 Thank you so much for that very kind introduction, and thank you so much for showing up. I have a jacket that I'm not going to put on because it's too hot, so I apologize. OK, so well, we weren't sure about the noise, so we weren't sure. I've read a lot of press questions about it. I was going to tell you, open it. See how it goes. So as John said, I have a lot of material, and I didn't want to. I still wanted this talk to be comprehensible, I took out a lot of slides that I would ordinarily include, but I want to say that you should definitely feel free to ask me for more detail about pretty much anything and everything that I'm going to say, because there's probably more there. So hopefully it'll still be comprehensible this way, but there's a lot that I'm not showing you. Hopefully it's still a message that will make sense. And towards the end, I have possible skips, depending on how I'm doing with the time, because there are a few things I really wanted to tell you about that I think it's for interesting discussion, but we'll see. OK, so I'm going to start with how I approach this topic. I'm sure most of you, if not all of you in this room, have heard of the term digital divide, has usually referred to access, no access, to digital technologies, and very much a binary approach of great or, oh, well, if you're not online, that's a problem. And then once you're online, we've solved all the problems. And I take a critical approach to this and have done so for quite a few years, arguing that just because people are online doesn't mean that suddenly we've solved all problems of inequality, because there will be differences in how people use the medium. I apologize to those who are behind me. I'm a little hard to speak to all of you. So the term that I use is second level digital divide, or another possible term is digital inequality. And the idea, again, here is that there's a spectrum of differences among users. And so just to give you a little bit about the overall framework with which I approach this topic, so while non-users are still an issue, especially for looking at the question globally, that's not really my focus. So that's still an important issue, but that's not really what I deal with. I deal with the user. So I start out with the user and recognize the fact that users occupy a certain position in society, and they have certain background characteristics. Most of the literature that works on this takes that into account. What I also emphasize is that people's uses happen in a certain context. And this context has both a technical side and a social component. And the technical side, again, what types of technical context opportunities you have when you're using digital media, but then there's also the social side, which has different components that are going to be relevant anywhere from fairly directed issues. Say if you run into a problem or there are people around you, you can address to help you. But also informally, if you're in a network of people who are also very knowledgeable, then they can pass things on to you completely informally when you're not even really talking about digital media uses that will enhance your abilities. And so then what I argue is that all of these factors together contribute to people's level of user skill, so online abilities, and that there will be a difference in this among people. And then ultimately, skill and all those other factors influence what people actually do online and what they do with digital media. Now as a sociologist and someone interested in social inequality, what then really matters to me, and this is the $10 million question to which we're just starting to get an answer to, what requires longitudinal data, which is nearly non-existent right now, is does all this then feed back into people's life chances? So as a sociologist interested in social inequality, this question is really only relevant if we do see this feedback mechanism. Otherwise, questions could be interesting to other folks, but not so much of concerns to the students of stratification. OK, so this is where I'm not going to spend time talking about what are the processes by which this question might go, but if you'd like to hear more, I'd be happy to address it. So just to recap the questions that I'm interested in, first, are there skill differences among users? And if yes, what predicts those differences? Are there systematic differences? Second, how do people's IT uses differ? Third, what is the relationship between these two factors? And then finally, as I mentioned, does this all feed back into people's life chances? So these are the main questions that guide most of the work I'm doing currently. So again, this is where I would ordinarily talk a little bit about some previous work I had done that helped methodologically figure out ways to study skill because it's a very complicated issue. And people have studied it in various ways that weren't necessarily optimal. So I've developed some new methods. There's always room for more work. But one of the things that I discovered after initial work in this area, which was done on adults aged 18 to 81, was not surprisingly that younger people tend to be better skilled. So age was a very robust predictor of skill. Nonetheless, as someone who is exposed to young people all the time, I was quite convinced that this doesn't mean that young people are all uniformly skilled. So I was very interested in studying young people in particular. And another reason that young people are good to study, especially in the college environment, is that you can control for basic access. So you can take that issue out of the equation. So what I've been doing is collecting data at the University of Illinois, Chicago. And you might be thinking, why UIC? Well, I won't get into too much detail. I'm at Northwestern. If you know Chicago, this is actually not really convenient to study UIC. So this is not a question of convenience. The reason that the study is being done at UIC is because since I'm interested in social inequality, I need a diverse sample. And UIC is ranked top 10 among US research universities in ethnic diversity. And it's not the flagship campus of the UI system. So it really has quite a diverse student population that really helps with the questions of interest to me. So what we did as one part of this MacArthur funded project, and I'm happy to talk about the rest in Q&A, is we did a paper pencil survey. We administered a paper pencil survey on campus last winter. And we collected detailed data about people's internet uses, digital media uses, and very detailed data about their demographic background. The nice thing is that there's one course at UIC that's required of all students. So there's no selection bias as to who took the survey, because we administered the survey in this course. Now you're thinking, oh, well, that's easy. You went to one course, and everyone took the survey. Well, not quiet, because there are actually 87 different sections of this course. So we went to 85 different sections where people took the survey. We ended up with a pretty high response rate, which is helpful if you want to generalize at least to this campus population. And as I said, very diverse. Just a few words about the samples. You have some idea of who I'm talking