 Our special guest is Ray Junko. He's an associate professor at the Lochhaven University. He is now working on experimental study, oh, he almost finished, apparently, about the effects of social media on student engagement and their grades. But I'm sure we will hear more from you. And I'm very excited and very happy to have you here. Thank you. Thank you very much, Sandra. And thank you, Urs, and the rest of the digital natives team in Berkman for inviting me. It's great to be here, and thank you all for coming. Today, I'm going to talk about a study that's currently in progress, which is a study we're doing to look at the effects of Twitter on student engagement and student success. Like I said, the study is ongoing. We're in the final weeks of the study. It's been a semester-long study. And I'm going to tell you the details of that in just a bit. I'm going to take about half an hour to talk about the study. But please feel free to jump in at any time. And so since I'm only talking for about half an hour, we certainly have a lot of time built in for discussions at the end as well, too. So feel free to ask any questions at any point. First, I'd like to introduce the research team for this study, Greg Heiberger. He's my partner in crime. We developed and conceptualized this study together. And he is the person in charge of the courses at the institution where we are doing this study, the anonymous, although I'm sure you could figure it out, site where we're doing this study. Trisha Scarcia King is our qualitative research expert on this project. I don't really do qualitative. And as a matter of fact, as we speak right now, she is coding tweets, which I don't envy her one bit. I like the numbers part. And so she's doing some semantic coding of over 1,300 tweets. And we're not done yet. And Luda Sonchak is our research assistant. She's been a great help in making sure the database is ready and doing a lot of data manipulation. I also want to say thank you to a couple of people. Eric Loken, who helped me think about our engagement measure a little bit differently. Angela Daley, who's an instructor of two of the sections of the courses that are in this study. Manny Cueva, who made me a Photoshop expert overnight so I could make the cool graphics for this presentation. And the Consortium for Service Learning in Higher Ed and the Corporation for National Community Service for their support of this study. OK, so what I'm going to do is I'm going to talk about engagement in general and the research on engagement and what I mean by engagement. And then I'm going to tell you about the study. I have some preliminary results from the study and we could look at those and then we'll have discussion at the end. So what about engagement? A few people already today have asked me, what do you mean by engagement? Well, when I talk about student engagement at the college level, I'm talking about two things. I'm talking about academic engagement in the classroom, which includes students participating in class discussions, study groups, asking questions, working with a professor, things like that. And I'm also talking about extracurricular engagement. So involvement outside of the classroom in student clubs, activities, things like that. In one slide and about two minutes, I'm going to give you a meta-analysis of 30 years of research on engagement. And these two books by Pascarella and Taranzini, I called them the higher ed bibles about how college affects students. They talk about lots of different indices of how college affects students, psychosocial, moral development, those kinds of things. I'm just going to focus on the engagement research right now. So here's what we know, 30 years of engagement research. We know that academic engagement, so that means engagement in academic work and in the academic experience, leads to increased or improved critical thinking, improved analytic competencies, and intellectual development. Social engagement, so class discussions, engagement with faculty, and engagement within the academic community, leads to psychosocial adjustment, so better psychosocial adjustment and maturity. And then lastly, extracurricular engagement leads to retention, meaning students will stay in school, they'll do better, they'll get their degrees. Educational attainment, nontraditional careers for women, I kind of bristle at that at just the statement, but it's careers in math, in science and engineering, and a more positive social self-concept. So any questions about that? That was the quick crash course on engagement. All right, so next step is social networking and engagement. So there isn't a lot of research out there about this. Some of you already know that things that happen online tend to kind of look like things that happen in the real world, they just are happening online. There have been two studies that looked at correlations between social networking website use and traditional indices of student engagement. These two studies are the Highberger study, that's Greg Highberger from this study, partner in crime, and the Higher Education Research Institute, or Harry. And Highberger looked at Facebook and student engagement, and Harry asked about all social networking websites, so MySpace, Facebook, et cetera. And basically what you can see from this slide is that the high users of social networking websites reported that they reported more engagement than the low users. So you'll see that the high users said they had a higher, very high connection to their institution, more than the low users, the high users also said they were more satisfied with their social life, they participated in greater than one student organization, they were more apt to spend greater than six hours in a student organization per week, they had more of them had a higher, very high connection with their friends, and more said that they interact daily with close friends than the low users. So again, these data are correlational. Two samples, the Harry samples, about 31,000 students at over 100 institutions, right? So you think, wow, that's a lot. The Highberger study is about 400 students at one institution, still the results seem to match up. So then what's next? What was the social networking? Okay, the social networking in the Harry was MySpace, Facebook, or anything else that people wanted to report as a social networking website. You called it a social networking website. Yeah, yep. They also had a list of the higher education research institute, I don't think they did that. There was no comparison between Facebook users versus MySpace users. They lumped them all in, okay. All right, so the next step was, so as we progressed through the research, we're thinking, okay, it looks like there might be some relationship between engagements. So why don't we test that experimentally? Because we can say there's a correlation. Can't really say there's causation. So I thought, let's do this. Let's come up with a study of testing that hypothesis. And so I thought Twitter would be a perfect technology to do this. Main reason for that, can anybody guess what my main reason for using Twitter? They used Facebook, MySpace. Just assuming. Certainly. It's a great tool for engagement. You