Can social media predict the election winner? Here’s what we learnt

The short answer: yes, active users is what matters.

Mauro Pirrone
Minely
4 min readJun 4, 2017

--

During the last few weeks, Minely has been working with ICON on a project to develop a Social Activity Scoreboard for Malta’s 2017 Election. In this project one can see a number of charts ranking the sentiment of the two parties — PN vs PL.

We analysed public data from several Facebook pages, including:

As reported in the press release by Malta Independent, the latest visualisation plots sentiment registered by unique active users each day. Social sentiment is a measurement based on aggregated social media data which gives an inkling of a person’s frame of mind about a specific topic. This visualisation plots the sentiment distribution of 137,603 Facebook users and normalises their data in such a way that multiple actions by one user are attributed singly and do not influence the general data based on volume.

Positive engagement on social media posts doesn’t really matter. What’s important is how many unique active users you have interacting in a positive manner.

By positive engagement, we refer to the positive reactions made with a post, such as the clicking the “Like” button.

Daily active users by pro-PL and pro-PN sentiment

The daily active users analysis shows how PL has been in a constant advantage over the last 100 days. The same data had been presented on a monthly basis, showing PL at 50.3%, PN at 43.06% and Undecided at 6.8%

Monthly active users by pro-PL and pro-PN sentiment, including Undecided users.

After eliminating the undecided users, data shows PL at 53.8% and PN at 46.2%

Monthly active users by pro-PL and pro-PN sentiment, excluding Undecided users.

Although data shows the trend towards a PL-win, it under estimated the huge PL victory. The final results set the PL at 55.04% and the PN at 43.68%, with the other 1.28% attributed towards the other political parties.

This brings a margin of error of around 2% when comparing social media trends against the actual results. Although the data is not representative of the entire electorate, this clearly shows that social media data is a very good indicator to analyse trends with a high level of confidence.

About Minely

Minely’s approach has been from the start not to focus too much on trying to predict the election winner. But, it is clear that social media can give insights on the trends that emerge prior to that. Minely had previously worked on analysing election data for Hillary and Trump, and more recently predicting the top 6 countries that made it to the Eurovision leader-board.

Minely is a big data analytics platform with a full range of social connectors including Facebook, Twitter, Instagram, Pinterest and YouTube. It has in-built features for natural language processing and artificial intelligence. Contact us to learn more how Minely can unlock the power of your data for a unique competitive business advantage.

About ICON

ICON is one of Malta’s most experienced software development and digital marketing companies. Having been in the business for almost two decades, the company has worked with various leading, national and international clients alike in an extensive range of sectors including financial services, telecommunications, hospitality and retail. ICON offers Big Data analytics solutions through its partnership with Minely.

--

--