about in the next few slides with the results. So mostly young. We're looking at first years only. Pretty good diversity in terms of parental education. I'm using that variable as a proxy for socioeconomic status. And then also on race and ethnicity, we have considerable variance, much more than lots of other schools. So this is definitely the wired generation. Almost everybody has a cell phone. They've been online on average for six years. They spend several times. They're practically constantly online, or pretty much everybody goes online every day. So we really are controlling for that in this population. And almost everybody has access at home, actually. Now, obviously, that access might not be the same, but they do have access at home. Just don't be in the dorm in this country. No. That's a very good question. Actually, 53% of the population of this sample lives with their parents. And it turns out that that's a relevant variable in all sorts of analyses, one of which I might or might not get to today. But that is actually a very interesting variable. And in that sense, this is not representative of all sorts of other schools. But it's interesting to be able to explore the context of a living situation on some of the variables of interest to me. So home just means where you live do you have access to the internet. OK, so I will have just this one slide where I give you a little bit more detail because so much of what else I'll talk about depends on this. And so this is how I measure skill. And you're welcome to ask me more questions about it. I've worked on this for years and I'm ready to defend it. But I'm also happy to get feedback about it. Basically, a lot of traditional literature just asks people how they rate their own skill. And instead here, I use a list of it's actually 27 items where I ask this question. And then it's asked in two different questions. And I ask people to rate their level of understanding of different internet related items. And I'll show you those in a second. And then I summarize all those variables and that's an index of skill. And there are a couple of articles that talk about this in more detail and compare it to other measures. OK, so on these skill terms that are these internet related terms that I use can be broken into what I would consider more basic internet related terms and then more advanced web 2.0 types of terms. Now we could certainly classify some of these differently. It doesn't really make much of a difference in the end. For the analyses, I just thought it might be helpful to look at it this way. So OK, so people know some terms better than others, but this graph actually doesn't really tell us anything about whether there's any variation in the sample. But there is variation in the sample on this measure, which is helpful because then there's something for us to try to explain. And then when it comes to web 2.0 related terms, I mean it's interesting if you look at these side by side that certainly web 2.0 related terms are much less understood in this group. So a lot of things that I think a lot of people in this room probably take for granted that we know what it is. A lot of these people don't really understand. So we have a scale of 1 to 5, and a lot of these things score pretty low. So people don't really know what these things mean per se. Now you could argue, some might not know the term fishing, but they would still recognize a phishing email. If you ask me later in Q&A, I could tell you more about why I know that these students don't really get phishing, period. But we don't have time for that right now either. OK, so on this variable as well, we have considerable variance. So there is variation to try to explain, which is helpful. OK, so then what I did was I wanted to see what predicts differences in scale. And what I'm going to do is not show you the actual regression output, the numbers. But if anyone's interested, I have them. So later I can show you the numbers if people are really intrigued by that. I'll just show you the overall of findings. So basically what I'm doing here is using regression analysis where skill is the outcome and what I find. And so first I just look at gender, race, ethnicity, and parental education to see if any of those predict people's level of skill. And what I find is that women claim lower level of understanding of these terms as do students of Hispanic origin and African-American students. And then lower SCS also relates to lower knowledge. On the other hand, compared to whites, Asian-Americans tend to know more, especially with respect to the basic skill measures. But also, even though we're controlling for basic internet access and experience, it turns out that the more types of access you have actually does matter in how much you know about the internet. So autonomy, I define as freedom to use the technology when and where one wants to, recognizing that, well, you might have a computer at home, but if you share it with five siblings and your parents, that won't be quite as autonomous as if you have it at home alone, or not to mention multiple locations then. And then in this case, both how many years you've been a user and how much time you spend online matters for knowing more about the internet, which makes sense. Another way to think about differences in internet use is types of things people do online. Now, this is one of those things that you could measure in many, many ways. And I have, and I'm just going to show you one way of doing this, but this is another thing we could talk about at length and all sorts of classifications. So these are of the activities that I ask students about, which includes the things on the bottom of the slide. The top are the things that more than half of the students do at least weekly, which really is a mix of recreation or entertainment types of activities with also what we can call capital enhancing activities, so things that students are probably benefiting from and learning from. Not to suggest that from entertainment people wouldn't be learning, which is why classification of these activities gets complicated. So I'd like to make the argument that the more diverse ways in which used the internet might have beneficial outcomes. So I create an index of types of things people do weekly, and then I'm interested in what explains diversity of uses of the internet, at least using this measure. And so again, using regression analysis, I show that women tend to do fewer things online as do students of Hispanic origin. And again, there's an SES relationship, so people from lower socioeconomic status backgrounds do fewer things online. However, similarly to what predicts skill, Asian-Americans do more things online. And also, if you have more autonomy and more experience in terms of how much time you spend online, you do more things. And then skill shows a positive relationship with diversity of uses. So people who know more about the web do more things online. Of course, the causal arrow here is complicated, and it probably goes both ways. OK, so now that's one part of the findings. I wanted to tell you about the participation gap and what we're finding there. I think this is an issue a lot of people here have thought about a lot, so I'm not going to talk in detail as to why we think