 I admire your work and I'm very pleased and honored to be a part of this task force proceedings. I need to give thanks to some sponsors of this research. I will tell you that in Canada the research has been cobbled together. Non-contributing members but in name only are the Information Systems Security Association, the FBI's InfraGuard program, which is about raising public awareness for infrastructure protection. Of course the National Center for Missing and Exploited Children. Time Warner has given us mostly in-kind support and 5,000 in-cash. Arbor Network supported our research with 5,000 in-cash and Semantic did too with 5,000 in-cash. But most importantly we had upwards of 20 school districts each contributed a precious $2,000 so that we could carry out this research for a total cost of $65,000 all of which went for graduate assistance. The number of professionals that are involved in this effort number in the dozens. Today what I want to do is cover some basic points and overview of our work beginning with the headlines, the major takeaways so that we're not bearing our message in lots of details. I'm also going to come right out and tell you what the limitations of this research are and then I'm going to brag about what the results are because they are mind boggling in terms of the massive amount of data and the breadth of what it is that we covered. I've been asked to focus several of my remarks in areas of cyberbullying, lying in deceitful behaviors and sexual offending and then I'm going to provide some observations of contemporary digital youth culture particularly with respect to alarming aspects of what youth and young adults are up to today and then hopefully there will be some time for your questions and comments. I'm getting started a little bit late so I'm going to go very quickly through some of this information but you will have access to this briefing through Andrew and Dana in short order I presume. Major point number one, today's students of all ages experience many forms of cyber abuse. They're listed here for your reference and it's really important to understand that they experience these activities as victims and as offenders. When I use the word abusers or offenders I mean essentially abuse of information systems either in violation of an AUP acceptable use policy or in violation of a state crime law and so forth. Point number two, cyber abuse, crime and victimization appears to be a function of the number of types of devices used to access the net. The amount of time spent online, types of online activities and levels of social computing engaged in, prior successful offending experiences, previous victimization as in retaliation and so forth and declining age of onset for engaging for any purpose online. And when you think about it ladies and gentlemen, that's pretty intuitive. There's no surprises here, right? The other thing I want to say in qualifying my work up front is that I'm not the bad guy here with respect to social technologies that enable social computing. I'm not poo pooing the internet. I am certainly not saying that you should not be using computers and other types of IT devices to engage each other online, facilitate their education and so forth. But I am going to tell you what the facts are based on this research. I will conclude, hopefully there will be time to go into a little bit of depth about contemporary youth culture and the fact that there are aspects of what our children are doing today that are very alarming if not disturbing in terms of civility and underpinnings for civilization itself. That is to say trust for commerce, for relationships, for education and all of the long-term investments that we as a society need to think about. They are not being checked in substantive ways with regard to education, supervision, and role modeling enforcement and technology, right? That's not an indictment of anyone in this room. It's just our lack of understanding as a society. Dana has cautioned me not to slip into any kind of a policy mode. I'm trained as a public administration person and as a doctoral policy analysis type in public policy and as a science and technology social scientist. So I need to make that comment up front to sort of qualify what it is that the purpose of this work has been about and what it has involved. So how dare I say these things? What is the basis for it? Well, very quickly, let me just tell you that as far as I can tell, it's the largest most comprehensive cyber crime offending and victimization study ever conducted. It involved 14 school districts in upstate New York, a convenient sample of over 40,000 kindergarten through 12th grade students constituting approximately 50% of the student population in those school districts and they were not selected in any way other than that would be consistent with the results of random sampling. In other words, we didn't pick this school building or that set of students. It was a matter of operational convenience at the school district level. Some districts chose to sample all of their students. Some said, nope, we're going to do stratified random sampling because of the impact on our operations and we're only going to try to do 10% and so forth. But it's a huge sample, right? We employed approximately 50 experts from these school districts and other academic institutions including the Harvard Medical School Division on Addictions, consultants from the University of Rochester and various people with a variety of expertise, child psychology, human development, school psychology, sociology, criminology, public policy and so forth. We really tried to get all perspectives in this thing. As you can see here, we were able to engage in a passive opt out parental consent process via US mail and this was an online survey. I'm very pleased to say we believe it's actually a pioneering survey with respect to engaging talking computer, that is to say audible surveying instruments for the younger children in kindergarten through third grade who cannot reliably read much less, understand the meaning of questions. And that's pretty cool. These are a list of the school districts that