Real-Time Research – Analyzing Twitter Posts During Games-Learning-Society 2009

Real-Time Research – Analyzing Twitter Posts During Games-Learning-Society 2009


>>So this presentation is on Real Time Research. I’m Mark Friedman, Principal Technical Analyst at CTC and the title is Analyzing Twitter Post During Games Learning Society Conference in 2009. Why is this title really supposed to be Validating and Learning Initiatives with Real Time Collaborative Research? Because this really is a just a story of attending a conference and doing some real time research using Twitter. This happened in Madison, Wisconsin, in June of 2009, and it since been published in a book by Hulu– Lulu, which is Carnegie Mellon self-publishing press. In this session we’re going to learn how to accomplish simple learning research, produce a quick turn around presentation in 72 hours and conduct complex research on games and learning. So, we started off in a conference and I decided to attend this session at 3 great presenters, Dr. Constance Steinkuehler, Dr. Kurt Squire, and Dr. Eric Zimmerman. All some of the founders of games for serious gaming and learning. And I said, “I’m going to go attend these sessions and this ought to be great.” 3 of the best people. And unfortunately when you walk in there, they said cut off by 6 and I’m like, we’re in trouble. And there’s going to be a lot of action and lot of work, and sure enough, they told us that 36 hours from now, our 6 teams are going to be presenting our findings of a research project, and then they handed us some candy to try to sweeten the deal. But we ended up picking on the cards where behaviorism, social network and statistics. And when I saw those 3 cards, being an MBA and not a PhD with an IT degree or ISD or even a Human Factor’s PhD with cognitive analysis, I saw Excel and something I could do late at night because I didn’t want to miss the conference and I wanna do statistical analysis. And when I saw social networks, of course I thought of Twitter. And in 2009, it was a lot less popular than it is today. Behaviorism, we weren’t sure behavior we wanted to model, but we figured that would come out about 11 or 12 o’clock at night as well. So, those are what the cards look like. They give you a little theory and a topic and a method and you have to produce your entire project in 72 hours with people you don’t know. There were 5 PhDs on the team with me, so I was well surrounded, that was nice. And 2 of us actually wrote the paper together. It’s Leviton, who’s a PhD up in NY University in New York City. So, we chose to investigate the nature of Tweet content that were tagged either GLS or GLSO9 ’cause in true fashion of the web, nobody can decide what the tag should be. And I/ITSEC right now, it’s probably I/ITSEC 11, I/ITSEC 2011. There’s probably lots of different tweet tags and you’ve got to track them separately and then aggregate them yourself. So, we said– and like I said, 11 o’clock at night. We said, what will our question be after we captured all these twitters ’cause we have like 800 twitters to compare. Do Twitter post during a 24-hour period like 2 p.m. till 2 p.m., do they refer to me, you, or us? Me would be, of course, the self and the id and the you would mean they’re watching somebody else of interest and they’re interested in what they’re saying and they want to talk about it. And the us would be more collaborative. And so we’re trying to really figure out just how collaborative Twitter makes a group of people at a conference. You think everybody there was to talk about the same topic, but maybe not. What we found out is a little bit different. So, we’re trying to figure out the spirit of social network. We were going to observe the messages and we’re going to then challenge the perception that Twitter is really nothing but ego-blasting and self-display. I see chuckles in the crowd. So, a tweet really is a expression of personal action, maybe. It could be a you as if this context, the you, is something in a keynote. Going to the keynote of the opening ceremonies of I/ITSEC, a lot of people be in the audience tweeting about the you, the general, the host, the retired rear admiral, that sort of thing. Or the call to action, who would like to play leader in the arcade or yes, I’d like to do that sort of thing or who wants to go together for dinner. That’s a very common thing at conferences, how to get together for after hour social interaction. And then, if it wasn’t one of those other categories, it might have been unidentified. So the 253– 235 tweet we analyzed in that first day, 43 were the me, about the writer, 116 about the you, and 69 us, and there’s only 7 that we could not analyze. So it actually was a pretty good breakdown to get to our 100 percent, who graphically, of course, you know, Excel is very good at midnight. And this actually put us on the top performers of the teams, of the 6 teams, we’re the only people who had a graphical representation. Now, we’d go up there and make a presentation in front of the 500 people of the conference. The full statistics were with GLS over the entire period of conference was over 550 with GLS, 540 GLS09, and we talked about when the time differences were. So IRC, I captured six 631 over that time period. In the morning, we found more tweets were about themselves. And we thought what’s very interesting is about them getting ready to go to the conference, running late, missed the bus, got to find Kinko’s, trying to attend the session at 10 a.m., it’s all about themselves. And it’s very interesting. The you tweets were more prominent during the sessions because they were there to learn. It’s pretty normal and even at I/ITSEC or being at games learning society, with the word learning in there, and you’re at Madison University, University of Wisconsin, Madison, you’re there to learn. You’re there to interact with people and to learn from people, you don’t get to see your normal job. And then later in the night, it becomes