this is important, but it's the next step where differences could occur, and that's why it's of interest. I could definitely talk about this in more detail if you'd like me to justify this study. But so what we're interested in here is who creates content and who's actually sharing that content, right? So people have always been creating content, and there's considerable literature on arts participation and creativity, and so that's not new. But what's new is that the barriers to entry in terms of sharing that with people have really been lowered. So we're interested that if you control for having access to digital resources, then who's actually posting the material that they're creating. And this work I'm doing in collaboration with my graduate student Gina Malekko. OK, so first, this is what I call my reality check slide, which is just to take a step back to realize how widespread or not some of these resources are, because we tend to get very excited about these possibilities, but it turns out that people aren't necessarily using all of these that much. If we had more time, I'd start having you guess certain percentages, but we don't have time so we won't play that game right now. But so basically, in decreasing order of popularity, so you have a few of these sites that are quite popular in this group, but then very quickly you drop down to very low numbers and not too many people in this sample using some of these sites. And remember that these are 18-year-olds, so if anyone is going to use some of these sites, you think it's them. So the fact that they aren't leaves wondering how widespread some of these services are with other populations. Does use mean visit? Does use mean anything? In this question, the way this question was phrased, these figures represent either occasional or frequent use of this service. So it's not, have you ever been there? Do you use it sometimes or often? OK, so this is where, again, I could show you more information, but I think I only have two minutes, so I'm just going to skip to the findings. So basically, we ask people, do they create content and not restricted to digital content or online content, but are they creating content? And so we got information about that. And there, the only, so if we look at gender, race, ethnicity, and parental education, what we find is that the only statistically significant predictor of engaging in creative activities at all is parental education, and this is completely consistent with the arts participation literature, so that's not surprising. However, what we then asked, which is new and quite unique in our data set, we don't know of any other data set that really gets into this, is then who shares. And the way we asked about this is we asked people if they post. So we didn't ask, are you sharing this, say with a friend? We asked you, post videos, you post music. And so restricting this analysis just to the people who are actually creating content, what we find is that the one statistically significant predictor is gender. So only looking at people who actually create, it turns out that women are less likely to share than men. No effective race or ethnicity or SES. However, what's really interesting is that if you then expand the analysis and you add information about skill to the model, it turns out that it's actually skill that explains the variation in whether people post or not, and that actually obliterates the effect of gender. So if you have a woman and a man of the same skill level, then they're equally likely to post. It's just that women tend to exhibit lower level skills for our measure, and that's what actually explains that they're posting less. So I'm happy to chat about that more because I think that's a really interesting finding. And then finally, okay, what I was gonna show you, but I really wanna keep it to 20 minutes, so I won't, but I've also done, what? Live a little, come on. I just, I don't like it when people go over, so I don't want you guys to get annoyed. Go over, please. I'll put it on the answer there. Don't stop. Okay, well then, just this one more paper that we have out of this so far that I wrote on, and this is actually, this should be out, I think, tomorrow, the special issue that Dana Boyd and Nicole Alston are editing of the Journal of Computer-Mediated Communication on Social Networks sites, and I have a piece in there that looks at predictors of using different social network sites or whether people use SNS in the first place and then which one they use in the sample. And so, I look at the usual suspects that I've been talking about and actually here, John, I include the information about living at home because it turns out it matters. So, first, if we aggregate any SNS, so we just ask about it, SNS use, it turns out, so women are more likely to use a social network site, and people who live at home are less likely to use a social network site. But then, oops, that should, well, hopefully you can see the, you can be blackened out, but, so yeah, Facebook, MySpace, Zanga, and Friendster that I also looked at, and this is not the most informative slide I realized because I should at least have direction of the relationship, but basically all I'm signaling here is what are the statistically significant relationships, and so, okay, so just to give you some information that's not really apparent on the slide. So, basically, for which I apologize, other than Zanga, for all social network sites, women are more likely to use it. I guess before I get into the details, I should say that partly what's significant about disaggregating by type of site is that there really are different predictors of who uses which sites. So, most of the literature either just looks at one site, but then we need to be careful not to generalize from that one site to other sites, or it's just generally a question about social network sites. But then again, that might actually hide some more particular relationships that you find if you go by service. So for example, what I find is that Hispanic students are considerably less likely to be Facebook users in this group, and considerably more likely to be MySpace users. Also Asians and Asian Americans are more likely than anyone else to use Friendster, and one idea maybe that Friendster is actually quite popular in some Asian countries, and there's a considerable immigrant community in this sample, so they might be connecting to a family there, that's one idea. And then SES relates, and those of you who know Dana's work in this realm, it's actually similar to what she's finding, so people of lower SES backgrounds or students in this group are more likely to be MySpace users, and students whose parents have higher levels of education are more likely to be Facebook users. So it's not random as to who's using which service. And then finally, living at home, it's really just for Facebook that it matters, and basically if you live with your, actually here when I say home, this isn't clear given what I talked about before, so this means living with your parents, I apologize, this isn't clear, I just realized. So living with your parents, compared to any other living situation, say living with roommates or alone, if you live with your parents, you're considerably less likely to use Facebook. And I think