Sentiments Virus spread

InsightsSocial Media in the World of COVID-19

As America took to Twitter in an attempt to socially navigate the global pandemic, we were there too: watching COVID-19 conversations and taking notes. We combined emotion analysis (including emojilytics) with geosocial and COVID-19 case data to discover how the changing pandemic affected our reactions and influenced the cultural conversation. Scroll down to see how emojis communicate mood shifts over time from the beginning of the pandemic.


Jan 01Feb 23

Watching from afar

13% news about Wuhan
11% aid to China

From wondering whether there would be a global outbreak to conspiracy theories, to debates over protective measures, people in the U.S. were still trying to figure out how to react, what to believe, and working hard to separate fact from fiction. Most conversation centered around news updates of the coronavirus outbreak in Wuhan, as well as the Chinese government's response. Mentions also call for aid and solidarity efforts.

Extra stuff we found interesting

used 2x
more than

used 2x
more than

used 50x
more than

used 2.4x
more than

Jan 01Feb 23

Conversation spread, emotions were mixed

15 reported cases
4.8M total COVID-related tweets

As of Feb 23, there were only 15 reported cases in the US, but the conversation had spread all over. We witnessed a fluctuation in different emotions across the country. Fear dominated at 4.2x the rate of normal expression, while positive emotions like love and joy were only 20% of normal levels.

Feb 24Mar 08

From concern to anger

34% anger about gov't & politics
19% anger about changes in plans

As news of the coronavirus hitting other parts of the world spread, the majority expressed that the U.S. government wasn't doing enough, many were upset about disrupted life plans, and others were mad that everyone else was overreacting.

A global look

Non-U.S. flag emoji use was up 1000% as we talked about where the virus was spreading.

37 cases14,232 mentions
200 cases14,278 mentions
23 cases14,415 mentions
104 cases14,506 mentions
73 cases14,817 mentions
337 cases14,871 mentions
203 cases14,871 mentions
176 cases15,190 mentions
80,823 cases15,418 mentions
64 cases15,464 mentions
274 cases15,601 mentions
1,040 cases16,011 mentions
502 cases16,468 mentions
673 cases16,650 mentions
1,136 cases17,608 mentions

Feb 24Mar 08

More cases spread anger

450 reported cases
3.8k angry tweets per day

As more COVID-19 cases reached our shores the nation was filled with anger. We reacted to our government's response, the healthcare system, and the availability of PPE. We were also pissed at the virus itself.

Mar 08Mar 12

No laughing matter

80% Less emoji laughter
18k Spike to “get in on”

Though the Twittersphere never stops LOLing, the coronavirus conversation hit its least funny point, with laughter expressed 80% less than usual.

But thanks in part to the boom in popularity of TikTok, some of us still found clever ways to laugh, and some even took their last chances at a little action before the lockdown.

Shy on the sly

Wondering what those emojis even mean? We were too! Click to find out.

Mar 08Mar 12

When reality hit, fear dominated

13.9k fear tweets per day
4.5k angry tweets per day

By March 8th, with more states going into lockdown, we reacted to losing jobs, closing businesses, and life put on pause. There were 1500 reported infected, but COVID had affected us all.

Mar 12Mar 23

Promoting humanity

3.7x uptick in displays of love
3.2x uptick in displays of prayer/­concern

As the death toll rose, anger and fear turned into love and concern for others, with many social users feeling the need to share positivity with the world.

While usage was still lower than normal

Our love and concern was on the rise:

5.1K 18.7K
7.7K 24.2K

Mar 12Mar 23

Humanity gave us hope & togetherness

38.8k love tweets per day
7.8k sad tweets per day

Despite deaths jumping to over 700 across the US, we responded to COVID by sharing positivity. We wished for the safety of others, supported charitable causes, and expressed gratitude to the essential workers and healthcare providers guiding us through the pandemic.

Mar 23May 01

Adapting to the new normal

29% uptick in happy birthday tweets
16% uptick in wedding & grad tweets

In a time of social distancing, 2020 has seen us turn to social media and other technology to celebrate together.

Oh also remember this?

Trump’s comment about disinfectant garnered these increases in reactions from Americans on Twitter:







Mar 23May 01

But everyone's experience was different

23.1k love tweets per day
6.2k angry tweets per day

While expressions of positivity were becoming the dominant emotion across the country, a look at different cities tells a more nuanced story. The most impacted cities reacted 5% angrier and 6% sadder, while the least impacted expressed 8% more love and gratitude.

May 01Jun 07

Two sides of the story

15% of all heartbreak sharing personal loss
19% of all heartbreak over cancelled events

The “new normal” still isn’t quite back to normal. Through it all, those of us in the most impacted states have been a little more heartbroken, and us in the least impacted were a little angrier. And as the situation evolves, we continue to have a lot to think about.

What Twitter was angry about

The states least affected by COVID-19 were significantly angrier about economic impact and cancelled plans and less angry about testing failures. Here's the breakdown:

Testing failures
States with 3 or fewer reported cases
States with 30 or fewer reported cases

The most impacted states use 💔 41% more often than usual.

May 01Jun 07

We began adjusting to a new normal

6.4k love tweets per day
3.4k anger tweets per day

As we adapted to the new normal in the US, the online COVID conversation had already dropped to just 1/4 of it's peak levels just weeks prior, now with only 5.5M original tweets in the month of May.

Jun 07Jun 30

A SecondWave

30:1 pro- vs. anti-mask sentiment
34:1 no/slow vs. quick reopening

As cases resurged, the debates continued. While a passionate few spoke out against wearing masks and advocated for reopening, the anti-maskers lost the debate by a landslide on Twitter.

To mask or  not to mask

Do people really care about wearing masks? Just look at the COVID-related hashtags! Pro-mask hashtags were used 133k times in June, including #wearadamnmask, while #maskoff was used just 12k times.

Jun 07Jun 30

A resurgence fueled our anger

24.3k new cases per day
86% angrier than when lockdown started

Conversation increased along with new cases in new areas, with total conversations per day doubling from Jun 7 to Jun 30. This time we were less afraid; more angry and sad. Sadness hit its highest intensity so far, and anger was at its highest since before lockdown, especially in new areas now hit the hardest.


Emojis are the language of the internet. The way we use them for self-expression has never been more prevalent than during this time. As the situation unfolds, our emojions will continue to tell the story.


As the crisis continues to unfold, so will our emotions. We continue to share our feels online, now more than ever in a time of virtual human connection.

Contact us to learn how we can help you understand the attitudes and emotions of people most important to your brand.