participated, the number of in the population during the time in which they surveyed between May 2007 and January 2008. The N, as you know, stands for the number of respondents and this is the number of respondents pre-child ascent. So in addition to parents providing opt in permission for their child to participate, every single child according to state and federal laws governing human subjects requirements also had a child ascent where they voluntarily agreed to participate in the survey and they could drop out any time they wish. We believe that based on US census data and what we know about other research that's been done for a long time in Rochester that this is generally representative of the US population. I'll qualify that at a moment. So let's jump to some survey results here. Well, that's okay. Let's just breeze through this real quickly. What you're seeing up here is essentially consistent with the Pew American Life Internet Survey There's no real surprises here. The large point is that very young children use a variety of devices to do various things online and in the process they have experiences which make them uncomfortable and they generally or oftentimes don't tell parents about those feelings and experiences. At the second through third grade level, which was our second cohort out of five different student age grade cohorts, similar findings, notice that we surveyed over 5,000 children in this age group, right? And you've seen these kind of statistics before. The punchline is, again, they use a variety of devices to access the net from home in various places within the home and when they do, they engage in a variety of activities. Also at the second and third grade level, we have solid evidence of victimization and offending experiences during the previous school year. All children were asked about what they experienced in the previous school year so they have a basis of reference, right? And I'm going to let you just read these statistics for yourself for a moment and then I'm going to flash to the next screen. Let's move ahead to the fourth through sixth grade level. They're interacting in a variety of ways, IAMing, email, chat rooms and so forth. And by the way, when we think about social computing, we're interested in gaming and all of these other ways in which children interact, right? It's not just about profiling web sites. And in case there's any representatives from MySpace or Facebook in here, everywhere I go, I emphasize the reality that there are dozens and dozens of these kinds of sites but those too have been demonized for good and inappropriate reasons, right? But clearly this is a much larger issue than those two firms. Now, let's look at some specific victimization data at the fourth through sixth grade level. What you see here is a series of questions that were asked of students and we asked them to identify what they had experienced in the last year. In other words, during the last school year, as someone used your password without your knowledge, right? We think that's a form of abuse. If they voluntarily share their password, right, and then use that password and access to the system for inappropriate reasons, we don't get at that, right? So this is sort of outright victimization, if you will. And then you can read the categories across for yourselves. The large point here is that victimization is related to social computing. The ways in which students interact online, right? The hang-up with adult sexual predators going after underage people, boys and girls for all kinds of salacious reasons is an important but very small problem as a part of a much, much larger cyber crime problem in society and we now know that it involves youth. It's youth exploiting youth in a variety of ways, not just adults who are preying upon children for sexual or other reasons. And here's our evidence of that. If you look at the last four bars to the right, these are sex-related victimization issues. Someone requested pictures of you without your clothes on, showing you pictures of themselves without their clothes on. They asked private things about your body and they told you private things about their body. If you experienced any of these things and you knew who did it, who was that person? That's this data. If you experienced that and you knew who the offender was, who was that person? The punchline is that it's kids' own kids. They're exploring their sexuality, they're maturing into this stage of their lives and they are experimenting with all kinds of social interactions online in largely unsupervised ways that expose them to this content. And hence it's no surprise that we are seeing the onset of pseudo-child porn and of course prosecution that's taking place around the country. The most sensational arguably is from last month in Westport, Connecticut in which a 12-year-old girl was charged for sharing indecent sexual content. I know I have this slide that identifies the perpetrator for middle school, so let us press on. At the middle school level, at this age, youth are experiencing every known form of cyber abuse, offending and victimization as a group that is absolutely true. It's also true that schools are the safest place in the world, at least in America, for kids to be accessing the net. By and large, their systems are better protected, increasingly against proxy servers, that kids can hook up to from school to their own computers and so forth. Cell phone utilization is controlled from policy and monitoring and supervision and policies against this kind of activity. And classroom activities are also closely supervised, right? And yet this group is very active in terms of their mobile computing. Social computing is mobile computing and mobile computing is social computing and they make no cognitive distinction about being online or offline. They just use their devices and engage each other all the time, right? This is what they do. So within the past school year, about one in four have lied about their age online. About one in ten pretended to be someone else. Seven to eight percent circumvented security measures, such as internet filtering or blocking software