this is interesting because, so going to college, so the way I tie this into digital inequality discussions is a big part of going to college is creating networks, meeting people, hanging out. If you live with your parents, you're probably doing less of that because you're at home. So the idea would be, well, but there's Facebook, you could at least keep in touch with college networks through Facebook, but not if you're considerably less likely to be on Facebook in the first place if you're living with your parents. So I think that's an interesting story. So just to wrap up, hopefully I've made a convincing case that there are skill differences when it comes to internet uses and skill differences actually do relate to what people do online and how they make use of their access to digital media. And so it's important to remember that beyond thinking about giving people access that we need to think about ways of providing training and support. And there's some of my MacArthur, one of my MacArthur projects right now is actually thinking about this further and I can talk about that as well. And papers, although not what I just talked about, but other papers leading up to all this are available on our newly launched lab website. So feel free to check that out. I noticed when you were measuring use of social networks that you were measuring race and ethnicity, but not income level, you know, and at parents' educational level, do you think perhaps if you put income levels in there somehow that you would find that the young people living at home were also the lower income demographic that had to do with what social networks that they were? So I'm using parental education as a proxy for socioeconomic status. I can tell you as a person who works for non-profit sets allows you to predict your own. Okay, well, I can tell you as a sociologist that it actually is used in the literature considerably and works reasonably well. And as well as we can do that proxy, it's also certainly the case that with this population trying to ask about income is not very useful for several reasons. One, if you ask about household income, which is what people usually do, it's very hard for people in this population to even interpret household similarly, depending on whether they live with roommates, if they live with parents, what does that tell you? Not to mention that 18 year olds are not reliable sources of information on their parents' income. So unfortunately, we can't really ask about income. We collected zip code information about their last residence before they moved to campus or where they live now. And we've tried to use that by taking medium of the zip code area, but it's just, it's a really difficult thing to approximate. So we're going with what the sociological and social science literature has suggested works. Okay, yeah, I'm sorry. I didn't mean to be sharp on that. It just seems to me that some of the stuff that you're doing with the people who are living at home is likely to be more about income than ethnicity. Yes, I agree with that. And I didn't make any claims about whether it had to do with ethnicity or income. It's a regression analysis, which means that all the other factors are held constant when you're exploring the relationship of that variable to the outcome. So the fact that we have findings for people living with their parents means that at that point you're holding the other factors constant, so it shouldn't matter actually. Unless you wanted to argue some interactions, which we could put interaction terms in and explore that. One thing that's great about what you've done is you've taken up a number of things. Right. I'm sure there'll be specific questions about it. What surprised you on that side? But what, coming into this with a bunch of hypotheses and then coming out with data, where were you surprised? I think there are all sorts of issues. I mean, I think I really, whether I was surprised or it's just something it hadn't really thought of. So I really hadn't thought of this whole living with parents, how that would actually influence social network sites. And I think it would be really interesting to do follow-up work in that arena as to how social context matters. We also find actually, I don't show that here, but in the detailed regression analysis, you do see that students who have access to a network-connected machine either at a friend's or a family member's house are also more likely to do some of these things. And so it really shows that the context of your uses matters. It's just really hard to quantify some of that. So I think that's where we really need a lot more work done. In terms of the gender, well, so the creativity and posting, I wouldn't have expected that gender would be the only predicting variables. So I thought that was very interesting. For me, it's great that skill tells so much of that story. I mean, that's my bread and butter setting skills. So when I ran the analysis and it came out, I thought, oh my God, you know what I'll say, I tweaked the data or something. But on the other hand, I have to say that I can think of other reasons why women would be less likely to post, and I'm not measuring those when they're not included in the analysis. So I'm not, I don't know if skills should really mediate the effect of gender completely. So I found that to be a little bit surprising. I actually would have expected gender to still remain a significant predictor to some extent. Just knowing there's a lot to grab on, can we stick on this sort of slide for a minute? Should I actually bring back the slide or the this one? Okay, yeah, or yeah, and I'm sorry. Well, actually, if you guys wanna see a little bit more, I also have a little more on numbers, I don't know if that's an interest to people. I had to- I'm gonna pull it up and then Jean will have the first question back. Oops, that's not, I'll get there. Oops, sorry, that's not where I'm trying to go. I want to follow those presentations but then they should- Endless. Well, I have other things that I just knew I didn't have time for, but where did I put the, those are the state outputs if anyone's interested. Oh wait, did I actually erase those slides? No, here they are. This is more detailed, but if someone had a question, I don't have to get into the detail of this, but this was just a little bit more detailed as to what are the numbers for people creating things and then what are the numbers for people posting. So basically two thirds of the people in the sample are actually creating something. But if, and if you look at aggregate, it's two thirds of women and men is pretty much the same, but so men are more likely to create music and film and women are more likely to create, to write. And then for artistic photography, not statistically significant difference, but women are just slightly more likely to, but well, the wrong numbers. Anyway, and then for posting, these are the numbers we have, which I'm just showing you so you have some sense of how popular it is among people. So makes sense that, so one would assume that writing is the easiest to post, so it makes sense that that's, if you control for creation, that that's the one people are posting the most, but those are some additional figures. And then here's when we have it by gender and