much more us. The second half of the day had a larger proportion of us than in the first part. And we think that has to do with the collaborative nature at the end of the day after you’ve taken care of yourself and you’ve learned what you could and your brain is full of that and now you’re ready to share it, and you’re ready to hear what other people saw that you didn’t see. I went to booth 300, but you went to booth 2700. And the bonus trend was self-reflective tweets. We thought it was very interesting that people are apologizing in their tweets when they realized that we were analyzing them, they would apologize for their mistakes. We thought that was very funny ’cause once you know that you’re being watched, people are very respectful. Oops, sorry for spamming you. And I said, this is over 2 years ago. So, we had initial conclusions, we had reflective conclusions, and it’s ongoing because you could actually re-run this test again in another conference and maybe get the same results and maybe get different results. I got to insert this though. The purpose of this Real Time Research is to possibly give a PhD thesis a topic to work on to go to further development. We’re not supposed to set the world on fire in 36 hours with the research topic. It takes 6 months to do real methodology for a PhD thesis. We’re supposed to pick up the ones that work and don’t work. So you throw 6 ideas up in the air and maybe 1 or 2 stick, and ours was one of the 1 or 2 that stuck. And so, there are people out in Madison I think have been working on their PhD thesis on this topic since that time because plenty of PhDs in 1970s, 1980s didn’t have the Internet to test like this, social media. And then we’d go down our avenue for 6 months and fail and have to reset themselves and do another PhD topic. So this was really the first side of filter. So the initial conclusions were that the you, me, us research is nothing but a drop in the sea of possible investigations of social networks. Lots of– that’s very true after what we just did on the paper this year, Shawn. Observing people’s communication trends and analyzing comparative statistics, it’s become an integral communication channel for professionals at conferences in general. It’s also a real time communication as it was originally designed. Even text messaging wasn’t nearly as popular in 2009 as it is today. As I recall, Verizon didn’t even have an unlimited plan back then. The most they have is 250 a month. Now we have kids with 10,000 a month. But Twitter really was the way to communicate with people with short text messages back in summer of 2009. We also realized that it enhances the depth of discussion in a conference. Because instead of just listening to what the author says or presenter says, having a tweet by somebody else might spur you to think of a different angle, a different way to look at that same factoid. It’s also liberating. And it democratizes the information process. You don’t have to just believe that the person on the stage says. You don’t have to believe me exactly. You can have your own opinion. You might disagree and you might tweet about it, and I might find that later. It reshapes the entire presenter-presentee power hierarchy. Then went a bit longer as one conference go– we’re actually twitted a week later or 2 weeks later, and I was watching it. It said, we noticed that GLS09 of your presentation couldn’t produce a Twitter, one liner, it didn’t exist. That’s pretty funny, that the virtual world was more important to the people at the conference than what actually took place. So if your presentation at the conference couldn’t generate some buzz electronically, you might as well not have even presented. You worked pretty hard to present and yet electronic buzz is what people actually judge you by. I think that’s very interesting. And I gave Mr. Stubbs credit for what he wrote. He sent me a copy of it when I caught it. The ongoing is, in over the last 2 years, we’ve reflected upon this as we wrote the chapter for the book, it’s How Might the Social Network Driven Approach to event attendance affect professional conferences? Are you going to have to attend in person or are you be more online conferences? I know since that time there’s been at least three massively online courses, MOOCs. There’d only been one part of that point. There’d been– help me out on this, CCKO8. CCKO8 was the only one prior to June of ’09, when this project took place. There’s been at least 3 more since. If I’m correct, there’s been over 2 or 3,000 people participating in each of those, maybe upwards of 3 or 4,000. So, there is a trend possibly of people going to learning conferences online as opposed to going in person, if the most value you find is electronic media and electronic commentary anyway. This is a very interesting quote by Steven Johnson in a magazine that same month as this took place, is literally injecting Twitter into a conversation fundamentally changes the rules of engagement. You just don’t have to go there and listen to the stage on the stage and take what they say and write it on your trip report. You actually get some other people to give you some other exchanges and your other private changes might be more valuable than the formal ones. We won the award. They have to make up these titles though, Most Surprising Finding. People weren’t– so they don’t have like Best in Show and Gold Award but our RRTs– Real Time Research, GLS5, and our team was actually considered pure awesome. And this was presented in front of the, like I said, 500 people at University of Wisconsin, Madison because they were shocked how well the results were concrete enough to tell a PhD student they ought to do more research. So it wasn’t important that we found the answer, it was important that the answer was findable.

One thought on “Real-Time Research – Analyzing Twitter Posts During Games-Learning-Society 2009

Leave a Reply

Your email address will not be published. Required fields are marked *