can suck people's data without being their friends. No, there's that one too. That's another great aspect of it, which actually I realized later. But we did ask students to put their accounts on lockdown. Well, the main reason, those are all great reasons, the main reason was that there wasn't as much Twitter penetration in society as a whole and certainly on college campuses. So that was a good thing. We could actually have a control group, an experimental group, we could actually take a population of people not using on Twitter and randomly assign some to an experimental group where we will encourage them to use Twitter and then randomly have others in a control group that won't be using Twitter. The other reason was that Twitter has been very popular in education and popular in higher ed. So I thought, well, let's see if there are any effects. So we did this study in courses that are called first year seminars. They were first year seminars for the pre-health professionals at one institution. And first year seminars are courses that help students transition into college. They learn college success skills, but they also learn about their major and how to engage in that major and more about the careers that they're gonna go into. And these courses have also been shown to improve retention and to improve student engagement. So, hey, we're hoping, right? Hoping for even more engagement. So we had seven sections of these first year seminar courses. We randomly assigned four of the sections to a group that was gonna use Twitter. And three of those sections to a group that wasn't gonna use Twitter. And the reason that Ning is on this slide is that students in all sections use Ning that was just a part of the class. Ning was an already existing part of the class. We just added the Twitter component to what we call the experimental group. Here's what the experimental design looks like. So we have a control group and an experimental group. Both the control group and the experimental group during the first week of the study, they took an engagement instrument that we developed based on a, or adapted based on a larger instrument. And I'm gonna show you that engagement instrument in the next slide. It was adapted from something called the National Survey of Student Engagement, which is something that's used to compare institutions on their student engagement. We had not seen a time where the instrument or parts of the instrument looked at short-term engagement. So we certainly are keeping our fingers crossed about that. And so as time, you'll see the arrows are time. And then the experimental group, we taught them how to use Twitter. At the beginning of the semester, I went out to the research site and we had sessions for each section on how to do Twitter. And then Greg also did some follow-up of that in the following weeks and the control group did not. And then at the post test, which is actually happening this week and next week, we are sending the engagement instrument out again. And so they're all going to take it again. We are having the instructors and the TAs for each of the classes ranked students based on how engaged they think they were both inside and outside of the class. And the reason that we're doing that is that we wanna have some metric to compare this engagement instrument to, to see if we're capturing this construct that we believe is being captured, which is engagement. And they were also gonna look at grades. I thought, why not? There's been, you know, some media attention to social media and grades. And I thought, let's see if, if there's any possibility that the, that the social media use could impact grades. Any questions about that? Okay. So here's what the engagement instrument looks like. I'll leave that up for a minute so you can look at it. And basically, students were asked to rate themselves on a four-point Likert scale, how often they did each of the following things. So these things include ask questions in class or contributed to class discussions, participated in a community-based project as part of a course, which actually it was part of this course. Discuss grades or assignments with an instructor, talked with career plans with an instructor, whether they discussed their ideas from their readings with their classmates, faculty members or other people like family outside of class, whether they worked with faculty members on activities other than coursework, like their research, whether they had serious conversations with students who are very different than them in terms of religious belief, political opinion and personal values, whether they attended on campus activities, whether they exercised, which was also an on-campus activity, and whether they participated in activities to enhance their spirituality, and whether they tried to better, how often they tried to better understand someone else's view by imagining how an issue looks from his or her perspective. So the initial analyses, I ran some initial psychometric analyses on this instrument and found that the Crombox Alpha was a 0.76, okay? I may be talking like Martian right now for some folks. What that means is the Crombox Alpha is a measure of internal consistency, and it tells us how well an instrument is measuring a discrete underlying construct. So 0.76, it's pretty good. The range in social science research for good is anywhere between 0.7 and 0.8. So we're right within that sweet spot, and certainly the idea is that this underlying construct that it's measuring is this engagement. We also, with the funding, we provided incentives for students to participate in both groups. We asked, for instance, in the Twitter group, we'd post a tweet and say, if you respond to this tweet by this certain time, then you'll be entered into a drawing to win X amount of money via a gift card or an automatic deposit into their student flex account. We also, because we wanted to keep things equal between groups, we posted that same information on the Ning site. So everything we posted on Twitter was also posted on the Ning site to control for the effects of the technology and receiving information. Now we did not try to replicate the natural conversations that were going on Twitter in Ning because that seemed to happen a lot more on Twitter than it did on Ning. So we used those incentives earlier in the semester because they weren't required to use Twitter and that was a nice way to get students involved and as you might imagine, there was quite a core group that was really, really interested whether we provided incentives or not. So here's how we used Twitter in the classroom. We used it to continue class discussions. The first year seminar course was one hour a week so it doesn't leave a lot of time for continuation of discussion when you're only meeting students one hour a week. So we did that on the Twitter feed and we only had one Twitter feed for four different sections. It was a low stress way for students to ask questions about material. They discussed a common reading on the Twitter feed and the common reading is the book Mountains Beyond Mountains by Tracy Kitter. It's about Paul Farmer and his medical relief work in Haiti and he, the relief graduated either from Harvard