what's statistically significant. This is not in the regression, this is just the raw, my variant. Just to make sure I know, if online sharing by gender doesn't mean how sharing has to be mixed up with somebody else's thing. No. So the, yeah, the way we ask about it is, so it's either your own video, you posted your own video or you posted your posted video, you'll be mixed from other people's work. Yeah, we don't. So in each case, we think it's clear. I mean, you never know what the person's thinking, but it's in the text of the question that it's posted your own video or posted video you'll be mixed from other people. So it's not, yeah, not something you just stole. So Gene and James. I'm curious if your model or your definition of skill includes the idea of attitudes and perception and whether that is, and if not, if it's an independent variable, whether that's a variable, that's kind of a compounding variable in terms of both, for example, this distinction you make draw between, this possible distinction between men and women, whether women might be reading themselves differently more because of the perception rather than actual difference in the skills or whether, and whether an attitude or perception of your own efficacy or your own talent might lead you to, might have a different impact on your willingness to share something rather than your actual skill level. And so whether the model includes some. Right, so, if you hit the nail on the head, this really is the very complicated part of the skill measure. And when I mentioned that earlier work I'd done is how I developed this measure, what I was referring to was that earlier I had, and actually I still am with this group, with a portion of this group, 100 people, we do observations. I mean, that's what I did years ago is I actually, I did observations one-on-one where you see the actual skill, right? You can really measure that. And then I correlated that with various survey measures to see which ones predicted the actual skill best. And that's the measure I use here. Now that doesn't mean that it's a one-to-one at all. It's really difficult to find better measures. I don't, I'm starting to look to other parts of social science that have nothing to do with internet use, where people might have tried to deal with this. Methodologically, it's really tricky. We definitely know that women tend to estimate their skills in a whole bunch of domains lower than men. And it's not clear for most of the work that's been done on this is whether it's women who underestimate their skill, or if it's men who overestimate theirs, or if it's both. But the point is that there is a discrepancy. The point is that there is a discrepancy. So there's definitely a discrepancy. And obviously it will influence this measure, even if it is better than the traditional measures. All that said, I think, so one of the things that I advocate is that we should have training. And I think that sort of addresses it potentially almost either way. Because whether it's that you really don't know how to do it, or you think you don't know how to do it, hopefully if you get more training in it, whichever might be going on, hopefully would push you towards doing more of it. So that's the best way I can address that, but it's a huge challenge of this area in my work in particular. Thank you. Have you looked to see whether this, how this relates to the general creativity of the same cohort that is perhaps in a broader sample, not just college age? And is there any sense that these online sharing systems are actually drawing out more creativity about from people who might otherwise not try their hand at some of these artistic areas? Yeah, so that's a really good question. So the first part as to, can I compare this? I don't know of other data sets that have even come close to this level of detail about this issue. If anybody does, I'm really eager to hear. But I mean, I do a lot of work trying to follow who's doing survey research in this area. And I just don't know anyone who has done, especially for any generalizable sample. So I'm afraid, no, I can't compare. The other question is a very interesting one. And I think very relevant because it is certainly the case. But I think in that case, we would almost need to have data from quite a while ago. So I think there it would need to be longitudinal comparing to times when these opportunities were not available or really find a population where the opportunities are pretty low to compare. So I do think that it is probably the case that some people are creating now because they have the potential to share. In fact, I think that's where gender might quite a role. So that I think there may well be especially men who would not have created things before, but now create things in the first place because they know that they can then share that. I can't really address that with this data set, but I do think that that could very well be going on. And it's an interesting area, yeah. Max, sorry. Oh, I'm sorry. To go into something JP, I guess briefly touched on, I wonder if any of your data looks at the illicit side of teenage computer use in the sense that, I had friends who knew how to run a torrent before they knew how to edit Wikipedia. And just the idea that for a middle income person, they might actually gain skill level in order to get music for free rather than have the disposable income to pay for it. Is there any, I guess there's a problem in having people fill out a survey saying yes, I do illegal activity online, but is there any way to see if their skill level is being increased by these activities that are less than ideal? Right, so yeah, we don't really, yeah, we don't ask about things like that partly to avoid IRB issues, partly because it's just not really been the focus of my work per se, although as you say, it could relate to skill development. I mean, torrent was one of the words we asked about just since you mentioned that and the level of understanding is extremely low or maybe, did you not see it on the slide? You didn't? Was it, okay, okay, yeah, because I thought it was on there. And so that, for example, is quite low so I don't think that's necessarily driving much. And it's also, I mean, another issue is, this is an issue that requires more thought and work. How do skills in certain domains translate into skills in other domains, right? So let's say you do become a really efficient torrent user or who knows what exactly is sharing in certain ways using P2P or whatever. Does that mean that you are also, does that really translate into certain other skill domains that might work to your benefit in other ways? And I don't, I'm not sure and I don't know of any work on, I mean, there isn't too much work on skill, period. There's not gonna be much work looking at the details of skill, but that's one direction that where this could go is to see how these things translate. I have a question about non-jewsers and I wonder if you're also ready to find any person that doesn't use internet and why? She's not using the internet. Why is it presented, should it happen? So as I said early on, I focus on users and the reason that I study a university population is because a school like UIC everybody is going to be a user. The university life pretty much requires it and you get an email account. So no, I don't, I haven't really done work on non-users in years. I don't. Oh, are there people in New York? I have two questions. First was, if somebody made a comment on someone else's site, would that be shared in some way? That's a good question. I mean, it's not, I think that would be a stretch. It's just not the way we phrase it because we say, have you posted, have you posted any of your poetry or fiction? And I just don't think people think of their comments that way, but I mean, I guess I haven't studied in detail whether it's- Comments are probably excluded. Comments should be excluded from this. Yeah, on a different part of the survey, we do ask about whether specifically with respect to people's closest friends, how they keep in touch, and one of the options there is whether they comment on their blog. But that's sort of, that's a whole separate paper that I'm not ready to talk about. I'm not like, that is the way I have to talk about it. My second question was, what struck me was delicious there? 