class: center, middle, inverse, title-slide # Examining the Twitter Conversation Between CNN, Fox News, Reuters, Sean Hannity, and Keith Olbermann During the Height of the 2020 Election ### Kaitlyn Fales ### 17 May 2021 --- # Introduction - The US political atmosphere is hyperpolarized; the media is no exception (Abramowitz 2011; Jurkowitz et al. 2020; Levendusky 2009; Mason 2018) - The 2020 election was unique in a variety of ways - Social media continues to rise in popularity and change the way that political media is consumed (Owen 2019) ### What does this paper hope to accomplish? --- class: inverse, middle, center # Literature Review --- # The Rise of Social Media in Politics - Political media and infotainment in the 1980s, and the Internet in the 1990s - Obama was the first to utilize social media in his 2008 campaign - All political advertisements and posts about politics are protected under the First Amendment, even if it is misinformation --- # Partisan Bias and Its Relationship with Political Media in 2020 ### What is bias? - Bias is "a tendency to believe that some people, ideas, etc. are better than others, which often results in treating some people unfairly" (Cornell University Library) ### What is confirmation bias? - Confirmation bias is "our subconscious tendency to seek and interpret information and other evidence in ways that affirms our existing beliefs" (Cornell University Library) ### News Gathering vs. News Analysis - Fact reporting versus building a larger narrative out of facts --- # Partisan Bias and Its Relationship with Political Media in 2020 - Out of 30 news sources, Democrats tend to trust 22 of those sources, while Republicans tend to trust only 7 (Jurkowitz et al. 2020) - 65 percent of Republicans trust Fox News, and 67 percent of Democrats trust CNN, but CNN is one of multiple sources that Democrats claim to trust (Jurkowitz et al. 2020) - In a Gallup poll, "69 percent of respondents said they are more concerned about how media is affecting others" ("Most People..." 2020) --- # Donald Trump: A Twitter Political Figurehead - Donald Trump's "rule by tweet" policy (Owen 2019) - Account was permanently disabled on January 8th, 2021 (Twitter Inc.) - 2019 analysis showed that Trump's most frequently used words on Twitter were "great", "president", "country", "people", "America", and "Obama" (Maksimava 2019) --- # The Contributions of this Paper - Examine the role of traditional media outlets and commentators on Twitter during the height of the 2020 election (9/1/20-1/20/21) - Analyzing the tweets of CNN, Fox News, Reuters, Keith Olbermann, and Sean Hannity during this time period - Variety of analyses, including frequently used words, correlations among frequent words, emotions analysis, and sentiment analysis --- class: inverse, middle, center # Methodology --- # Choosing the Twitter Users - Use of the [Ad Fontes Media Bias Chart](https://www.adfontesmedia.com/interactive-media-bias-chart/) to initially examine three news sources - CNN bias score of -12.15, reliability score of 42.43, and Fox News bias score of 17.19, reliability score of 32.90 - Introduction of Reuters as a neutral source, with a bias score of -7.37 and a reliability score of 51.27 (second best source when considering bias and reliability together - Pryor 2020) - Choice of political commentators (Hannity and Olbermann) was based on what two think tanks rated as some of their most hated/biased commentators --- # Data Collection - Scraped tweets in Python using the snscrape package - Every tweet from this time period for each user was used in this analysis, whether or not it actually dealt with politics and/or the election - Tweet Sample Sizes by User ``` ## CNN Fox News Reuters Sean Hannity Keith Olbermann Total ## User 14595 2765 61510 2961 7875 89706 ``` --- # Data Cleaning and Text Mining Analysis - Two copies of the data were loaded in, one for the frequently used words portion of the analysis, and the other for the emotions and sentiment analysis - For frequently used words, the data was loaded into a corpus and then cleaned - For the emotions analysis and sentiment, the tweets were not cleaned, and the emotion and sentiment scores were calculated --- class: inverse, middle, center # Results and Analysis --- # Frequently Used Words by User - Top 10 Most Frequent Words by User - Word Clouds --- # CNN <!