undergrad or Harvard Med. We sent out class reminders and because these students were in the pre-health professionals, were in the pre-health professional majors, we not only sent out class reminders about the first year seminar class such as this assignments coming up but we sent them reminders about their other classes. Like, don't forget, there's a chemistry exam or if you need chemistry help, this is where you can go, things like that. We sent them campus event reminders and we used social oomph which is now called Twitter later. I don't know which one, which name I like better but you can set up tweets in advance and so we use that so we wouldn't forget that there was something coming up and we'd have to send it. It was also a way to help students connect with the instructors and with each other. We saw a lot of that happening. Use the Twitter to help them organize their service learning project because they did have to do some kind of service learning project and what I was actually hoping might happen but didn't wanna push too much was organizing study groups and so at one point a student said, hey, is anybody studying for this chemistry exam and I popped in on the Twitter feed and said, hey, why don't we get a study group going and they actually created a study group and as far as I know they've kept that study group going. So that was I think a neat consequence of doing that. This is what the assignments look like for the Twitter. We did have required assignments that happened later in the semester. So one assignment was students had to attend a sophomore or they actually got to choose one of a few things and they were to attend a sophomore or upper class student panel of students in their same major and tweet questions about that panel. They shadowed healthcare professionals and they were asked to tweet about shadowing those healthcare professionals. Then they had a couple of questions that they're actually working on right now and they were questions for their final project which include discussing the Mountains Beyond Mountains book and about their responsibility towards people who have less than they do. They had another final project where they watched a YouTube video and read an article about the Hearst family who works at the Pine Ridge Indian Reservation who also does medical relief work and the 100 people, a world portrait, have you seen this? If there were 100 people on the planet, how many would speak English? How many live without electricity? Things like that. And the third and final project question was to react to the statement that Paul Farmer's critics say that what he's doing is a band-aid for the problem. So discuss that. So here's some preliminary data. First, I'm gonna give you data that we collected right at the beginning of the semester. So went out with the engagement instrument and that was done online. And then I'm gonna give you some information about what was happening on Twitter as well. So here's what the sample looks like. 60% were female, 30% were male, 28% of our sample said that neither parent had a college degree. The participation rate in each group is as follows, 95% of students in the control group participated for a sample size of 55 and 97% of students in the experimental group participated for a sample size of 72. You'll see the race and ethnicity breakdown wasn't, I'm not thrilled about that, but then it also kind of looks like the population of this institution, which is really not like the, I guess average institution of higher ed, excuse me. So you'll see that our sample is predominantly Caucasian. So 91% of those in our sample were Caucasian. We, it looked like we did under sample Asian Americans, over sample Latinos as well. And I have yet to run some statistics on that because I just got that data. Oh, and let me back up too, I'm sorry. We are working on getting their income data. So their income data is coming either this week or next week, we're getting actual family income data from financial aid instead of self-report income data. And we're also looking, we're also getting data about their high school GPA and their ACT scores because they don't take the, they don't require the SAT for this institution. We're gonna use those variables as covariates in the analyses. So here's what they said about technology use. We asked about using their cell phones, so both texting and talking on their cell phone, Facebook, email and instant messaging. And this is what we found. These are average minutes per day. We asked them to tell us what your average hours and minutes per day were using each of these. I converted to minutes. I'll convert it back to hours as I'm talking about it. So by far texting was the most common activity. Students saying that they texted on average 190 minutes a day. So about three hours a day of texting. Yeah, that would mean that I might be texting right now. They're texting a lot. Facebook, about an hour a day. Talking on their cell phone about an hour a day. Email using email about 40 minutes a day and using instant messaging about 22 minutes a day. Which is interesting, I did a large sample survey a few years back and the IM use was much higher than this. As you might imagine, now there's text messaging and that's much more common than it was three, four years ago. So that time IM on the computer is, I think, being used texting. So there must have been some people who texted six hours a day? I'm wondering. Yeah, I'm also wondering about, yeah. And I'm also wondering about how they understood the question and we're gonna do some follow up about that. What do they mean? Because if you asked me how many minutes I text a day or how much time I text a day, I would probably aggregate the amount of time actually spent texting, right? So even if conversations went over a few hours, I would probably just count what I was doing. So that might mean like 10 minutes of texting. So we're gonna ask them about that. They might mean it's a conversation. They might mean it's a conversation, right? Or they might be doing it all the time. And as a couple of students suggested that, they pull out their phones at different times throughout the day, even in classes, we kinda do that. This is a histogram just to show the distribution of what students said the amount of time was they spent on Facebook. So the y-axis is the number of students and the x-axis is how much time they reported. So one hour a day was most popular. About 35 students said they spent an hour a day. But interestingly, 12 students said they spent no time at all on Facebook. And this is for both groups, both the control and the experimental group. One person said 15 hours. Yeah, yeah, one person said 15 hours. And again, when I see a number like that, I think... If you've had a number higher than 24, they would have picked that. They would have, yes. I had an open-ended question once before the survey instrument, not for this study, for another study, before the survey instrument could let you set parameters on it and ask students to give time, and they would always answer more hours than a day. So I'm guessing what that person thought, and again, we're gonna follow up, is they thought how long is Facebook open or something