The 2016 presidential primary campaign has evolved from a horserace to a skirmish. Super Tuesday represents a critical junction for each remaining campaign, and is sure to winnow the field even further. The candidates continue to tussle on stage at debates, on the ground in battleground states, and on social media.

To better understand the relationship between big data, mobile, and social media on the 2016 campaign, since mid-January TechRepublic has logged publicly-available Twitter data related to candidate accounts. We’ve discovered that winning and high-performing campaigns know how to translate big data and social media activity into actionable tactics that covert supporters to voters.

READ: Big Data Primer for IT Pros (Tech Pro Research eBook)

Here are three trends that the Twitter data is telling us heading into Super Tuesday.

Front-runners have emerged.

As early primary contests played out over the past month, clear front-runners emerged on Twitter. The following is a list of account average follower growth since January 26, 2016.

  1. @realDonaldTrump 20,900
  2. @BernieSanders 10,400
  3. @HillaryClinton 10,300
  4. @MarcoRubio 5,320
  5. @TedCruz 3,950
  6. @RealBenCarson 1,870
  7. @JohnKasich 1,440

Election results in early primaries bear out the conclusion that as the field of candidates shrinks Twitter followers flow towards the frontrunners.

@MarcoRubio’s momentum is back.

The Republican field on Twitter is more volatile than the Democratic field, but so-called #marcomentum has returned. Rubio’s followers spiked near the Iowa caucus, then his momentum slowed in the days following the New Hampshire primary. Since the South Carolina primary, however, Rubio’s account activity started to trend upwards again, culminating with a spike following the GOP debate on February 25. While his account still has nowhere near the clout of Trump’s, and Cruz is nipping at his heels, Rubio is experiencing a surge of fresh followers at a rate far faster than his rivals.

@realDonaldTrump is on a roll.

Before the primary season kicked off, a common media narrative speculated that Donald Trump’s support had a ceiling, and that most of his support was bolstered by his media appearances and social media activity. Trump’s strength Twitter tells a different story. His account growth, relative to his competition, is almost literally off the charts. What accounts for this growth is unclear.

Bonus question: How “real” are campaign followers?

Demonstrated by election results leading up to Super Tuesday-not to mention his controversial rallies-Trump’s audience is lively. Does Trump’s online activity translate to real-life action? Leading up to Super Tuesday, it appears so.

A scan of third-party site indicates that in addition to @realDonaldTrump, Democratic frontrunner @HillaryClinton also has a large percentage of junk followers. Others accounts like @BernieSanders, @TedCruz, and @MarcoRubio appear to have higher quality followers. Though we can’t vouch for veracity of, the question is a good one: Where do zombie followers come from, and are the campaigns aware of the dead accounts?

Over the coming weeks we’ll investigate the zombie follower phenomenon, and we’d would love your theories in the comments below.

TechRepublic’s Election Tech 2016 coverage

The winning and losing campaigns can be seen as analogous to emerging companies fighting for market share in a competitive industry. By building a historic dataset of Twitter information using low-cost tools we hope to make simple data analysis accessible, understandable, and useful for small companies trying to understand the advantages of leveraging of big data and social media.

We sliced some data in-house, and also used to aggregate and interpret engagments such as Likes, Tweets and Retweets, and new Followers on Twitter and Instagram.

TechRepublic data and charts were produced with assistance from political scientist William P. Stodden, using Microsoft Excel.

Over the course of the campaign we will continue to perform simple data analysis. In the future we hope to correlate sentiment with follower actions like retweets and likes. We hope to uncover additional and unique insights.

If you’re a data scientist, social media professional, or inquisitive TechRepublic reader we’d love your ideas on how to inspect campaign social media data. Please leave a comment below, or ping us on Twitter @ TechRepublic.

Read more