2%, you know, one of the simplest apps to use and one of the most obvious, what the benefits are for every user. And why is that so low compared to something which is quite complicated which is high? You know, it's funny because you say, and I agree with you, you say that the benefits are obvious or something, and it turns out they're not. When I talk to students at Northwestern, where you might predict somewhat higher use of some of these services, I mean, it's definitely a much more elite group of students who've had much more resources over the years. People have absolutely no idea what I'm talking about when I say delicious or if I say social bookmarking. And then when I explain it, and maybe I'm doing a horrible job but I don't think that's the case, they still don't really get why you would do it. And then if I go into a 10 minute explanation, eventually they get why you might want to do it. But so I actually think part of the issue is that people don't get the relevance either to their lives or other people's lives or just in general, what's up with some of these services. I mean, the other thing is Second Life which I also try to explain because that was even less used and less known. And people laugh and they mock. It's really fascinating. And then I have to come up with examples of their lives that to the next person would be pretty mockable but to them has meaning. And so that's usually a little, that's helpful in getting them understand and open up a little bit as to why these services maybe as well. The YouTube number was real high. Does that mean they've made videos? No, no, no. That just means that, no. That just means that it's a site they visited often. So that definitely doesn't mean that they're posting, yeah. So if I were looking at the things that drive lots of usage that presumably then produce skill in use of the internet, which is kind of what I gather you sort of done in identifying what the activities are. And if I were looking for an activity that I know commands a lot of usage and that might have a gender differentiation, I'd probably pick pornography. In other words, if you looked at the amount of time that males get exposure to cutting edge internet stuff and access to all the different functionality, I would probably, based on everything I've seen over the decades I've been involved in the web, I'd probably say that porn usage probably overwhelms two thirds of the other characteristics you've looked at for web usage. And you might find a huge differentiation between gender and around that. What you might find is the training that's going on is happening to go back to the point of illicit activity. It may turn out that what you've done is focused on like delicious and things that may have utility, may be interesting, may not be interesting, but there's a whole bunch of those. But you may have missed actually one of the key usage drivers, the trained skills. Right, well, that's a very interesting point. I mean, I have to admit that I don't have much experience with the recent innovations in porn online. So I don't really know what it, to me, maybe naively, I just think porn is probably viewing images and maybe that's really naive. And I don't, because just in viewing images, I don't necessarily understand what would be the skill transfer, but again, I don't really have much experience. Every generation of the net, we knew this at AOL, every generation of the net, the earliest adopters, whether it's flash, whether it's social networking, the earliest adopters, because that's where you generate the large usage, turns out to be pornography, right? And in fact, it's not, I mean, you know, again, speak about a gender difference. So if you say to me, and you're the prime investigator of these issues that porn is just looking at images, then I would say to you, you probably ought to spend a little more time. Okay, I'm sorry. It's okay. It's okay. It's okay. It's okay. It's okay. It's okay. It's okay. It's okay. It's okay. It's okay. It's okay. It's okay. No, I appreciate that. And maybe I, I mean, you know, it's one of those tricky things of how to actually pursue in terms of, well, anyway, so I'll think about how I can become more educated in that realm. I mean, I could see how, I see where you're going and I can see how even just the goal of finding the latest could enhance information seeking skills or could enhance collaboration or, I'm sorry. Chatting online, you take the advanced functionality of every single thing you listed on that. The earliest instantiation of a lot of that functionality tends to be in porn. No, it's a really interesting point. I mean, but yeah, it would, another issue of course with studying porn is beyond IRB is how forthcoming people are with that. In an earlier study that was on older adults as well that I mentioned, I mean, it was a small and it was just a hundred people. I did ask about porn and got pretty low numbers. So not what one would suggest. So that suggests that with respect to that particular activity, people are not as forthcoming as they might be. Well, these were actually not students, but yeah, I don't think you want to tell anyone too much. Back to us, I was saying to your comments that, you know, the skill of all fired and looking at pictures and whatever the iteration is now, the back door to that is, do you know how to clear web history? Yeah, that's true. Yeah, no, absolutely. Especially if you're living at home. Right. Very sophisticated people who are very good at using TOR or an anonymizer or something that, you know, there might be correlations between, would be considered, you know, not subterfusion, counter or whatever we're calling it, technology that comes around from an unusual source. No, I think that's a really good point. And I think, I mean, the more I think about it, just even sitting here, I'm realizing the implications. So I'll definitely think about it more. I mean, I gotta, yeah, I don't know how much I'll educate myself on some of the specifics, but I think it's definitely, and I mean, I think you make a really good point, even in terms of cookies and this sort of, you know, yeah, I mean, I think that's a very good point. How do you think, and gambling? These are two of the reasons people