like that. Because I can't imagine, right, that they have a browser window open, Facebook is in the background, right. We have that Facebook in the back of the mind for the whole day anyway. They have their browser open, so. Sure, that's a great idea for a study, right, to try to tap how much they are just thinking about, you know, like they're processing in the background, the background process of thinking about Facebook, that it, and if it's open on a tab, you would assume that it's there, right. It's in front of you. Checking every minute. Sure, sure. So maybe, and we'll find out more about that. Yeah. Interesting. So email, where do you also have to email a client usually in the background running? Whether or not it's different with Facebook because you also expect to have IM on Facebook, which some people use, right. Mm-hmm. And I think also in the world of multi-basking, I think these kids might define what they're doing as possibly being able to do more than 24 hours in a day, right, because they're doing multiple things at once. Sure, and actually I just was telling some folks that we just submitted a manuscript about multi-tasking, and certainly if you add all the things up that they're doing, it is more because they're doing them all at the same time. Right. Yeah. Sorry, did you have any question about that software is that they integrate all the platforms? So if you tweet in that platform, the tweet appears in your Facebook, in your Twitter, and whatever. Because how? Twitter's API gives you fantastic information. So as part, what we did to analyze the tweets is downloaded the tweets from the friend timeline directly, and that was kind of fun because I had to kind of figure out how to do that. And in the XML file that you get from the Twitter API, it tells you the user client and how they're using that. So that was actually really neat. So for instance, 25% of our students sent tweets via text message, at least at last count. And whereas what you were saying is, you were talking about email, how that's different versus email. One of the things that I have found in my work is that students tend not to use email because their email account is generally the one that the institution gives them. And they don't wanna check that because it's just garbage from us, right? It's like, hey, you know, you gotta do this or you gotta do that or don't forget to register. So they prefer the Facebook is the social outlet and email is the, you know, when you get junk mail in your mail, you know, that's their email. It's hard to get them to respond to things on email. So here's another histogram of time they spent texting and talking just so you can see the distribution. The green bar is the time texting. Again, the Y-axis is the number of students who reported each time level. You'll see that the most popular amount of time spent talking was 15 minutes a day. And that was actually the way I coded that here was one through 15 minutes a day. And then texting, 30 students said that they texted about an hour a day. Just, I guess, an admission here, I truncated this at 10. There were a few outliers and we did have an outlier say texting 20 hours. And that, I think, either they're in trouble with their sleep, which could be the case or they really didn't understand that question and what we were talking about or they kind of made a mistake in answering that. But again, we'll ask them that. We had a conversation when they were texting maybe once every five minutes and it was continuous for several hours. I bet they would say it was several hours. Well, sure, but there was somebody who said 20. Which is not shown here because I kind of, you know, this chart looked a lot nicer without a 20 on it. That's why, that's why it looked too skinny without a 20 on it, with a 20 on it. But that's there and we'll investigate that. Okay, so that was information we collected at the beginning of the semester. Now here is Twitter activity up till our weeks in the study ran from Monday to Sunday. We started the study in the second week of the semester. We used semester weeks in discussing this. So you'll see that the first week I'm presenting is week two when I talk about the data. The last week we have data for is the week before last because I wasn't gonna work on this while I was driving back from Thanksgiving but also there wasn't a lot of activity during that week but we will have that data. So here's information, we didn't at the onset think that we were gonna collect data about how many people each student was following. And in about week five I looked at that and said, oh, this is kind of interesting. So, you know, we started because it really is a pain. Following data is not captured by Twitter in their API at the time that it happened. So for instance, if you download the friends timeline from Twitter and you have 50 friends, let's say I have, I'm following 50 people at the time I download that, that's what you see, right? If then I get five more friends and then download it again at some later date it'll show that 55 but we didn't start downloading the Twitter API until later. So there's really no way to go back and see how many people someone was following at say week one because Twitter will leave it as their current number in the API. So you'll see that that increased as the weeks went on. So you'll see at week 12 which is a week and a half ago that the average was 60. Kind of interesting but this is kind of most interesting to me, this is the percent of students who tweeted in the experimental group. Okay, why is it at 85 on week two? That was the week I was in their face, right? Say this is how you set up a Twitter account, everybody say hello, hello, goodbye. You lost me, hello. The neat thing though is that with doing generally nothing we had some incentives in these weeks, three through six and seven that there was participation about 30, 24 to 40% of students participated and I should say that none of these students used Twitter before the study began, which we liked, that was good. One student had an account once and didn't use it but everybody so and didn't currently have one when we started the study. So we thought that was pretty interesting participation rate. These are the average number of tweets set by week. And by the way, sorry, let me back up and say there's 85% at the end there. That is when we required all students to do an assignment. So they all had to do an assignment where they had to tweet. And that's why we have the 85% participation rate at a percent of students tweeting there. We can call participants. Week eight that you got that spike. Week eight was an assignment, but it was not something that all of them had to do. They could pick one of a few things, okay? So and some pick this. So that was the spike, not until week 12, which is when that was, we hoped that they would tweet, but we did not say you must tweet until week 12 when we asked them to do their final project. So 15% at week 12 failed to tweet. Yeah, okay, just checking. 