Most Active Stories
- Controversy Over Blackbeard's Queen Anne's Revenge Continues
- Deep Water Shipwreck Discovered Off North Carolina Coast
- The Front Bottoms, 'Laugh Till I Cry'
- Clinton Won't Go As Far As Rivals On Minimum Wage Or Rule Out Oil Pipelines
- Updated Atlantic Hurricane Season Outlook Predicts Below Normal Season
Tue January 29, 2013
Finding Learning Tools In Digital Footprints
Originally published on Tue January 29, 2013 12:10 pm
Throughout Tell Me More's series, "Social Me," Rey Junco shares the research he's done as a faculty associate at the Berkman Center for Internet and Society into how how young people interact online.
While some parents might think that their kids are wasting time on social media sites instead of doing their homework, they could actually be helping to shape their own education. Junco describes for NPR's Michel Martin a potentially revolutionary educational tool. The technical term is 'learning analytics.' Junco says that is "the use of student-produced data to predict outcomes and tailor education."
Students leave a trail of data behind them as they navigate the Internet. Junco believes that there is a big opportunity here. His own research has already shown that "we can use things like how much time students spend on Facebook and what they do on Facebook to predict academic outcome."
Junco says this is already at play in our daily lives. "Think about how a website like Amazon.com knows the kinds of things you like, based on your shopping habits." He says we should imagine the same idea, applied to education. "We've got data well before a student will flunk a first exam or a quiz," he says. "So we can make some predictions about the things that they're doing, and how we might intervene before we get to that point."
That said, Junco admits that educators are not using learning analytics much. He explains that it is generally applied to course management systems — basically a fancy term for an online discussion board — to monitor the level and frequency of a student's interaction. But, this data is "very, very basic," Junco says, "so the predictive models are not as accurate as they could be."
Junco explains that this analysis requires resources. But he does point to consumer applications and services, like digital pedometers, that can monitor what you do online. "We know from the research that the act of monitoring and tracking your own behavior changes your behavior," he says.
MICHEL MARTIN, HOST:
Now let's take a look at how young people are interacting online. It's the latest in our series of conversations around social media and young people. It's a short series we're calling Social Me. Yesterday we talked about the ways young people are exploring different online identities. We talked about the fact that there could be a positive side to that, as well as the negative side that we hear so much about.
Today we want to look at how websites kids visit and the apps that they download could be used to enhance their educations. To tell us more about that, we're joined once again by Rey Junco. He is a faculty associate at Harvard's Berkman Center for Internet and Society.
Rey Junco, welcome back. Thanks for joining us once again.
REY JUNCO: Thanks for having me back, Michel.
MARTIN: Now, there's a technical term for what we're talking about here. Learning analytics. In layman's terms, what are learning analytics?
JUNCO: Well, learning analytics is the use of student-produced data to predict outcomes and tailor education, so it's often thought of as a practical application of big data in education. So in other words, it's collecting large amounts of data to identify patterns that will help tailor education more precisely for each child. So if you think about how a website like Amazon.com knows the kinds of things you like based on your shopping habits, imagine the same thing for education.
So I think of learning analytics as the ultimate formative assessment. We're always talking in education about how formative assessments are very important. It's important to assess frequently and to make adjustments. We've got data well before a student will flunk a first exam or a quiz and so we can make some predictions about the things that they're doing and how we might intervene before we get to that point.
And so, for instance, in college, you might be taking a class where there are only three exams and then a final, and so it's very important to identify what some of these risk factors are and we can do that with learning analytics.
MARTIN: Are educators doing this now?
JUNCO: Not much. There are some applications that are collecting data. Generally, what you see is learning analytics applied to course management systems and course management systems is basically just a fancy way of saying an online discussion board and they are looking at how many times a student will respond, how - to a discussion - how quickly they respond to a discussion, how often they log on. So just very, very basic levels of data that they're looking at, and so the predictive models are not as accurate as they could be.
MARTIN: What is your research in this area about?
JUNCO: I think we could be doing so much more with learning analytics, so for instance, students are providing a lot of what I call trace data, so data on their use of technologies, data that they leave behind just by the nature of what they do every day, so their Facebook use and their Twitter and their Instagram use and their mobile phone use.
So some of my research has already shown that we can use things like how much time students spend on Facebook and what they do on Facebook to predict academic outcomes. So I'm actually engaged in a project right now where I'm collecting trace data on college students to create more precise, predictive models, and then with more data comes the ability to be more precise and to tailor better interventions to help support students.
MARTIN: Is there anything that just - forgive me for using this term - regular people could do to use this kind of data now? For example, I mean, you know, you often see these days people who, you know, are privileged enough to be able to have like a tablet, computer or, you know, even a smartphone. I mean, you see people waiting in line. You see people at the grocery store. A kid gets antsy. What do people do? They pull out their phone or they pull out a tablet, they hand it to the child, who can play and entertain himself or herself for quite some time.
There are all kinds of apps out there that kids are already attracted to. Some of them they're finding on their own or they're finding out about from their friends. Is there anything, you think, that people could do on their own to use this time more productively?
JUNCO: Sure. Well, there are two things.
MARTIN: Without being corny about it. I mean obviously, you know, the kiss of death is the educational app, educational software. You know perfectly well your kid's going to run screaming from that.
JUNCO: Well, certainly that would be great. The learning analytics kind of process takes some resources, so typical apps don't have any kind of learning analytics integration. However, there are consumer level applications and services that let you monitor what you do online, and we know from other areas - so, for instance, there is this, you know, movement about collecting a lot of information about yourself, the quantified self movement. Right? And so how many steps you walk every day using something like the Fitbit or the Nike FuelBand to check in on how well you're doing towards your fitness goals.
And we know from the research that the act of monitoring and tracking your own behavior changes your behavior, so just the act of monitoring changes your behavior. So there are some consumer applications out there that you can sign up for. You can install software on your computer. You can install software on your cell phone and it gives you a report at the end of the day and it says this is how you used your time. And some of these will say here is how much time you spent in unproductive activities and here is how much time you spent in productive activities.
MARTIN: Is there any sign that kids like this too? I mean I think that people who just heard what you said, you know, with the pedometer, some companies even giving them to their employees now to urge them to, you know, walk more during the day, or those FitBands or those wristbands that people have bought. A lot of people, I'm sure, got those for holiday presents.
Is there any sign that kids like those too, that kids respond to those too?
JUNCO: To things like the Nike FuelBand and Fitbit...
MARTIN: Yeah, yeah.
JUNCO: ...and things like that?
JUNCO: You know, I haven't seen any research on that.
MARTIN: Well, here's an area for you.
JUNCO: Yes. And actually I'm thinking about it from a more learning analytics perspective, not from the quantified self and the health data perspective, but just finding out how amenable they are to learning analytics.
MARTIN: See, I've given you a new idea.
JUNCO: Yes, you have. Thanks, Michel.
MARTIN: Rey Junco is a faculty associate at the Harvard Berkman Center for Internet and Society. He joined us from WPSU at Penn State University.
Rey Junco, thanks for joining us.
JUNCO: Sure. Thank you for having me.
MARTIN: Tomorrow, we'll be talking about an issue that many parents are concerned about - and teachers. How children are protected or are not protected by digital privacy laws. So please join us for that conversation. Transcript provided by NPR, Copyright NPR.