use it the most. So I think you'd find some of the supportive amounts of the person who went to our faculty. The other is that people can stop at where a team, next door, just clearly tracking the way in which people get computing stuff for exactly this reason, you know, want it. And so you're able to, so that's the case, that's what people say. So I think that would also go to the skill thing. So then there might be some very interesting evidence for sure. And I actually do ask about gambling, so I could try to look at some of this with gambling. Sally, then Ralph. Foreign conversation, but I'm sort of curious in all the, the question that you posed at the very beginning, which is, what kind of feedback cycle might it be? What kind of data are you looking to generate or what works? Mm-hmm, yeah, so for the feedback cycle, I mean, partly that's just theoretically, you know, what are the processes that might be going on? And I'm happy to talk about that in terms of data. So that kind of work absolutely requires longitudinal data and we don't really have longitudinal data right now. Fortunately, the MacArthur grant that I just got with my colleague Peter Miller a few weeks ago is a planning grant to figure out to assess the feasibility of a longitudinal study on digital media used by youth. So this year, what we're doing is taking a survey of what work, as in not a survey data collection, but surveying the field, what has been done to figure out how that could be pulled off, whether it's to work with an existing longitudinal study, because I mean, social science, there's some extremely high quality studies that have been done. Unfortunately, they're usually not on youth and they're usually already oversubscribed. So, I mean, the best you get with an existing social science survey about internet is to use the internet. You're lucky if they include a question like that. Never mind any details. The current population survey of the census hasn't asked about this since 2003 at all. Their last report was a nation online, so supposedly by then we fixed all the related issues. There was no need to study it anymore or measure it anymore. So we're just really lacking data. And so hopefully with this paper that Peter and I will write, we'll be able to chart out what's realistic, where we can head in that direction. Now, in terms of the processes, I think the idea is that you can enhance various types of capital for people, right? Whether it's human capital, it could be financial capital, it could be social capital, it could be cultural capital, and depending on the different types of uses, what you're doing online could enhance whether it's your skills that you can put to a job or whether it's ways to figure out great financial deals or whether it's just learning about things that you can then, I mean, cultural capital can work in not completely obvious or ways, but that can definitely be very important as people have shown. So those are all the various processes by which this might matter. But in order for us to know, we definitely need longitudinal data. Mr. Longitudinal Data himself. I have no longitudinal data. This is great stuff. Thanks. The thing that's I'm puzzling over in this is the direction of causality between levels of skill and the motivation for wanting to do something. Do I know how to do this because I wanted to do it and I went and figured it out or vice versa? And how do you look at that? Yeah, that's a very good question because directional causality in this whole puzzle comes up all sorts of places and it's very hard to disentangle, again, especially without longitudinal data as well. I think, I mean, in some ways the way I deal with it is that, so I guess, well, I guess your question could be motivated by different interests here. I don't know if it's- I was looking at the perspective asks, I mean, do you want to then assign people in class posting a video on YouTube to bring their skills up or do you want to send them to the computer lab to give them extra remedial work on computer usage? I guess that's kind of what's- As well as just understanding things better. Right. I mean, do we have, we controlled for indigeneity when we're drawing our inputs? Right, so I think, I mean, I guess the way I address this to some extent, which isn't to say that it doesn't matter because it certainly does and it's interesting and it would be important to know even just to know how to spread skill. But okay, so it's true that if someone's just not interested that could lead to different types of, say if we wanted to do intervention that that would require a different type of intervention from the person who's interested. They just really don't, they don't have the networks to go to, right? So that's a different scenario and you need to deal with it differently. And so that does require figuring that part out. At some level, as far as I'm concerned, I mean, I don't wanna say it's secondary because obviously it's relevant, but in some ways, well, okay, let me say where what I'm about to say is coming from. So some people argue that they're folks out there who just don't care, they're not interested. If they don't wanna do it, leave them alone. So I don't, I think it's more complicated than that because as more of society moves online, as more government services and very basic services require that you go online and that you be able to access and make sense of what's online, it's not optional, it's not a luxury good. So in some ways, there are other things we teach people in society and just because they might not wanna learn to do it, we still believe as a society that it's valuable to the greater good that they should know it. And so the question is, can we fit this into that category? And again, as more and more of life moves online and more services require the knowledge, I don't think if it should necessarily be dependent on whether the person actually cares to or not. Now, of course, that doesn't mean that. I mean, that still means that if they don't care to it, it will be harder to, it will potentially be harder to get them to learn it. And so, yeah, I don't, I mean, maybe the short answer is that no, in this data set, I don't have anything on motivation and I don't really know it. And I'm not saying it's not relevant, but I think some of this story is relevant, even regardless of whether it's a motivation to show or not. As a plug for a Harvard Law School class, I'm really intrigued and excited by your, what seems to be the ultimate objective of this study, which is to find out what the impact of these different levels of skill are on life chances. And I'm curious whether at this point you have a hunch as to whether the dividing line is going to be along the lines of technology or whether it's, there's also just the basic skills of knowing how to do, for example, social networking in any medium, whether it's face-to-face or online, whether that comes with skill or attitude or whatever. And so I'm curious at this point, whether you're looking for the independent value of knowing the technology in terms of being able to do things like social networking or whether it's kind of, these things are kind of glommed together as kind of technological social networking, which is different and kind than maybe face-to-face or just