How Twitter users reacted on Trump’s impeachment: A sentiment analysis by VisionRI

The sentiment analysis shows that the process of impeachment has hardly caused any damage to the popularity of the US President Donald Trump. Interestingly, the word ‘impeach’ is the second highest among top ten words used to express negative sentiment on Twitter.

As the US House of Representatives approved the impeachment of the President (POTUS) Donald Trump, he has become the third president in the history of the US to face the Senate for impeachment. The majority of Republicans in the Senate is also likely to provide him one more similarity - his removal is a remote possibility.

However, Trump's impeachment has a unique dissimilarity in comparison to two previous impeachments conducted by the Senate - firstly, against Andrew Johnson in 1868 and; secondly, against Bill Clinton in 1998.

On social media, people are expressing their opinions round the clock. In addition to the Congressmen and Senators, Americans across classes and communities are expressing their opinions on Twitter. Though they may not affect the process of impeachment, their opinions reveal the mood of people on impeachment and also about the upcoming Presidential election.

VisionRI attempted to decipher the public mood on the issue by conducting a sentiment analysis through a scientifically proven methodology. In this exercise, 16,000 tweets posted on Twitter hashtag #TrumpImpeachment on December 19 were picked up for analysis. The sentiment analysis has provided some interesting results which could provide an insight to understand the thought process in the oldest and one of the most vibrant democracies of the world.

Divided on Congress' vote Twitter users trust Senate

In the sentiment analysis, the Twitter users are vertically divided on voting in the Congress as the word 'vote' is a topmost word used to express both the negative as well as positive sentiments. Besides, the word 'vote' is also 1st among the top ten words used on Twitter to express surprise and 2nd among top ten words used to express the sentiment of trust.

In the voting, the Congress comfortably approved charges against Trump apparently because the members voted on party lines. However, the sentiment analysis shows an equal degree of positive and negative public sentiments on Twitter on the issue of impeachment.

The Congress on Tuesday accepted the allegations of 'misuse of power' against Trump with a majority of 230 votes against 197 votes. However, the allegations of 'obstructing the Congress' were approved by 229 votes against 198 votes. But, the sentiment analysis shows a neck to neck fight. This indicates that Trump has more support on the ground in comparison to the ratio of Republicans in Congress.

We noticed the highest degree of trust on the Senate among Twitter users. The word 'Congress', however, featured at 6th rank among the top ten words used to express the sentiment of trust. Besides, the word 'Congress' featured as 2nd highest among the top ten words used to express disgust.

The other words used to express trust are 'vote', 'good', 'finally', 'happy', 'congress', 'justice', 'hope', 'majority', and 'good'. Furthermore, the word 'impeach' has received second-highest negative sentiment in our study.

It seems, besides the Republicans holding the majority in the Senate, Trump enjoys huge popularity amongst his people. This is probably the reason; he is least concerned about the process of impeachment. While the Congress was voting to impeach him, Trump was addressing a campaign at Battle Greek in Michigan and enjoying the cheers of the crowd. "While we are creating jobs and fighting for Michigan, the radical left in Congress is consumed with envy and hatred and rage, you see what's going on," said Trump amid the cheers of the crowd. The Republicans hold 53 out of 100 seats in the Senate which require a two-third majority to adopt the motion of impeachment against the President. An apparently cool Trump is busy in the campaign for the upcoming Presidential elections in the US on November 3, 2020. This day the US will also vote for new Congress and 35 new members of the Senate.

The four main words used for anticipation are – 'time', 'good', 'finally' and 'happy' – which again indicate towards a favourable public mood for Trump.

Do Americans want Trump to leave the office?

Though opposition doesn't want to see Trump anymore in the White House, it seems the people on twitter are not convinced. In the analysis, we noticed that the word 'remove' has been used to express four sentiments – sadness, anger, fear and negativity. Here the word 'remove' is the topmost word used to express sadness. Besides, it is second in expressing anger and fear but fifth in negative sentiment. We do not find any positive sentiment, joy, anticipation, trust and surprise for the word 'remove'.

This indicates that Trump is not only secure in the Senate but also secure in the public and may get the second term in the office. However, the data also show disgust, sadness and surprise for the word 'winning' but they are not much significant in sentiment analysis. Twitter users have also used the word 'democracy' to express positive sentiment and trust while some have used the word 'government' to express fear.

Abusive words used frequently

In the analysis, we also noticed the use of abusive, insulting and derogatory words against the high dignitaries in the US on Twitter. The words 'bitch', 'ass', 'shit', 'hell', 'damn', and 'hate' etc. have also been used to express various sentiments by Twitter users.

However, some Twitter users were also seen cautioning against such abusive words. The analysis shows that the sentiments of sadness, anger, fear, disgust and negativity were also expressed for the word 'abuse'.

METHODOLOGY

The sentiment analysis is based on the data extracted from Twitter hashtag #TrumpImpeachment on December 19, 2019 by using API (application programming interface). We extracted 16,000 tweets for analysis but retweets are not taken into consideration. The 16,000 were extracted tweets were subjected through three methods of natural language processing – BING lexicon, NRC lexicon and AFINN lexicon – for sentiment analysis. Tweets having followers more than 5,000 are not included in the analysis.

Research Team: J.P. Singh, Editor-in-Chief, Neeraj Singh Mehta, Technical Lead, Esham Chatterjee, Assistant Manager - Economic Studies and Planning, Siddheshwar Shukla, Associate Editor, and Senior Sub-Editors, Parag Narang and Subhro Prakash Ghosh.

Disclaimer: The conclusions drawn in the sentiment analysis are subject to technological limitations of data extraction and methodology(ies) used for data processing. This study is based on analysis of tweets only and the retweets were not taken into account. All the available precautions have been made to maintain the objectivity of the research and its presentation in a scientific fashion.