The second and last day of Blog Indiana 2012 was on Friday. Blog Indiana 2012 is the fifth annual conference that brings together bloggers, writers, marketers, PR specialists, social media and SEO professionals from around the Midwest. I did an analysis of Twitter data for Day One of BlogIndiana 2012 and wanted to do the same for day two.
First, some statistics
The tweets were collected just before the first session keynote started — Friday, August 10th, 9:00 am — into the late afternoon on the same day at 6:00 pm. In total, there were 2,799 tweets — 34% more than day one — consisting of 45,497 words from 358 people. The top tweeter, again, was Jenn Lisak (Twitter: @jlisak) with 88 tweets! Here are the top 10 Twitter users and their number of tweets:
|Rank||Twitter user||Number of tweets|
Surprisingly, only four of the top ten tweeters from Day 1 were in the top10 for Day 2: Jenn Lisak (Twitter: @jlisak), Robby Slaughter (Twitter: @robbyslaughter), Erik Deckers (Twitter: @edeckers), and myself, Walter Jessen (Twitter: @wjjessen).
Similar to Day 1, most conference attendees using Twitter actually tweeted very little: 249 people tweeted just 1-4 times. Thirty-five people tweeted 5-9 times, 28 people tweeted 10-19 times, and 18 people tweeted 20-29 times. Each of the other groups with 30 or more tweets consisted of less than 12 people.
Let’s focus on retweets. Retweets are a way of saying “yeah, I like that” or “I agree!” and for the purposes of this analysis come in two forms: they can be a simple rebroadcast (e.g. RT: “original tweet”), which we’ll call a RT without conversation, or a retweet with comments back to the original tweeter (e.g. Agreed! RT: “original tweet”), which we’ll call a RT with conversation. Almost two-fifths (38%) of #BIN2012 tweets were retweets. Of those, 23% (245 of 1070) were a RT with conversation. The top retweeter was Randy Clark (Twitter: @randyclarktko) with 53 retweets.
#BIN2012 Day 2 tag cloud of tweets
As I’ve written before, I’ve always had a thing for tag clouds. I find tag clouds a useful visual representation of metadata; if they’re designed right, tag clouds can also look really good.
From 45,497 words, 4,358 were unique. I calculated the frequency of all 4,358 words and then “cleaned” the data, removing all words less than 4 characters, numbers, and words that were either common words (such as “that”, “from”, “with” or “have”) or gibberish (consisting principally of url strings from shared links). I also flattened out the top twenty terms — bin2012, @jaybaer, social, @douglaskarr, content, @muhammadinc, @kmullett, media, like, people, great, @nathan_hand, @allisonlcarter, good, @robbyslaughter, help, presentation, know, marketing and @lorraineball — since their frequency increased so much that they distorted the tag cloud.
I ran a quick analysis to see if the terms used on Day 2 were positive or negative. I again used the Subjectivity Lexicon from the Departement of Computer Science at the University of Pittsburgh. The Lexicon contained 2,304 positive terms: 7.76% of which were included in at least one tweet, and surprisingly almost half as many as Day 1. The Lexicon also contained 4,152 negative terms: 5.75% were included in at least one tweet, 18% more than Day 1. Assuming that tweets were mutually exclusive with respect to being positive or negative, #BIN2012 tweets were 1.3 times more positive than negative. Generally speaking, tweets were more negative on Day 2 of the conference than Day 1.
The top 200 terms were then imported into Wordle and a weighted tag cloud was generated. Feel free to download any of the files below and reshare.
Here are the top 10 terms (prior to flattening) from the tag cloud above:
Walter Jessen is a digital strategist, writer, web developer and data scientist. You can typically find him behind the screen something with an internet connection.