using technology. I think they're intertwined. And again, that's why we need longitudinal data, because then you can actually control for some of the initial ability to do some of these things or the initial propensity to do some of these things, which is one reason to try to study young populations, because then you can actually collect data early enough that some of these wouldn't be that widespread yet. I think, I mean, where I think this is going is, so I don't think that the technological ability will take over those things. I think it will interact with them. And in fact, I think what will happen is that if anything, and this is where it is a potentially troubling outcome, is that I see the differences going like this, right? So if you are good at networking, then you could really make a lot of something like Facebook or whatever the next great such services. Whereas if you're not, then it's not really helping you leapfrog your shyness or inability to network, it's just you stay where you are. And so ultimately the gap will actually widen potentially. That's how I think it's gonna go, but it's hard to have the data yet. I'll ask one quick other question, which you won't be able to answer. But I'd love to know how to even be able to do it. But I guess the question for me is in the network space, do you have any sense of whether the variation within usage is higher or lower than the variation in all the other explanatory variables? It's the same question as, is this a leveler right now, as you see it, or is this another tool for greater divergence in socioeconomic status and power? You mean digital media in general? Yeah. So I think similarly to what I just said, I do think it's gonna create a greater divide because I do think that those who have the opportunities, for those who have the opportunities and have the skill, there really is so much out there. I mean, I don't question all the opportunities. In fact, I take advantage of them all the time, but I'm a PhD, I'm not exactly the person who needs the extra boost, right? Whereas it's the people who don't have the education, who don't have the networks to figure it out, who aren't going to be benefiting, which isn't to say there aren't exceptions. I'd love to, I mean, partly what I think would be really great is to find those exceptions and figure out what happened there that they actually were able to take advantage. I think that would be really fascinating. But overall, my answer is that I see it going like that. Does the data say anything about that? No, because it's not lunch at all, so. Right, right, so there's the place. Thanks. In the piled data that you're checking for in sites, what's going on to use to move or emailing their friends that's very hard. Maybe yesterday you talked really briefly about the mobile. Sure. That's my question, so don't mobile. Well, in fact, we usually let people go around with their phones, and it's perfect as a last one, but I'd love to hear comments about what's going on. Sure, yeah, so I mean, quickly in terms of this survey, we actually do have a little bit of information on some other types of time use, especially having to do with social types of activities. But what John is asking me to talk about, so got a small grant from NSF to do a bit of supplemental work, collecting diary data using text messaging from the students, and the idea is, so there's been lots of studies that have used beepers or specialized PDAs, but what we really wanted to do was be as unobtrusive as possible in collecting information about the daily lives of students. And the way to do that is use text messaging not only, so 98.5% have cell phones and a very big percentage use text messaging already. So what we're doing is, so we wrote this work with Chris Carr, wrote a program where we ping them. This could be randomized, but what we did was just so far just a pilot. So every hour on hour 15, we ping them for a text back to us. We told them ahead of time that we wanna know four things. Where are you, who are you with in terms of number of people and gender of the people? What are you doing? Just anything, right? And then in particular, tell us about any digital media that you're using or any communication you're engaged in at that moment. And you would think, well, is it possible in 150 characters to convey all that information? Well, these 18-year-olds are very good at this and it's actually amazing how much they're able to give us in 150 characters. We don't have enough in this grant to really pursue this. It's a fascinating project that we really wanna take further. We're trying to figure out, I mean, partly what's really challenging is that it's open-ended, so how do you code that? So we're trying to figure out something with computational linguistics. Is there a way to code it automatically? We're not there yet. But part of the goal here was to see, so okay, we have the surveys, but if we actually ping them all day, how many times do we get responses where they are listening to their iPod or where they are using Facebook? And what we're finding is there are a lot of, I'm sleeping or I was just sleeping or watching TV, and again, this is a very particular population and we'll try to redo this at Northwestern and it'll be interesting to see what differences we find. But yeah, I'm very excited about that project. I think there's a lot of potential there and it really, it turns out it is quite unobtrusive, so they open up very much in terms of what they tell us. So it's very exciting. That's it. One last one. I can't hear you. You need to speak up here, sorry. Or stand up to us if you're going to be. Just thinking about cell phones, the 98.5% of the students have cell phones and thinking about the kind of widening of the speaking gap that you're thinking is gonna happen in terms of what services. Kind of throwing up the idea of all these services going mobile and that mobile phones are something that seem to be being adopted across the board kind of with the fact of video sharing, photo sharing, everything, so it's never gonna go to mobile. Yes, except, so we did collect detailed information on what people are actually doing with their cell phones and text messaging is very widespread in this group, but when you get into specifics of, say, video or web or email, there's actually, so I'm just working on this paper as well, that the whole digital inequality story, you can tell with cell phone uses as well, so higher SCS, more likely to be using cell phones for more things, so that type of digital inequality story actually also transfers to that. What year did you collect this data in? This year. This year. Fresh up to the minute. First time since I've seen. This semester, and we will definitely send this link to your program officers at NSF, but I'm not ready to say clearly the longitudinal study you've described is doing clearly your mobile work as huge methodological resilience, I think, into the future, so we very much hope you get to do this work in one way. I do too. And as those of us who are not social scientists, but relying on what you find as we make our policy arguments, we need it. So that's both for what you've done and for what you will do. Thank you so much. Great, thank you so much. Thank you. Thank you. Thank you. Thank you. Thank you.