinton supporters (a). Peaks in the left leaning news spreaders activity (yellow, dashed) tend to precede peaks in the activity of Clinton supporters (blue). A causal effect relation from the Trump supporters to the top 100 fake news news spreaders (b) seems also apparent. c Graph showing the maximal causal effects between the activity of the top 100 news spreaders of each media category (left) and the activity of the presidential candidate supporters (right) computed over the entire 5 months. Arrows indicate the direction of a the maximal causal effect (>0.05) between two activity time series. The width of each arrow is proportional to the strength of the causation and the size of each node is proportional to the auto-correlation of each time series. The center and left leaning top news spreaders are the news spreaders that show the strongest causal effect on the supporters activity. The values of the causal effects between each activity time series are shown in Table 4
Fig. 6 Granger causal network reconstruction between top news spreaders and supporters activity. a Activity time series corresponding to the top 100 left leaning news spreaders (dashed) and the Clinton supporters (continuous, right vertical axis). b Activity time series of the top 100 fake news spreaders (dashed) and the Trump supporters (continuous, right vertical axis). We show the residuals of the STL filtering after the removal of the seasonal (daily) and trend components. A causal effect seems apparent from the top 100 left leaning news spreaders to the Clinton supporters (a). Peaks in the left leaning news spreaders activity (yellow, dashed) tend to precede peaks in the activity of Clinton supporters (blue). A causal effect relation from the Trump supporters to the top 100 fake news news spreaders (b) seems also apparent. c Graph showing the maximal causal effects between the activity of the top 100 news spreaders of each media category (left) and the activity of the presidential candidate supporters (right) computed over the entire 5 months. Arrows indicate the direction of a the maximal causal effect (>0.05) between two activity time series. The width of each arrow is proportional to the strength of the causation and the size of each node is proportional to the auto-correlation of each time series. The center and left leaning top news spreaders are the news spreaders that show the strongest causal effect on the supporters activity. The values of the causal effects between each activity time series are shown in Table 4CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

1 retracted

Show all 44 references

Research works printed but not linked: 2

  1. Howell, L. et al. Digital wildfires in a hyperconnected world. WEF Rep. 3, 15–94 (2013).
  2. Cleveland, R. B., Cleveland, W. S., McRae, J. E. & Terpenning, I. STL: a seasonal-trend decomposition procedure based on loess. J. Off. Stat. 6, 3–73 (1990).

Cited by

See all 701

Metadata

Authors
Bovet, A., & Makse, H. A.
Affiliations
City College of New York (US)
Corresponding author
Makse, Hernán A.
Venue
Nature Communications
Published
2019-01-02
Volume
10
Issue
1
Pages
7-7
ISSN
2041-1723 (ISSN-L)
OpenAlex
W2792307011
Cited by
701
References
44
Citations by year
2026: 38 · 2025: 106 · 2024: 114 · 2023: 105 · 2022: 109 · 2021: 114 · 2020: 86 · 2019: 28
FWCI
130.25
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, other-oa)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Repository of the University of Namur (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • arXiv (Cornell University) (cc-by)
Funders
  • National Science Foundation
  • Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (165158)
Topics
  • Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
  • Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Complex Network Analysis Techniques
  • Social Sciences › Communication › Social Media and Politics
Keywords
fake news, 2016 US presidential election, Twitter, information dissemination, influential spreaders, election influence, Trump supporters, social media misinformation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Influence of fake news in Twitter during the 2016 US presidential election. https://app.cochat.ai/papers/10.1038/s41467-018-07761-2
The paper
Bovet, A., & Makse, H. A. (2019). Influence of fake news in Twitter during the 2016 US presidential election. Nature Communications. https://doi.org/10.1038/s41467-018-07761-2

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.