15% at week 12. You can end the sentence that failed. Yeah, okay. Okay. So these are the average number of tweets as we go through the study. Week seven, I'll let you know about that once we go back and check that out. That was certainly an assignment week, but it was just people were really involved that week. And we're gonna analyze that a little bit more when we're looking at the categorical coding of the tweets. You see that mean number of tweets go from one to three to spiking at nine because they were to tweet two things and then respond to two student tweets was the assignment. So they were to tweet two reactions and respond to two reactions. What we found was that they didn't stop there. They almost couldn't stop because other students drew them in. They would say, but what do you mean this? And what do you mean you believe in capitalism and every person for themselves and you shouldn't help anybody, and that kind of got discussions going that way. So here is the qualitative part of this that we're doing that Tricia is actually doing some coding on right now. That's dedication. She is actually recovering from a procedure right now and she figured she would code tweets. I would watch movies. Yeah, so thanks. We came up with four categories based on tweets halfway through the study. And these four categories, the main categories of tweets were supportive, release, academic and co-curricular involvement. Supportive tweets were just that, that students were supporting each other and just providing support. Release were tweets where we felt that students were getting something off their chest. From what I could tell at the early stages we had a lot of release tweets. Then academic tweets, anything having to do with an academic issue and obviously co-curricular involvement, any activity kind of related tweet. So I'm gonna show you some examples of things we have given the primary code of one of these four things too. Tweets that we've given the primary code of one of these four things too. And the first group of four is a conversation that two students had. So one student said, I believe I've gotten to the point where no amount of coffee can pull me out of this sleep deficit. We've all felt that before. Student two said, have you tried sleeping? Student one said, do you think that will work? Good question. And then student two said, aggressive nap therapy. It has to work. I think this person quoted their own new theory, aggressive nap therapy. So that one was a conversation. I just wanted to show one example of a conversation. All the other ones were individual tweets throughout the semester. So this next one, a student noticed one other student's status. And I'm not posting that original status because I don't think it's appropriate to post in an open forum. But the person said, are you okay? Not that I know you or anything, but your status sounds not so good. At the time and even afterwards we realized how powerful this was for this student to notice that there was something going on with another student and to provide some support. And they actually engaged in a conversation that for those very reasons I'm not sharing here. But we did have to talk to that student, have to bring that student in and just check in to see how that person was doing because of what we saw on here. So release tweets. These are probably the funniest ones. One of my favorite parts of the day is sitting in Biowork hem and everyone looks at their phones because someone has posted something on this feed, the tweet feed. I need some motivation to finish my two papers due this week. Anyone have any good ideas? And of course people did pop in with ideas and so did we, we popped in with some ideas for them. I miss my puppy. If only she could come visit me. It's like LOL cats. I have such a bad migraine right now, FML. And I think that Twitter should really consider lengthing the amount of characters allowed in tweet. I mean, 140 characters are just not in. I got to say, I'm a born skeptic, right? So I think, did they really come up with that? Cause that's a really good one. What's the breakdown for residential and day students? I'd like to start your class. This isn't. Oh, this range is not there. Okay. At the location. At the location is almost all residential. Okay. I just wondered. Yep, almost all. So these are academic tweets too about the book they were reading about, you know, students reacting to farmer leaving his family and doing relief work in Haiti. Hey, Blank, what was the actual value of the unknown phosphate for chem lab? Or when is chem lab due? Or when can we do this? We had a lot of those. I got 102 on my bio test. My evil plan of not studying is working. Evil laugh. Although that would probably get to, all of these are gonna have a primary code, a secondary code and a tertiary code. This one probably would get a secondary code of release as well. Then co-curricular involvement, a student talking about one of the lectures they went to, still looking for group members interested in a volunteer project. I thought that was kind of cool getting people involved that way. Getting ready for the game tonight, later after layer of warm clothes. I like this one because I don't like cold weather. So, you know, if I'm going out to the game and I see this on the feed, oh, good, I gotta get, you know, warmer clothes on. You can never be too warm. Working at the harvest table tonight, anyone else going? That was one of their service activities, volunteering to serve food for the needy in their community. And the last thing I'm going to show you today are tweets from us. This is just a selection of things that we sent out. So very early on in the semester, we sent tweets like, 10 must do things for the first week of college. Or, this is one of my favorites, the procrastination hack. Okay, that one we sent out a few times because it is one of my favorites from 43 folders if you don't know the procrastination hack. Like the procrastination hack. Tweets about activities like the Latin American pen a series with Cuban American singer who was there. One student asked, he said, well, I think I'm leaning towards pre-law. And so we asked, what's making you lean towards pre-law and had a conversation with him about that. Information about how to sign up for tutoring and information to help them with a tech issue they were having for a video they had to watch for class. And then the final project question in the chemistry resource room. And that's it. Well, retweets, they did. We did have retweets and we stopped, we didn't do that many to begin with and we stopped doing retweets and discouraged retweeting because as part of the study, we had them follow us and follow each other. So really, it was superfluous to retweet because it was on their friend feed. Did people follow other people outside of the class? Yeah, yeah, they did. Yeah, and we had quite the range on there, right? So the average was, what did I say? It was 60, but we had quite the range. I mean, there were people following. We had one student right at the very beginning. When we did the training, I went in there and I talked to them about how to get on Twitter and this is how it works and this is why we're using it. Here's the whole concept. And had them all put their Twitter on lockdown. So that was the default. I didn't want the default to be everything was open. If they wanted to leave it open, they had the choice, but