Coronavirus Early Reactions

How did President Trump's reactions differ from
those by health experts at the early phases of COVID-19?

By Ha Ta

The coronavirus outbreak has seemed to hit the U.S. hard. As of April 12, US death toll has overtaken Italy as world's highest, with New York state being the epicenter of the outbreak. The country is well on its way to end up with the worst COVID-19 outbreak in the industrialized world. How did a country with supposedly the best medical infrastructure in the world be so ill-prepared for the pandemic when world leaders were warned about it since early January.

This visualization takes a look at the contrast in reactions to the coronavirus between President Trump's and a range of health experts and scientists, the US Department of Health and Human Services (HHS) and the World Health Organization (WHO) in the period between Jan. 22 to March 11, 2020 through an analysis of the sentiment scores of their speeches, articles or tweets.

The project uses Syuzhet Package in R and the lexicon “Bing” to analyze the sentiments of the texts. The “Bing” lexicon assigns words into positive and negative categories with scores. The colors of the text boxes represent the sentiment of the texts. Shades of red color represent “negative” sentiment (having “Bing” score lower than 0). Shades of blue represent “positive” sentiment (having “Bing” score greater than 0). White color represent a “Bing” score of zero.

The sentiment scores are not reflective or intended to be reflective of any specific quality of the texts and the speakers, nor are they useful in representing and illustrating the complexities and nuances of these texts. They simply provide a unit of measurement to measure and illustrate a contrast in reactions from two sides. Thus, snippets of the respective texts as well as link to the full texts are provided in each textbox for your reference.

Detailed Timeline

*The sentiment scores in this visualization are not reflective of any specific quality of the texts and the speakers, nor are they meant to represent and illustrate the complexities and nuances of these texts. Thus, snippets of the respective texts as well as link to the full texts are provided in each textbox for your reference.

Timeline

Trump

Health Experts

22th

"It’s one person coming in from China. We have it under control. It’s going to be just fine,"

In an interview with CNBC’s Joe Kernen from the World Economic Forum in Davos

Sentiment score: 4

See the full text

22th

“We need to learn — and fast — about how [the coronavirus] spreads...”

Former CDC director Tom Frieden in an op-ed

Sentiment score: -9

See the full text

22th

WHO held meeting of the International Health Regulations Emergency Committee regarding the outbreak of novel coronavirus

World Health Organization

Sentiment score: -5

See the full text

24th

“China has been working very hard to contain the Coronavirus. The United States greatly appreciates their efforts and transparency. It will all work out well. In particular, on behalf of the American People, I want to thank President Xi!"

A status on Twitter

Sentiment score: 3

See the full text

28th

“Act Now to Prevent an American Epidemic.”

Luciana Borio, public health administrator. Scott Gottlieb, Former Commissioner of Food and Drugs

Sentiment score: -10

See the full text

30th

"We have it very well under control. We have very little problem in this country at this moment — five. And those people are all recuperating successfully."

In a Presidential Remark

Sentiment score: 10

See the full text

31st

Declares Public Health Emergency for United States for 2019 Novel Coronavirus

Alex Azar, U.S Secretary of Health and Human Services

Sentiment score: -2

See the full text

JAN

2nd

"We pretty much shut [the coronavirus] down coming in from China"

In an interview with Sean Hannity

Sentiment score: 3

See the full text

13rd

‘Tip of the iceberg:’ Why Covid-19 deaths will keep rising even as the coronavirus outbreak wanes

Interview with infectious disease experts

Sentiment score: -28

See the full text

10th

"A lot of people think that goes away in April"

In a Presidential Remark

Sentiment score: 2

See the full text

20th

“The infectious-disease epidemiology community and policymakers have come to the conclusion that it’s very likely that this virus is going to continue spreading throughout the world over time"

Dr. Michael Mina, Professor of Epidemiology at Harvard said in an interview with the Harvard Gazette

Sentiment score: -14

See the full text

10th

"Looks like by April, you know, in theory, when it gets a little warmer, it miraculously goes away"

In a rally in New Hampshire

Sentiment score: 7

See the full text

20th

"We need to start preparing for the next viral outbreak now"

David Bloom & Daniel Cadarette, researchers at Harvard's Global Health Population Department.

Sentiment score: -21

See the full text

10th

"We're in very good shape ... You know in April, supposedly, it dies with the hotter weather"

In an interview on Fox News

Sentiment score: 10

See the full text

25th

“As we’ve seen from recent countries with community spread ... it has moved quite rapidly.”

Nancy Messonnier, director of CDC’s National Center for Immunization and Respiratory Diseases said in a press briefing, reported by STAT News

Sentiment score: -5

See the full text

23th

"We have it very much under control in this country"

In a White House Remark

Sentiment score: 4

See the full text

26th

"U.S. isn’t ready to detect stealth coronavirus spread."

Article on Politico

Sentiment score: 0

See the full text

26th

"USA in great shape!"

Tweeted on Twitter

Sentiment score: -3

See the full text

28th

"The writing has been on the wall. The chances of stopping [coronavirus] spread have dwindled to a very, very low probability."

Dr. Michael Mina, Professor of Epidemiology at Harvard

Sentiment score: -12

See the full text

26th

"This is like a flu."

