This is Part IV of a four-part post. Part I talks about collecting text data from Twitter while Part II discusses analysis on text data i.e. text mining. Part III outlines the process of presenting the data using Tableau and Part IV delves into insights from the analysis.
Discussion:
- According to Wikipedia, Twitter was created in March 2006 by Jack Dorsey, Noah Glass, Biz Stone, and Evan Williams and launched in July of that year. The service rapidly gained worldwide popularity. In 2012, more than 100 million users posted 340 million tweets a day, and the service handled an average of 1.6 billion search queries per day. In 2013, it was one of the ten most-visited websites and has been described as "the SMS of the Internet". As of 2016, Twitter had more than 319 million monthly active users (Source). It is interesting to note that the former U.S. President Barack Obama is probably an early adopter and the low counts of reactions between 2007 and 2011 can be explained by the the low number of users trying out as with the early stages of a typical product life cycle.
- We can also see that Obama made use of Twitter heavily during the U.S. Presidential Elections in 2012, with his activity on Twitter much higher compared to other years. The number of reactions by other users has also increased significantly relative to the previous year, especially retweets, showing the propagation of information, thoughts or sentiments.
- The sudden huge increase in 2017 (both absolute and percentage-wise) in number of reactions, more specifically likes and retweets. For now, without additional data sources, we can only guess what could possibly lead to this phenomenon: (1) Many wanted to express their gratitude towards Obama for serving as the POTUS and to show their continuous support for him, (2) Twitter had made some changes in features in the year, (3) Users of Twitter became more expressive in 2017, (4) 2017 has been a tumultuous year that in turn propelled people to come together and stand together as one. It is unlikely that the increase in activity was due to increase in number of monthly active users as the increase has been slowing (Source). Before Nov 2015, there wasn't a "Like" option on Twitter but instead it was"Favorite" (Source), and if both are stored as the same data item, this change might explain the uptake in number of likes from end 2015 but still not the discrepancy in user behavior in 2017.
- Wanting to see if Obama is an isolated case, I did a cross-comparison with another public figure, HH Dalai Lama (below), and is surprised to see a similar spike in 2017. Similarly, the number of likes overtook the number of retweets across time. High volumes of activities, if sustained, is something Twitter can leverage on to highlight its usefulness in enhancing information propagation/ network connectedness.
- Another area of analysis that we can look at is whether there is a difference in the popular tweets by retweets versus likes.
- Looking at the frequently mentioned words in Obama's tweets across time (shown in the word cloud), majority of the tweets centered around campaigning and urging citizens to watch certain broadcasts within the day (hence high frequency of the word "today"). "OFA" also showed up a lot, showing significance of the organization during his term.
Alternatively, click here to view the interactive Tableau dashboard.