we started off that way. And we had a few students say, I'm gonna become internet famous. And I'm gonna do this. So we did have students that kind of did that. They didn't do something like, I don't know, crash a state dinner or something like that. But they were trying to get more people involved in stuff like that. We actually jokingly said to them, we'll give you all the rest of the funding that we have if you become internet famous out of this, but it didn't happen. But the people were very likable people though and they're still tweeting strong. How many outside followers did sort of take a look at those people, the ones that were trying for the internet fame? I'm curious as to how many followers did they have? Yeah, yeah. Haven't looked at that. I've looked at how many people they were following. But we certainly have that and we could sort of- Are you curious to know how close they came to their goal? Not very close. No, no, no, it wasn't. I mean, 80, maybe 100. I mean, that was the largest number I saw following back was maybe 120. What were they tweeting in class as well? Sort of questions, like relevant to the lectures? Not during our class. Not during this class, I'm saying our class, I feel so- It's a very smaller discussion class, right? You're the ones, the seminars? The seminars were- They're not lecture format. Both, they're lecture and discussion. So yeah, they weren't, we did not encourage them to tweet in class. Although they could if they wanted to, but they generally didn't. I have two questions. One, why did you, was it an IRB thing that you made them lock their profiles or their feeds? It was not an IRB thing, it was a me thing and I actually serve on my institution's IRB, but this wasn't at my institution. How's that? No, I thought it was safer in terms of confidentiality because of what might arise and wanted them to have the knowledge to do it and to start off with that as the default and then go public if they wanted to. I don't have the actual data on that, but my guesstimate is like 95% kept it as private. And it was also about things that we were asking them to post too, that I wanted to make sure that we were being sensitive to the information that they wanted to share on the feed. And then did you have any recommendations for how teachers should use Twitter in the classroom based on what you found? Some, well, I expect that I'll have more and I think probably the main idea behind how we ran Twitter in this study was that anybody could do the same, that anybody could do it the same way and that it would be very close to what a faculty member in higher ed, how they would do it, maybe with the exception of the lockdown. So, yes, and I can certainly get into more of that. How about we wait a few minutes and I'll go back? When you classify the tweets, like academic, supportive or whatever the category is, did you measure how much they were doing each of them in terms of percentage because are they really using tweets for academic purpose or no, it's just like a social tool because maybe to force a change towards academic, maybe it doesn't work because if they are 80% they use it for supportive or socially or other types of use. Yes. It's really far too much work that really is expressive. Yes, we are measuring that. We don't have all of the tweet data yet and our qualitative person just began coding them because we're not even done with the study yet. We still have two more weeks to go in this semester. And one of the things that I haven't decided that I'll just put out there, the study in terms of the dependent variables of engagement grades and the ranking ends this week and next week. But in terms of analyzing the Twitter stream, I thought, hmm, maybe it'd be interesting to take a look at that a little longer to see how much engagement there is after we just leave this all alone and see whether students are gonna continue doing it or not. I haven't decided that yet because I also have to talk with the person who's doing the coding if they want another 800 to 1,000 individual tweets to code on. The company Twitter gave you the access to the tweets. How do you know how they're doing that? I'm not sure what the question is. Because I know that there are people trying to do projects around tweets and the company didn't give them access to the database of tweets. They are actually having to, like, with some kind of software, a spider, whatever, getting the tweets like manually, kind of almost manually, so. Yes, we're downloading the tweets manually and it's not fun because the Mac is not set up well for XML support and Excel, there's no XML support, so I have to run Windows and I don't like doing that. But yes, you have to manually do it. At one point I thought, oh, maybe I could find a code or to code something automatically that calls on the API because when we first did it, Twitter will only return the largest number of tweets on any feed Twitter will return for you is 200. Beyond that, you have to do it by page and there's only 20 a page. So we realized that we could do this. I realized I could do this about halfway through the semester and had hundreds of tweets. So that was quite the process to do it manually. But I posted a blog post on my Tumblr about how to do it, how you can download the tweets. I was asking about the classroom involvement is one of the co-founders of Berkman Center, guy named Jonathan Zittrain, uses sort of a similar thing in his class where he has a back channel communication among the students and it tends to be very successful, I guess because kids or his students can engage in a conversation then he can just sort of just turn around and see what the top rated questions are and things like that. Do you have any ideas? Cause when I read about this speech, it's a Twitter in the classroom and I immediately popped into my head. Yes, that is that I'd like to do next. I'd like to look at some differences. I'm gonna be looking for funding for this. And I wanna look at the differences between back channel, so the groups would be back channel communicating, not back channel communicating, so just letting it happen naturally, required and non-required. So required back channel communication, non-required but encouraged back channel communication and then required just letting it happen, so use Twitter and then not required. Required and not required, use Twitter. That's fantastic. I wanna know the results. Thanks, well it's, knowing what I know from doing this, I don't know, does this study sound simple to you? It sounds simple to me. I think simpler is better than it first time being. Simpler is definitely better. One of the things I found is this is a lot of work. We've got reams and reams of data. You have to do something every day on it, right? I'm not good with plants and growing things. It's just not my thing, right? I'm from a city and it's like that. You kinda have to water it every day. You have to give it nutrients and sunlight and love because if not, the Twitter feed just kinda, so that's certainly gonna have to secure some funding. But if you'd like afterwards, I have in my OmniFocus, I had one of those times where I woke up early in the morning and I had this idea about the back channel