Chapter 5 Results

5.1 Youtube

5.1.1 Initial Investigation

Here is the result for total views per candiate online:

Here is the result for views density per candidate per state:

From the series of plots using choropleth maps, we can compare: a) In each state, which candidate has more popularity online, in terms of views. b) For each candidate, in which state he/she has more popularity online, in terms of views.

5.1.2 Topic Analysis

Here are word cloud plots for top topics in 2020 presidential election on Youtube:

5.1.3 Popularity Analysis

  • Here are the core plots for our study to show which candidate are actually more popular online, based on video viewers’ likes and dislikes aggragation:

  • Here are core plots for our study to show which candidate in which state are actually more popular online, based on video viewers’ likes and dislikes aggragation:

5.1.4 Comment Analysis

Here are word cloud for comments per candidate on Youtube:

  • Here is one of the core plots for our study to show the total diverging sentiment score per candidate on Youtube:

5.1.5 Demographics

Credit:castfromclay.co.uk

Credit:castfromclay.co.uk

5.2 Twitter

5.2.1 Tweet Analysis

Here are word cloud plots for top topics in 2020 presidential election on Twitter: (Bonus viz on “twitter president”)

  • Here is one of the core plots for our study to show the total diverging sentiment score per candidate on Twitter:

5.2.2 Demographics

Credit:castfromclay.co.uk

Credit:castfromclay.co.uk

5.3 Facebook

5.3.1 Post Analysis

Here are word cloud for posts per candidate on Facebook:

  • Here is one of the core plots for our study to show the total sentiment score per candidate on Facebook:

5.3.2 Demographics

Credit:castfromclay.co.uk

Credit:castfromclay.co.uk

(Please note: We have more demographics analysis in Chapter-6, as interactive components.)

Github