A country’s response to a monumental crisis is reliant upon political leadership that presents a united front. This was not the case when COVID-19 first struck the United States. Democrats and Republicans failed to see eye to eye on appropriate mitigation strategies at that point in time.
The public takes cues from elected Congressmen about behavioral changes they need to adopt. Contradicting facts might mislead citizens into making wrong decisions, endangering themselves and others around them.
A recent study conducted by Ohio State University and University of South Carolina recorded a huge number of polarizing tweets by members of Congress during the onset of the pandemic. Researchers investigated tweets between January 17 to March 31, 2020 to prove the bias perpetuated by politicians on Twitter.
A difference in the number of tweets put out by both sides was first observed. Democrats posted much more than Republicans in the beginning with 19,803 tweets. On the other hand, the latter posted 11,084 tweets during the time period. According to their calculation, this is equivalent to 71 tweets for every Democrat and 45 tweets for every Republican member of Congress.
The lack of political consensus became more evident when community spread was first detected on February 26 in California. The political divide deepened further from March 13 onwards when a national emergency was declared.
“The differential emphasis on the issue itself, independent of differences in word usage, suggests that Democratic members were sending earlier and stronger signals to their constituents that they should be concerned about the crisis,” the paper published in the journal of Science Advances on June 24 said.
Excluding the volume and tone of tweets, mixed messages were sent to the respective constituents. Different word usages were noted in both parties’ messaging. For example, the word “health” appeared in 26 percent of the tweets put out by Democrats, while the word “health” showed up in 15 percent of Republican tweets.
“The words most frequently used by Democrats concern public health and direct aid to workers (e.g., health, leave, testing) while the words most frequently used by Republicans concern national unity, China, and business (e.g., together, United States, China, businesses),” the researchers explained.
On the pattern of polarizing tweets, the researchers could not find consistency until there was no denying the seriousness of the pandemic.“However, polarization quickly rises, peaking during the week beginning February 9 – roughly two weeks after the first reported case in the United States and well after the virus had begun to have devastating effects in multiple peer democracies,” the researchers noted.
“From there, polarization declines slightly in early-to-mid March before rising again later in the month as the parties debated the various relief packages designed to mitigate the economic damage caused by the pandemic,” the researchers added further.
The researchers employed natural language processing and machine learning techniques to classify the tweets based on the texts and date. They then proceeded to identify partisanship of 76 percent of tweets. A matrix of texts attached to dates was created to devise a random forest training algorithm with the help of 70 percent of the data. This was applied to identify the remaining 30 percent’s party affiliation.