versus not back channel and I wrote it all out on my OmniFocus, my task management application. So I have what came out of my head if you wanna talk about that, I'm certainly happy to do that. So I wanted to go back to your question about the recommendations. I think it is definitely a powerful tool to help students connect with each other and be engaged, however, I don't have any final data about that yet. So I'd like to have some final data so I can say, look, yes, we did have some differences and if you use it this way, then this is what you'll get and which was part of the idea to begin with. But certainly it seems that at least a group of students and I'll let you know more about, I'm guessing it's gonna be the traditional digital divide kinda issues which is gonna be socioeconomic status in differences on who used and who didn't use. But I also suspect there's some personality variables which we haven't measured and actually I know of only one study that was posted recently where they looked at extroversion versus introversion on Facebook which I'm really interested in because it does seem that extroverts are seem to be more extroverted online as well but it also does help the introverts. So I'd like to look at that. I've got all these grand plans just not enough time to do them. So do you have any preliminary comparisons between the control group and the Twitter group? The engagement instrument just went out yesterday. So, no, yeah, and I'm really, I'm itching to just download that data tonight and just run some analyses on that but I'm gonna keep myself from doing those because those that respond first, those are probably gonna be the folks who are more engaged. The following are tweets we'll find out when you do. Yeah, you follow my tweets which are not the account tweets, right, right, right. I'll be really excited. And one of the things we actually asked them, one of the, I don't know, side questions that has nothing to do with this study, we asked them about their high school GPA early in the semester to give us what their high school GPA was. I'm just gonna run a quick correlation between actual and what they reported just to see and I'm sure I'll tweet about that because that's not a paper. I'm not gonna write that up or anything like that. Yes, Joe. Do you have any idea how many professors at university use Facebook or Twitter in the classroom or some sort of other form of constant communication between professors and students? Some sort of online, or any institution? Oh, in general, I don't know. At my institution, there are a couple of professors who use it and the ones that do don't have many student friends on Facebook and as far as I know, none use Twitter. And at this institution, where we're running this study, I'm not exactly sure from what I could tell, this was the first faculty member to use it as part of a class at this institution. I remember there was something about a year ago where a student was tweeting in a class at NYU or something and the teacher got mad and asked them to stop doing it and remove the post. Was there any apprehension among the faculty about what might come of this? Whether it might turn into a Gawker type of thing or something like that? There was a bit of apprehension about whether it would be disruptive. When that student tweeted about one of my favorite parts of the day is when I'm sitting in camera bio and everyone takes out their cell phones to look because someone has posted something to the feed, I spoke to my colleague who's on site and I said, how awesome would it be just to take a video camera in there and just set it up, not let anyone know and send a tweet. We didn't have time to do that, but I mean, that was just kind of an aside that there's only tangentially related what you said. But generally, no, we didn't have any complaints. It seemed, we know they're scheduled because they're all taking generally the same classes. It didn't seem like they were tweeting during class, maybe one or two would tweet something during class, but it didn't seem like they were doing that. There definitely seemed to be times of the day that where they tweeted between classes. Lunchtime was big time. Lunchtime and around two to four was big and then after five was big and very early in the morning too. Well, not very, very early. We're on the research and this is obviously one study and they have done several studies, a little bit in context, so like two steps back and what are we studying here and in what way does it contribute to our understanding of technology and use of technology when it comes to learning and how young people think and communicate with each other, just kind of a big picture of you probably to round it up. I'm gonna go high, so if I need to go a little lower just let me know, okay. The way I conceptualize it is that maybe those that are not involved digitally as much as people in this room or our students really feel like technology is a negative factor in student psychosocial development. I don't think that's the case. The work that I have done over the years shows there are certainly positive benefits, corrective even benefits of using technology. There's certainly data on using technology for connectedness to help people with their mood. So actual more psychological things like happiness versus depression and technology helping with that. So as the years have gone on, it does seem like there are some benefits and there's not enough research on that, but there are some benefits that can be gleaned from using technologies and encouraging students to use technologies, but also they're using the technologies. They're digital natives, they're using the technology, so why not think about how to leverage those technologies in ways that make sense and have some educational and growth value for them? What made you, I mean, do you think it would be easier to get students to adopt something like this if it were like a Facebook application? I mean, they're already on Facebook 15 hours a day rather than try and implement something that they've never used before. One of the aha moments I had during the study was I didn't conceptualize this when using Twitter, but Twitter has much less information about them, right? So the buy-in for Twitter is lower than it is for Facebook. If a faculty member sits up there and says, all right, everybody, Facebook friends me, which I would never recommend, right? Students are gonna be like, what's wrong with this jerk? I'm not on Facebook right now. Because it's much more personal, it's photos, it's information about me. Twitter is all about being truncated, right? I mean, you only get 140 characters for your bio, right? I can't say a greeting on the phone in 140 characters, right? So I think it's a lot easier for students to say, oh yeah, I'll use Twitter. Plus the nature of Twitter is that it's public. Is that you're gonna, these things are gonna be exposed. There's not as much TMA at least in this group or the groups of students that I've seen as there is on Facebook, right? I mean, you look on Facebook, some of your friends feeds are like, wow, they said that, right? But it's kind of normative for Facebook versus not. So I think there's a lot less of a buy-in