that parents often rely on, and five percent cheated on schoolwork. The victimization data for 7th through 9th grade is as follows. And by the way, let me just stress that we also consider physical activities along with cyber activities in aspects of our research. Because again, social computing is my mobile computing and vice versa, right? And so much of what happens occurs to and from schools in the shopping mall and importantly out of the home. And as students age and they mature, not only are they entitled of course to more privacy and more freedom, right? And because that's inevitable, anyone who has teenagers knows that, right? It's also less possible for the parents themselves, even if they were aware, trained, educated, and had the time to install, maintain tools and so forth, to carry out the sort of monitoring that might be necessary, given mobile IT devices, right? Unless we're going to turn them all into log analysts and that kind of thing. It's a very, very difficult challenge you have before you. Once again, notice that the sex-related victimization experiences are over on the far right, right? So someone requested nude pictures, someone showed you a nude picture, they requested sexual chat with you, or they actually engaged in sexual chat with you. Since they're only in middle school, right, we considered that a form of offending merely for purposes of illustration here. Now, if you experienced one of those types of activities in the previous school year and you knew who did it, who was the person? Was it a man? Was it a woman? A boy? A girl? A classmate? A friend? An online friend? An online stranger? A parent or guardian? Another family member? A teacher? Other known and unknown? Yes? Correct. No, any of these. If you were a victim of any type of these offenses and you knew who did it, who was that person, right? And they had to choose one or more of these categories, right? So this is what shows. And the big picture here is that it's kids on kids, right? I have, in the last three months, briefed the National Intelligence Council, the Department of Justice, the National Governors Association, and I have heard similar groups like this, and everywhere I go and everybody I talk to and everybody that is paying attention empirically to this issue understands that many of our resources have been sort of misallocated towards things that maybe we shouldn't be investing quite as much in. And that's something for this task force to consider as it contemplates technology R&D investments. Now, here is the offending data for middle school, right? Have you given out your password? Have you hacked a computer? Have you pretended to be someone else? Have you intentionally embarrassed someone else? Have you bullied or threatened them? Have you lied about your age? Lied about your gender? Have you requested nude pictures of others? Requested sexual chat? Then you have piracy issues, followed by a variety of academic dishonesty. Plagiarism, buying papers, selling papers, cheating on school workers, cheating on exams, and once again circumventing or blocking or filtering software, right? They're doing it. Now, I've recently completed a factor analysis of offending. If there's any statisticians in here, any research methodologists? What a factor analysis does is it turns the computer loose on the data and it tells the computer to separate groups of offenders in statistically significant ways, right? So what I'm saying to you is that at the middle school level we have identified four discrete types of online offenders, right? We have generalists who do a whole variety of offending. Pirates who illegally download music, movies, and software, as it turns out, and we have academic cheaters and deceiving bullies. Now, there are exceptions to the norm in which you'll have a kid who crosses over from one type of offending to another. That's implicit given that we have a generalist category of offending, right? My point to you is that by middle school, children are beginning to specialize with regard to their preferences and the use of technology for certain types of offending behaviors, right? So I'm periodically bemused when Congress comes down on higher education and says, you'll stop that pirating or else you'll cut your research fund. You're including ninth graders in the middle school? Seventh through ninth grade, yes, sir. That's what this is. And I react to that and I say, well, again, I'm not supposed to talk about policy, but what do you expect? They start pirating in fourth grade when they're 10 years old and now we understand that it's a discrete form of offending that they begin to specialize in by the time they're in middle school. And for the researchers in here, these four categories explain 55% of the total variance in the data, which starts to become really important. You can think of it this way. About half of these kids are engaged in some form of offending in these ways as a matter of routine. It's what they do when they're online and they decide to offend, right? Now let's go up to high school. Again, this data is largely consistent with other studies that have been out there. High schoolers spend a good amount of time online. And by the way, they do a lot of other things, not just offending. They're engaged in research in various forms of recreation. When they get up to this age level, they're shopping for colleges and universities. They're buying things and they're selling things, right? And all of that can be productive and totally appropriate. Our research is different because it goes beyond how they use technologies and for what purposes, and explores offending along with optimization in the context of technology use. And that's the real value of what you're seeing here. Also at the high school level, we know that these numbers of students are engaging each other online, including strangers. The concept of friending, befriending other people, right, is a verb as well as a noun in the vernacular of youth, right? If we go further and we look at deceit and cheating and so forth, we ask about multiple offending experiences. So have you not only done it, have you done it