Josh Minot

Logo

PhD student, University of Vermont

Follow me on Twitter

View My GitHub Profile

Home | Contact

POTUS anchored sentiment

As the 2020 US Presidential election came and went, I was curious how the conversation around Trump and Biden responded to current events. Using the LabMT[1] dataset (which underpins the Hedonometer) I scored daily word frequency-distributions for Tweets containing the 1-gram biden and trump. From this, we can see how both sentiment time series responded to COVID19, the George Floyd’s death, and the January 6th insurrection at the US Capitol.

POTUS anchored sentiment Fig. 1: Anchored sentiment for tweets mentioning Biden and Trump.

The selection criteria introduces some considerations. For instance it will include tweets that are retweets from either the Biden or Trump accounts. This is notable especially in Trump’s considering his tweets’ tendency to attract significant attention in the form of response activities—in effect leading RTs of his content to have a disproportionate effect on the ambient sentiment measurement.

References

[1] Dodds, Peter Sheridan, et al. “Temporal patterns of happiness and information in a global social network: Hedonometrics and Twitter.” PloS one 6.12 (2011): e26752.