Indian Prime Minister Narendra Modi in a surprise move on 8th November 2016 declared the two largest currency bills (Rs. 1000 and Rs. 500) illegal. This action is primarily targeted towards reducing fake currency that’s in circulation in India and the massive amount of black money that corrupt people hold. It’s too early to discuss whether the move is good or bad. However, it doesn’t stop people from giving their opinions.
There are two primary hashtags on Twitter that are in use – #demonetization and #demonetisation. Note the difference in the spelling. The first is Americal English and the second is British English. I thought it might be interesting to analyze the tweets that used one over the other hashtag. It’s possible that some people used both but the chances are that they didn’t because the hashtags are long and take up too much space.
My hypothesis was that the tweets with hashtags with British spelling will be more negative because they are likely to originate from India. Like the people in most countries, Indians are also quite vocal on Twitter. Besides it is massively gamed by all the political parties. Currently all the opposition parties in India have opposed the move so I expected that they must be funding negative publicity on Twitter. They are likely to use the hashtag with British spelling.
On the other hand, many non resident Indians are likely to be unaffected by this move. The apparent benefits to the economy are large while the cost to individuals who don’t live in India is small. So the people who are using the Americam spelling are likely to be more positive about demonetization.
However, many foreign publications such as CNN, NY Times, WaPost, etc. may also tweet about this with the American hashtag. From my past experience, these publications have a severe bias against righ wing parties and Modi in general. This will likely make the comparison more difficult.
I downloaded 5,000 tweets for each hashtag. After cleaning the tweets and running a simple text analysis for identifying the sentiment as positive, negative, or neutral, this is what I found.
So we have almost twice as much tweets with positive sentiment for demonetization. Note that these tweets used the American English spelling. Now let’s take a look at the British spelling.
Well, I get the samle pattern with about double the tweets with positive sentiment than negative sentiment! So I can’t reject the null hypothesis of no difference. In other words, this will always remain a blog post 😉
Just for fun, I also plotted the wordcloud for both.