once, twice, thrice, four or more times, right, in the previous school year? And what you see here is you see patterns of offending that begin to emerge, right? So it's not just a once upon a time thing, right? If you commit types of offenses and you're successful at it, well, that's social learning theory, you know? I mean, you just have your built-in rewards, right? And if you're in it with groups and groups of you are doing it and engaged in it, then that's just reinforced, right? And that's what we know, of course, from sociology and child psychology and so forth. At the 10th through 12th grade level, we also did a factor analysis. Now we have seven discrete forms of offending, each as the IT people in the room know have implications for technology, right? There are also implications for information security and internet safety and, I would argue, cyberethics, right? Now we've got hackers and fraudsters and pornographers who not only download but share and seek it among their peers, right? So you can think of these people as not only sexually curious, right? They are on the offensive in their search for sexual opportunities and experiences, right? They factor out as a discrete type of offender because by and large, and we have not, this would be a limitation on this presentation, we have control to this point for those who are 18 or over, but we know that 10th graders through 12th graders include 16, 17, and 18-year-olds. So some of those kids, depending on the constraints of their state laws, may be guilty of creating, distributing, and receiving child pornography, right, in violation of state or federal crime laws. Notice that as the number of categories of offending goes up, the amount of the variance increases, expectably so, so now we've got 74% of all the sort of noise out there this data accounted for. You can think of it this way, the majority of kids are engaged in one type of this offending. They've begun to specialize and this is what they do, right? So a little more here, we're almost done. Someone pretended, so this is victimization stuff. Someone pretended to be you online once, twice, thrice, four more times. They stalked you, they intentionally embarrassed you, and so forth. So here's my thoughts on this kind of thing. Let me try to wrap this up. I think that as children in modernized society come into the world, they are increasingly afforded with opportunities to engage online. We know that the age of onset is declining. We know this from our own research over the past four years involving college students as well as this major study. A whole series of surveys actually. When we started surveying youth in 2004 at the college level, their recollection was they started using computers at age 10.4 years, right? And more than half of them were the primary user of their home computer. When they were 10, which would have been in 1994 right after the internet was commercialized. When we started surveying 40,000 K through 12th grade students in May of 2007, the age of onset was 7.8 years of age. And in four months, by the time we wrapped up the survey, or began the second phase of our surveying in the last nine or 14 school districts, the age of onset dropped to like 7.6 years of age when they started computing. So they're beginning earlier in life. The population of youth increasingly are internet users and as I've indicated at younger and younger ages. Somewhere along the line, they become victims. We know that cyberbullying now begins, we know it begins in the second grade as operationalized to mean someone was mean to you or you were mean to someone else online. That's what children at that age level can understand. As we get older, we ask more in-depth questions. And then they sort of squirrel around out there and they slip into abusive behavioral patterns. They may not even recognize that it's wrong criminally or civilly or otherwise ethically, right? And then we also know that there's an overlap, that on any given day a child can be a victim and an offender and go back and forth just depending on what they experience. And maybe they've had enough, right? Maybe they sort of get out of it and just hang out online, become solitary, right? Searching for solos and escape from whatever they're experiencing. And we know that that kind of thing happens anecdotally but not as a direct result of this research. So here's the deal, folks. I think that all youth and all generations of youth as they grow up and mature, they experience things, they rebel. I grew up in the 60s. I remember what that was like as opposed to today's youth. Today we have a digital youth culture, which is arguably a subculture of something called Internet culture, right? And it is their space in their mind, right? And so social computing happens for these youth and young adults in all kinds of ways that are productive and fun and it stimulates them and it brings out the creativity in them and it complements the ways in which they learn, right, and so forth, right? And they always want to be online, right? They need to chat and they need to interact. And yet we know that there's alarming aspects of this. The insubility, the promiscuity, the abuse, the sending me messages and that all of this is learned. And it's as if this culture is perpetuating these attitudes and behaviors because that's what's accepted and normed out, right? You asked kids, you heard it earlier before lunch, you talked to a lot of kids, and I know it from our own research of a qualitative nature not that I'm presenting on today, a lot of kids don't even know what cyberbullying is. Oh, that's something that adults make up, right, so they can crack down on us. So we're also able to determine that ages of onset for different types of abuse kick in and may peak at certain levels. We know, for example, that cyberbullying peaks in middle school, right? And we know that females are more engaged than males, although we don't know why, right? Not exactly. We can surmise that from our data. Dana can tell you more about that perhaps. We know all of these other things, right, about their culture that leads speak as a way to disguise their messages