-- --> --- # Fox News <!-- --> --- # Reuters <!-- --> --- # Sean Hannity <!-- --> --- # Keith Olbermann <!-- --> --- # CNN <!-- --> --- # Fox News <!-- --> --- # Reuters <!-- --> --- # Sean Hannity <!-- --> --- # Keith Olbermann <!-- --> --- # Frequent Words Correlation Analysis - For each user's top ten words, a correlation analysis was run, and the results show any other words, or root words, from their tweets that were correlated at or above the 0.25 level --- # CNN ``` ## $presid ## joe vice ## 0.41 0.37 ## ## $trump ## donald ## 0.36 ## ## $covid ## vaccin ## 0.27 ## ## $new ## york studi ## 0.39 0.25 ## ## $elect ## joe ## 0.33 ## ## $year ## old ## 0.41 ## ## $biden ## joe ## 0.81 ## ## $state ## unit ## 0.33 ``` --- # Fox News ``` ## $trump ## presid ## 0.25 ## ## $biden ## hunter joe ## 0.44 0.36 ## ## $elect ## result ## 0.28 ## ## $senat ## runoff georgia ## 0.37 0.30 ## ## $vote ## count ## 0.28 ## ## $state ## battleground key secretari ## 0.36 0.26 0.26 ``` --- # Reuters ``` ## $new ## york ## 0.33 ## ## $coronavirus ## case ## 0.28 ## ## $presid ## joe biden elect donald vice ## 0.45 0.34 0.34 0.28 0.26 ## ## $trump ## donald joe biden ## 0.38 0.28 0.27 ``` --- # Sean Hannity ``` ## $biden ## hunter joe ## 0.42 0.32 ## ## $covid ## relief ## 0.33 ## ## $new ## york ## 0.43 ``` --- # Keith Olbermann ``` ## $trump ## olbermann minut brief must analysi youtub covid ## 0.57 0.31 0.31 0.30 0.27 0.26 0.26 ## ## $new ## youtub show short minut covid ## 0.39 0.33 0.29 0.26 0.26 ## ## $video ## youtub show minut short analysi worst brief trump’ debat ## 0.51 0.44 0.42 0.36 0.32 0.29 0.28 0.26 0.26 ## ## $coup ## attempt conspir must plot plotter ## 0.42 0.32 0.31 0.30 0.26 ## ## $full ## youtub show minut short debat analysi vanish brief ## 0.60 0.53 0.43 0.39 0.33 0.32 0.30 0.26 ## ## $just ## ran paraphras movi bullshit hvpypqdnd knockout debat vanish ## 0.43 0.43 0.41 0.41 0.35 0.31 0.29 0.29 ## happen network hvpypp ## 0.28 0.26 0.26 ## ## $will ## coup” “hair farc dye cultist appeas ## 0.31 0.31 0.30 0.30 0.28 0.27 ## ## $biden ## counsel china show “fraud” ## 0.30 0.27 0.26 0.26 ## ## $pleas ## pledg tomjumbogrumbo via alert seen ## 0.61 0.61 0.58 0.52 0.51 ## rescu miss found nyc pound ## 0.50 0.49 0.48 0.48 0.47 ## chip cat dog avenu kill ## 0.44 0.42 0.38 0.36 0.34 ## nycacc love site thursday die ## 0.33 0.33 0.31 0.31 0.29 ## train manhattan ’ve can scare ## 0.29 0.29 0.26 0.25 0.25 ## ## $version ## short show youtub minut debat worst lose ran ## 0.56 0.49 0.49 0.45 0.38 0.28 0.27 0.26 ## china knockout ## 0.26 0.26 ``` --- # Emotions Analysis - The syuzhet package was used to classify meaningful words in a user's tweets as one of the eight main emotions: anger, anticipation, disgust, fear, joy, sadness, surprise, and trust using an NRC dictionary - For each user, a data frame was created to classify each of the words in each tweet with an emotion --- # CNN <!-- --> --- # Fox News <!-- --> --- # Reuters <!-- --> --- # Sean Hannity <!-- --> --- # Keith Olbermann <!-- --> --- # Sentiment Analysis - Shiny Application Construction - ANOVA test and Tukey Multiple Comparisons --- # Sentiment Analysis Shiny App - [Shiny Application](https://kaitlynfales421.shinyapps.io/Election2020_Twitter_Sentiment/) --- # One-Way ANOVA Test ``` ## # A tibble: 5 x 2 ## Username avg_sentiment ## <chr> <dbl> ## 1 CNN 0.0490 ## 2 FoxNews -0.152 ## 3 KeithOlbermann -0.403 ## 4 Reuters 0.0345 ## 5 seanhannity -0.231 ``` ``` ## Df Sum Sq Mean Sq F value Pr(>F) ## Username 4 1588 396.9 334.3 <2e-16 *** ## Residuals 89701 106495 1.2 ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ``` --- # Tukey Multiple Comparison <!-- --> --- class: inverse, middle, center # Conclusion --- # Conclusion and Future Research - Because this was a visualization-based analysis, it is difficult to say that these differences are occurring because of partisan polarization - However, there are granular, word-for-word differences between the users, and partisanship is involved in that simply based on the users chosen in this study - This research sheds a light on the difference between fact reporting (news gathering) and news analysis - Many future research directions, including the study of more users, and adding in a misinformation analysis --- class: inverse, middle, center # Thank You!