President Trump

is changing the political landscape by using his tweets to talk directly to the American people1 in an attempt to move his policy agenda. By using Twitter, he is changing the discussion for Congress, the Judiciary, the media, and his own administration. This change has implications for policy based on the significant reactions to these tweets by both politicians and Federal employees (Graham, 2019).7 Sentiment analysis2 and natural language processing can help assess the potential impact of how these words(tweets) shape policy by his administrators, as well as the impact on his favorability ratings for re-election. Natural language processing (NLP) is a methodology that allows researchers to analyze the sentence and language-related data using computational techniques to determine specific outcomes. NLP has a wide variety of uses, for example, it is used to predict words and texts on mobile phone messaging apps. It is also used to translate web pages and help users when conducting online searches. Within NLP, we specifically chose to use a tool called sentiment analysis, or opinion mining. It is the automated process of identifying and extracting the subjective information that underlies a text. This can be either an opinion, a judgment, or a feeling about a topic or subject. The most common type of sentiment analysis is called ‘polarity detection’ and consists of classifying a statement as ‘positive’, ‘negative’ or ‘neutral’ (MonkeyLearn, 2020).9 This project will evaluate Trump’s tweets and determine trending using current data science techniques including NLP, Word Clouds and Sentiment analysis (IEE Explore, 2018).8 In addition, we will also use the Trump approval rating, as measured by various polls, as an additional variable for comparison (trump-approval-ratings, 2020).6

​

RESULTS SUMMARY

​

We analyzed 3 policies to see if Trump’s tweets changed the behavior of staff, administrators and policy managers. It is clear that just like President Kennedy changed the communications landscape in the early 60’s by using television Trump is advancing his agenda by using Twitter. But like all changes policy around presidential communications must change. Social media must now be included. Leta Lohrmeyer wrote in a paper at the University of Nebraska, “Results support the idea that the definition of public forum needs to be expanded and updated to include social media”2 It is clear from the results that this president has redefined the communications methods for the oval office and “Some would say that his use of this platform has reshaped the presidency.” (Shear et al.2019)1

 

As politics moves into the social sphere the ability to touch voters directly is producing results that have never been seen before. It is clear that the politicians will have to agree on what the rules are and create a level playing feel for all potential candidates. Tweeting at critical times does impact policy and forces administrators to respond to the “tweet crisis” as demonstrated in 3 specific cases (Border Security, Impeachment, AmericaFirst-TPP). The Cosine model was used to evaluate the consistency in the language of the tweets and was significant in setting the explicit tone for policy and resulted in direct action within President Trump’s Administration, largely based on the pressure of supporters. Consistent polling data shows after 2017 shows how either the sentiment analysis score or the language has an effect on the opinions of the public. Instead, the polls are clearly divided by party lines.