In a White House Coronavirus Briefing

Sentiment score: -3

See the full text

28th

"It’s going to disappear. One day — it’s like a miracle — it will disappear"

In a Presidential Remark

Sentiment score: 9

See the full text

FEB

2nd

"They're going to have vaccines...The U.S. is, right now, ranked by far No. 1 in the world for preparedness"

In a rally in North Carolina

Sentiment score: 9

See the full text

2nd

"40-70% of the world's adult population could be infected"

Marc Lipsitch from Harvard University, one of the country's top experts on viruses.

Sentiment score: -8

See the full text

4th

"It wasn't something that was going to affect us...It’s very mild."

In an interview with Sean Hannity

Sentiment score: 10

See the full text

3rd

This virus is not SARS, it’s not MERS, and it’s not influenza. It is a unique virus with unique characteristics.

WHO Director-General's opening remarks at the media briefing on COVID-19

Sentiment score: -11

See the full text

8th

"We’ve done a fantastic job with respect to that subject on the virus ... No I'm not concerned at all"

In a Presidential Remark

Sentiment score: 3

See the full text

6th

"Testing for the coronavirus might have stopped it. Now it’s too late."

William Hanage, a Harvard epidemiologist

Sentiment score: -13

See the full text

10th

"Just stay calm. It will go away, be calm. It’s really working out."

In a Meeting with Republican Senators

Sentiment score: 14

See the full text

10th

"We are all at risk of infection right now, some more than others, depending on where they live or the jobs they do. And it is only going to get worse."

William Hanage, a Harvard epidemiologist

Sentiment score: -18

See the full text

11th

"This is the most aggressive and comprehensive effort to confront a foreign virus in modern history."

Remarks by President Trump in Address to the Nation

Sentiment score: 17

See the full text

10th

"Some of the best epidemiologists in the world are estimating that between 40 and 70% of adults will end up getting an infections."

Dr. Ashish Jha, Harvard Global Health Expert

Sentiment score: 0

See the full text

MAR

Summary of methodology

Sentiment analysis

The project uses Syuzhet Package in R and the lexicon “Bing” to analyze the sentiments of the texts. The “Bing” lexicon assigns words into positive and negative categories with scores.

The sentiment analysis of texts from Trump with the Bing lexicon returned scores between 2 to 17. Meanwhile, sentiment analysis of texts from health experts, the HHS and the WHO returned scores between -2 and -34.

The binary comparison of this visualization hinders the ability to see the complexities in the responses, both by the President and by health experts. It also comes across as assuming that the contrast in reaction is a good versus bad comparison, when in fact, the reality is much more complicated than that.

The small snippets, together with the varying degrees of positivity or negativity are added in the hope to offset the simplicity and personal bias to an extent. When the method of analysis was (negative and positive spectrum) chosen for the project, it was intended to help highlight the contrasting reactions, but what it lost was richness of the sentiments (anger, anticipation, disgust, fear, joy, sadness, surprise…). These are the shortcomings of the visualization’s analysis method.

Origins of articles, speeches and tweets from Trump, the WHO, the HHS and health experts

Texts analyzed in this project are taken from transcripts of interviews and rallies by President Trump or published on President Trump's Twitter, the WHO's website, the HHS's website, the White House's website as well as on the following publications, which has been checked of the “media bias check” on mediabiasfactcheck.com:

  1. CBS News: rated as “LEFT-CENTER"
  2. CNBC: rated as “LEFT-CENTER”
  3. Fortune: rated as “RIGHT-CENTER BIAS”
  4. Fox News: rated as “RIGHT BIAS”
  5. Harvard Gazette: rated as “LEAST BIASED”
  6. Politico: rated as “LEAST BIASED”
  7. STAT News: rated as “LEAST BIASED”
  8. The Washington Post: rated as “LEFT-CENTER”
  9. The Conversation: rated as “LEAST BIASED”
  10. Wall Street Journal: rated as “RIGHT-CENTER BIAS”
  11. WBUR: rated as “LEFT-CENTER”

Below are descriptions of each type of categorization by mediabiasfactcheck.com:

LEFT-CENTER BIAS

These media sources have a slight to moderate liberal bias.  They often publish factual information that utilizes loaded words (wording that attempts to influence an audience by using appeal to emotion or stereotypes) to favor liberal causes.  These sources are generally trustworthy for information, but may require further investigation.

LEAST BIASED

These sources have minimal bias and use very few loaded words (wording that attempts to influence an audience by using appeal to emotion or stereotypes).  The reporting is factual and usually sourced.  These are the most credible media sources.

RIGHT-CENTER BIAS

These media sources are slightly to moderately conservative in bias. They often publish factual information that utilizes loaded words (wording that attempts to influence an audience by using appeal to emotion or stereotypes) to favor conservative causes. These sources are generally trustworthy for information, but may require further investigation.

RIGHT BIAS

These media sources are moderately to strongly biased toward conservative causes through story selection and/or political affiliation. They may utilize strong loaded words (wording that attempts to influence an audience by using appeal to emotion or stereotypes), publish misleading reports and omit reporting of information that may damage conservative causes. Some sources in this category may be untrustworthy.