What happened on Twitter right after Trumps victory

For Britta Muehlbach it was traumatic. On November 8 she was watching the news on election night together with friends at their apartment in New York City. “I felt anxious, the whole time. I did not know what was going to happen.” Muehlbach sensed that it could go either way, although her friends were expecting Clinton to win over Trump.

They switched back and forth between CNN and MSNBC until about midnight. At that point Trump had not yet reached the required 270 electoral votes, yet it was already clear the big swing states were going to the Republican presidential nominee. Muehlbach left her friends place knowing that Trump had won.

“Honestly we were all in shock. I was not surprised, but I was in shock. Nobody started crying, wailing or cursing. There was like a numbness”, Muehlbach describes. It was only the next morning, when she checked her phone, that the news sunk in. The first day she could not stop crying. And the next day at work, when watching Hillary’s remarks, there were more tears.

That Muehlbach was not the only one in shock, became clearly visible days after the election when the protests started against the new president-elect. But if we take one step back, and look at the common worries and emotions of a whole country, the whole world even, right after the big news broke through. Where do you look? That is right, social media.

That is exactly what students of the University of Amsterdam did. During a data sprint, lasting from January 9 until January 15, they researched behaviour on Twitter after Trump’s victory. At 2:30 AM Trump got the 270 electoral votes he needed to win the presidential election. The students focused on behaviour and emotions of people during the fifteen minutes following that event.

“Twitter ‘claims’ to be a real time news medium, but that is just a part of it”, says Michael Stevenson, scholar in new media and digital culture at the University of Groningen. “Twitter does not ask you: how are you feeling? But rather: what is happening?” Instead of looking at the news, or what celebrity is trending, we could look at the bigger picture regarding emotions and Twitter, and find the bigger narrative, says Stevenson.

Finding the data

During their research the students of the University of Amsterdam tried to find this bigger narrative. They did that by collecting a sample of tweets by using a nifty tool called DMI Twitter Capturing and Analysis Toolset (DMI-TCAT). The selection of tweets about the U.S elections was made by using the keywords ‘Clinton’, ‘Hillary’ and ‘Trump’. Because of the rate limit on collecting data set by Twitter, 11 to 15 percent of all the tweets about the election night could be collected.

The bigger themes

The most frequent words used by Twitter-users, can be a good indication of the bigger themes during the 15 minutes after the election outcome. Looking at the word cloud above, two bigger themes can be distinguished. The cloud is pretty much self-explanatory: Big words were used more often.

Not surprisingly, those themes were sexism and racism. Words like, man, woman, sexist, sexism, stood out in the word cloud and together form the sexism theme. The words race, racism, black or white form the racism theme.

Trump didn't win tonight. Racism won. Homophobia won. Islamophobia won. White supremacy won. Sexism won. Sexual assault won. Hatred won.

— Vanessa Romo (@_uanessa) November 9, 2016

How do you explain to young girls in tears at Hillary HQ that women voted for Trump and misogyny is the way of the world? I won't. It's not.

— Dana Delany (@DanaDelany) November 9, 2016

Muehlbach acknowledges that those themes were talked about a lot by the people in her personal environment as well. “The immediate reaction for people was: we elected a racist and sexist president.” According to her, these issues are still in the background but now the focus is more on issues such as environment and health care. People have a lot of questions about the defunding of planned parenthood, the repeal and replacement of Obamacare and what is going to happen now that a denier of climate change is going to run the country, Muehlbach says.

Sentiments

Two new samples were extracted, using only tweets that were retweeted. The overall retweeted tweet sample of those 15 minutes contained 2895 tweets. After reading the tweets, and picking out the ones that didn’t match the theme, the students found a total of 54 tweets regarding sexism and 79 tweets regarding racism 79.

Donald Trump is not MY president. I will continue to sit down for the pledge. I will continue to speak up against white supremacy.

— mulher perigosa (@luminoussam) November 9, 2016

Something called a sentiment analysis tool was used to analyse those tweets. The tool is based on the NRC Word Association Lexicon (EmoLex). According to the EmoLex website: “The NRC Emotion Lexicon is a list of English words and their associations with eight basic emotions (anger, fear, anticipation, trust, surprise, sadness, joy, and disgust) and two sentiments (negative and positive).”

Overall sentiments of all the tweets sent in the 15 minutes after it became known that Trump had won the elections were divided in positive and negative. Our findings show that for all those tweets, the prevailing sentiment is positive rather than negative. However, looking at the tweets about racism and sexism, the sentiments differ slightly. The sentiments of those tweets are still more positive than negative, although there is less discrepancy between the two.

Timeline

Timeline all Tweets

Timeline racism Tweets

Timeline sexism Tweets

When comparing the timelines of emotions of all the tweets sent within the first fifteen minutes after Trumps win and the tweets about racism and sexism we saw some striking differences.

For all tweets, the first minutes show a big peak in tweets with the emotion trust. One explanation might be that most of the tweets sent during this particular time slot, use words that according to the EmoLex, are associated with trust. As seen in the EmoLex, words like ‘win’ and ‘winning’ are associated with trust. The majority of tweets during the first couple of minutes after Trumps win, might have had to do with ‘winning’.

Another explanation might be the number of factual news tweets after the election result. When you look at the graphs that show the overview of emotions during those 15 minutes, the emotion ‘trust’ spiked when the tweets were more or less factual news tweets, sent by news organizations. After two minutes, people could have switched from posting tweets that contained the news of Trumps victory, to tweets that are more emotional. This is shown in the graph as a point at which the ‘trust-tweets’ fall and the other emotions are divided proportionally amongst all tweets sent.

Whether it is really possible to read emotions, using a tool, is uncertain. Also, the validity of sentiment analysis is often quite reliable, accept when it is about political event, Stevenson adds. The best emotion-readers are still people. But if you read and try to understand tweets during the electionnight, you do get a feeling of the bigger themes and the underlying emotions people had during that period. “There is a mix of emotions, because Twitter is an emotional medium”, says Stevenson.