or a cost for students to join Twitter than there is for Facebook. I mean, I think doing Facebook would be neat too, but I think you have to be very careful about that and the way you organize it, maybe organize it as groups versus, or maybe a different account and maybe befriending students. A different account just for a class. Isn't that why teens aren't really using Twitter at young adults? Because it's not personal enough. So I can see where it'd be useful in an academic environment, but as you said, no one was using it before. Someone had set up a count and never really engaged with it. I'm glad you, and I know there's been research on the attitudes of kids towards Twitter versus Facebook, Twitter's for old people, quote unquote. That's just media. I have some slides about that. I don't believe that because the data don't seem to capture that teens don't Twitter. The data seem to show that Twitter's not very popular in the general population yet, and the age groups that are on Twitter reflect that. So I think it's very provocative for these two, Mashable and New York Times to say, teens don't tweet, but if you look at, this is actually same data from ComScore as sourced in one of those articles, in the New York Times article. The, that is unique visitors relative to their presence on the web. So 100 is average, if it's 120, it means that they're represented on that technology 20% more than they are on the web. So you'll see 12 to 24 are actually 20% more are on Twitter than relative to the web. And you'll see, and these are same data from, two to 11 is interesting, right? It must be a lot of 11 year olds. I, and you know the full disclosure here, my son's three and he has a Twitter feed, but he does not update. It's, you know, I think it's cool stuff he says. It's, you know, times between I journal about his growth and things he does and stuff that's cute, but it's hard to do all those things. So, you know, I bust out the phone and say, oh, look what he said, this is hilarious. You know, let me put that on Twitter. And of course it's on lockdown. Don't let anybody follow. It's just, you know, another way to journal, keep for me. And this is data from Pew, both in December, oh wait, you know this? Oh. Sandwich for my son. He said, daddy, you're the king burger, instead of Burger King, which is another one. Right, right. He calls it, my son calls it El Twitter de Liam, because we speak Spanish to him at home. And he wants to see, he always wants to see it because his icon is Thomas the Tank Engine. So he always wants to see El Twitter de Liam and, you know, he wants me to, you know, take videos of him doing funny stuff and then he wants to watch them, although I don't upload them online. I'm really, you know, I'm really crazy. You've seen my Facebook. His pictures of him are not on there. We're really careful about what we post on here. And here's also some recent data from Pew about people and different age groups on Twitter. Certainly you can argue about the way they ask the question about status updating, but, I mean, it does seem that there's... What are the answers? I can't read those. Oh, I'm sorry. Percent of internet users on the Y-axis and then age at the bottom, 18 through 24, is the first grouping of bars and 25 to 34 is the second grouping of bars. And what is the question they ask? The question they ask is, do you use Twitter or another service to update your status? Facebook. Their argument is that they asked about Facebook in the question before that. So my answer is we need more information than that, but there does seem to be at least some data that show that young people do use Twitter, that they're not using. On the previous graph, what is, I'm sorry, what is presence on the web mean? So, and can you say how often they're actually using a Twitter account? So if you have a Twitter account, but don't ever update it, does that count? I don't think that captures this. I don't think this captures that. An account and not use it and still be counted in the, okay. Sure, but that's what, wait actually, this is unique visitors to the main Twitter page. So when you go to login, that's what this is. The other one was self-report. What we still don't know, they're level of use. So I mean, the great thing to do would be to go out and talk to teens and do like what Esther Hargitai does and do the face-to-face survey and look at that on get a good representative sample and ask them, are you on Twitter? How often are you on Twitter? And do it that way, because I think collecting the data from Twitter would just be an enormous task that such a thing to do. I'm wondering if it's not harder to, someone else alluded to it, push Twitter out to students and expect them to use it for something that's really valuable as opposed to engage them on a platform that they're already bought into. And Facebook's a lot harder because it's a lot more closed. Right, right. Honestly, it would have been great to do this for Facebook, but think about what a group of people who aren't on Facebook would look like if there are a group of college students who aren't on Facebook. I mean, Facebook has penetration rates of anywhere. The studies show between 75 and 99% for college students. So there kind of goes the whole experimental hypothesis out the window. But yeah, that's, yes. I mean, I think the main question that comes to my mind is not so much how Twitter can affect classrooms. Why isn't there a website specifically for students to engage in this kind of like academic and extracurricular discussion in a school setting? I mean, I ask that, because I'm trying to make a website that's a very thing. But to encourage and create a social network for education, I suppose. And I guess you probably have the same sorts of question. Yeah, I think that's a fantastic idea. It makes me think of that aggregator. I'll have to look it up. I can't think of it right now. Where they, where you can set up a page for like anything, right? Like a conference. And it does its best to remove the noise from the signal when someone's using hashtag. So they look at retweeting and stuff like that. So it actually uses Twitter, but it filters it based on certain things. And if you've ever done hashtag conversations on Twitter, or do the hashtag, and then all of a sudden, hey, check out this website. You'll make a lot of money or porn. Check this out with the hashtag. This does, I think, a pretty good job of filtering that out. Ask me later, and I'll pull that up, because I can't think of it off the top of my head. Well, great. I think we have to wrap up. Thank you. That's great. Well, thank you. Thank you. For the session at 450 feet in this same room, we've shown Marciano, Andre. We will talk more broadly about the use of technology and education and curriculum building. So part of discussion will be, are these tools that you can use to get kids interested and enhance their learning experience and motivate them? But also, what is the substance that we try to teach and how does a curriculum in a digitally networked environment look like? So it will be the conversation at 4.15. Of course, you're all invited to try this. Thank you. Thanks again so much. Thank you. Thanks everyone. Thank you.