as well as abbreviate and do the rapid-fire messaging. You can't type out full messages when you have to do it this way. And the content of behaviors multiply virally. That's the term they use, right? So I don't know how you can get out in front of this thing except by thinking through it as the complex social problem that it is. Last slide, I think, Dana, here. Choices in parental oversight. So where are the parents in all of this? This often comes up, right? Well, our research is we were able to survey parents as well as youth and we're actually able to link the responses of each anonymously in such a way that preserves privacy and anonymity and so forth. Suffice to say that parents in general believe that they're doing a better job of supervising than the children believe that they are, right? The law. And that is especially true at higher age levels as they mature and grow up and expect more privacy and children of adults expect them to engage in more responsibility. You have my contact information, the business cards, the two brochures growing around or sort of propaganda sheets of brochures that we put together for a big briefing in New York City for the National Press. I would encourage all of you, if you want to see the full report which Dana's crew has reviewed and all the survey instruments that we've done. Everything that we've done is posted at this website, right? And you'll be probably reading more about it in various books and academic journals and stuff as we go forward. I have a quick question. Dr. Quay, were there any questions? You can call me Sam, that's fine. Okay. For the 15% of the 10 to 12 graders who invited an online stranger to meet, do you have a breakdown of how that was male and female? Not today, but we can do it that quick. Right? One of my limitations to answer your question more fully, I had two slides prepared. I'm really sorry, I don't know where they went off to. Cyberland, even though. One of my limitations was that we're in the very early phases of our analysis. And frankly, I think one of the reasons you're just now hearing about this research is because through the review, the literature review process, Dana and her team only really became aware of this research about three weeks ago or so, and it's because it's not published in academic journals. That's forthcoming. So you are really getting this stuff before it's out there in the mainstream academic world. What else? Any more questions? Sam. Yes, sir? Well, in general, the soundness of research findings ultimately is rooted in the methodology employed. I'm not familiar with the study. I didn't hear the presentation. 80% sounds awfully high to me. I'm not familiar with the study. I'm not familiar with the study. I'm not familiar with the study. I'm not familiar with the study. I'm not familiar with the study. On the other hand, we know that one of the flaws in all survey research, including our own, is the requirement to rely on memory. This is a retroactive response to questions. And sometimes what we can't remember very well will have a tendency to make up. And we also know that when it comes to these sensitive questions about sex and crime and so forth, that some children are prone to over- or under-reporting, so you really have to do some data quality kind of assurance things. I will tell you that in 2004, when we did our first college study, two out of three students shared their password on the RIT network system. And 50% of those students never changed it after they shared it. What you know is who they were sharing it with. You can't make the assumption that they were sharing it with another college student in RIT. So I told the university, I said, in effect, we've got an open system here. That'll make the MIT and Harvard, IT and security types really happy. Right? That's the reality. One more question. Yes, sir? Why are kids longing out for age and learning to access to it? Oh, that's a great question. I'm not sure, but I have a couple schools of thought on it that I'm kicking around in my mind. One is that they're doing it to access content that they should not. They lie about their age in order to get access, particularly probably to pornography sites, because of the controls that you put in, some of you rather, for age verification. I also have a hunch that some of the improper credit card usage that's showing up as credit card fraud and what we're qualifying as identity theft may actually be rooted in their desire to view pornographic material, whether it be legal or of the illegal variety. So that's one general school of thought. The other major school of thought, actually to me, is far more interesting. And that is that children may be lying about their age, their gender, their appearance, and putting up relatively true to false information on any number of social networking profile websites in order to protect themselves. After all, we've been hammering it home to them how vulnerable they are. They're not stupid. They don't want to be victimized. They just want to friend and do what's called for in their contemporary digital youth culture. And that may even serve a useful crime prevention strategy from a consumer awareness standpoint. But to me, that's also deeply troubling. What does it mean that we have millions of youth by extrapolation that are routinely engaged in dishonest communications with their friends and with relative strangers and complete strangers? What is the basis of our civilization? And of course, that's beyond the scope of this committee and anything that's currently underway. But it is something to ponder. And I would offer this. If you look at research investments by the National Science Foundation for trusted computing, research and development, and so forth, we're all about trying to ensure the identity of who it is that we're interacting with online because trust matters. You have to have trust. Well, we are perpetuating millions of kids today who are thriving largely unsupervised without mechanisms to reinforce the importance of trust. And it's deeply disturbing. And I don't know how to resolve it, but those are my thoughts on this. Thank you so much.