Analysing the US election using Youtube data

September 18, 2016
By

(This article was first published on Florian Teschner, and kindly contributed to R-bloggers)

Youtube is one of the channels the candidates for the US election use extensively to promote themself.
Using the public Youtube API and the R package tuber it is pretty straightforward to create a snapshop of the online discussion and sentiment.

First, I slightly exended the tuber functionality to gather the channel data for both the Hillary Clinton and Donald Trump youtube channel. (All code is posted at the end of this post.)
To do that, I first gathered all video IDs of all posted videos and then run queries to gather the video detail data.

Both candidates have been quite active in their channels, Hillary posted 180 videos, Donald 101. So let’s have a quick view on the aggregate stats per video.

plot of chunk unnamed-chunk-2

The plots show the mean (dot) for all videos and the standard error of the mean (wings) for both candidates side by side.

Here are my take-aways:
* The average number of views per video are pretty similar.
* Hillary leads in terms of likes per video.
* She also leads in the number of dislikes per video.
* Donald has more comments per video.

As both; dislikes/likes numbers are higher for Hillary, it might make sense to aggregate them to a “sentiment” measure.

plot of chunk unnamed-chunk-3

Overall, Hillary has more dislikes than likes, while Donald has a postive share of likes. This looks pretty surprising from a European perspective. (Just from my subjective perspective; even the -supposely neutral- media reports very positively on Hillary and through-out negative on Donald.)

Another potential interesting fact might be an engagement rate, measuring the number of “activities” per view. An activity can be either a like, dislike or comment.

plot of chunk unnamed-chunk-4

Here we find that Donald has roughly 30% more engagement per video view on average. Expressed differently; one out of every tenth view leads to an activity.

To sum up; it is great that youtube offers the possibility to easily gather sentiment and activity data for channels and videos. In terms of predictive analytics and election forecasting, I assume the data is pretty meaningless. (At least I hope so :).

To leave a comment for the author, please follow the link and comment on their blog: Florian Teschner.

R-bloggers.com offers daily e-mail updates about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...



If you got this far, why not subscribe for updates from the site? Choose your flavor: e-mail, twitter, RSS, or facebook...

Comments are closed.

Search R-bloggers


Sponsors

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)