Dartmouth Predicts GPA Based on Phone Tracking App - InformationWeek
IoT
IoT
IT Life
News
5/29/2015
09:06 AM
Connect Directly
Twitter
RSS
E-Mail
100%
0%
RELATED EVENTS
Cloud Security: Don't Go Blind While Playing in the Cloud
Dec 06, 2017
Attend this webinar to understand the security transformation from an on-prem fortress mentality t ...Read More>>

Dartmouth Predicts GPA Based on Phone Tracking App

Can a phone app predict your performance based on your behavior? This one can, at least if you are a college student.

Big Data: 6 Real-Life Business Cases
Big Data: 6 Real-Life Business Cases
(Click image for larger view and slideshow.)

We all know our grade point averages (GPA) would have improved if we spent less time on the weekend re-enacting our favorite scenes from Animal House and more time in the library. Yet, few of us led the kind of lives where we knew exactly where we went and how much time we spent there. But now our smartphones do. And that's why Dartmouth invented an app that can track student behavior and predict their GPAs to within a tenth of a point based solely on the info from their smartphones.

This is the quantified life. And if it works on campuses, it could soon work in the office, in our home lives, in the gym, and anywhere else. Think of this as the fitness tracker for life.

The Dartmouth app uses GPS and WiFi data for all parts of the small college town of Hanover, NH. The app knows when you are in the library, the coffee shop, the sorority house, and pretty much anywhere else. It figures if you're in the library you are probably studying, and if you're in a fraternity house on a Saturday night, you probably aren't.

(Image: Kane5187 via Wikipedia)

(Image: Kane5187 via Wikipedia)

Using your location data, the app predicts what you are actually doing and tracks your studying, sleeping, socializing, physical activity, class attendance, and even your stress levels, to get a picture of how you're doing. It can then predict your classroom performance without knowing anything previously about your talents, your IQ, your grades, your SAT scores, or anything else.

In other words, it is our behavior, not necessarily our talents, that best predict our academic success. Will you pass on the party to make sure you get that paper done? Do you show up for class? The Dartmouth researchers tried it with 30 students and could predict GPAs within .17 points. You have to figure they'll dial it in more with a larger group.

Here's a video that shows how it works, and a little bit about the types of behaviors that lead to academic success:

This is usually where we launch into the spiel about privacy, and how we don't like our smartphones tracking our every move. I'm not going there this time. First off, our phones track us anyway, so we might as well use it to our advantage. Second, this is one of those times it is for our own good.

It isn't hard to say that if we get more sleep, more exercise, go to class, skip the party, and study more that we'll get good grades. The trick comes when real life meets the ideal. And this app can help you course correct. Hey, you had a really good weekend last week, but maybe it is time to hit the books before it is too late? Or, maybe, all work and no play makes Jack a dull boy.

The limitation to this going global, of course, is that for now the app is tuned to Hanover. If you took the same app to Cambridge or Palo Alto you'd have to reprogram it for the locations there. And some colleges are easier than others because of how the social life works. It is easy to map fraternity row, but what if the hot spots are nightclubs in the next town over?

What if you took the same basic concept and applied it to an office? It might actually work better there. Is the best employee the one who sits at their desk all day? Or, does getting up and circulating in the break room, or at the desks of fellow employees, help connect you to the info you need? Did you realize you take two-hour lunches?

[ This CEO thinks performance evaluations should work more like a fitness tracker. Read BetterWorks CEO: Treat Feedback Like a Fitbit. ]

Here is where the privacy spiel comes in. This is a great app for a student to choose to use for his or her own success. It isn't such a great thing if your boss tells you to download it and use it in your office. It has to be voluntary, which is a lot harder to do when mapping out the environment of an office than a college town. Once the employer knows the locations have been mapped for people to use this app, they're going to want you to use it.

Still, the quantified life is coming. We're counting steps now. Soon we'll be counting conversations, sleep, social time, and everything else, as the Dartmouth app is doing. Is this going to help us become more organized and successful, or are we going to stress ourselves out

[Did you miss any of the InformationWeek Conference in Las Vegas last month? Don't worry: We have you covered. Check out what our speakers had to say and see tweets from the show. Let's keep the conversation going.]

David has been writing on business and technology for over 10 years and was most recently Managing Editor at Enterpriseefficiency.com. Before that he was an Assistant Editor at MIT Sloan Management Review, where he covered a wide range of business topics including IT, ... View Full Bio

We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.
Comment  | 
Print  | 
More Insights
Comments
Threaded  |  Newest First  |  Oldest First
zerox203
100%
0%
zerox203,
User Rank: Ninja
5/29/2015 | 11:02:10 AM
Re: Dartmouth Predicts GPA Based on Phone Tracking App
This is definitely a strong case for how a college campus can serve as microcosm for adult life. Sleeping, studying, exercising - the study didn't mention eating habits, but students do that on campus, too, and many of them even work there. I love being skeptical, but that .17 point accuracy is hard to argue with. Numbers don't lie. I suppose it's worth noting that Dartmouth is an Ivy League school so the results may have been skewed towards the positive end (the full report actually notes this)... or we may be looking at a more diligent than average group of students. I wonder what would/will happen if/when the study is repeated at other schools or with larger numbers of students. There's certainly proof here that the predictive model itself works.

There are a few things that left me wanting to know a little more. The video and the report make repeated mention of 'conscientiousness' and how it relates to improved grades. What the heck does that have to do with smartphone-based location tracking? Actually, of the four other conclusions they make, only two seem (potentially) tied to the location awareness -  shorter conversations (which I'm not sure how the phone would really measure with just location), and increased time studying. The others, "positive affect", and stress levels throughout the semester, seem to rely on 'periodic self-reports' instead, which they don't really describe how they carry out. It seems more like a hybrid study, which is fine, but this is not exactly a case of the Eye of Sauron phone. Certainly, as phones and wearables can pull more kinds of data together, though, we might get there sooner than we think.
vnewman2
100%
0%
vnewman2,
User Rank: Ninja
5/29/2015 | 12:22:35 PM
Re: Dartmouth Predicts GPA Based on Phone Tracking App
@zerox - as an aside, conscientiousness is the only personality construct reliably shown to predict job performance.  You make a great point, this trait is based on self-reports typically in assessments - I'm not seeing the link to location-tracking either.  At my school, the library was kind of a hang-out, just because you went there  didn't mean you were studying.

Plus, there will always be that person who does not go to class, goes out all the time, and still gets an A on the test, just to throw a wrench into things.
SunitaT0
50%
50%
SunitaT0,
User Rank: Ninja
5/29/2015 | 12:42:07 PM
Re: Dartmouth Predicts GPA Based on Phone Tracking App
@vnewman2: Not only that the app predicts your GPA based on where you went. But you can go anywhere on the campus and be studying, for eg you can sit at a coffee shop and study, go to a commonplace and study, do whatever you like. Moreover I think this app can be made social like others would be able to track what their friends are doing and what their GPA would be if they are doing what the app thinks they are doing. Increases competition in many different layers.
David Wagner
50%
50%
David Wagner,
User Rank: Strategist
5/29/2015 | 1:12:38 PM
Re: Dartmouth Predicts GPA Based on Phone Tracking App
@vnewman2- It is true that going to the library doesn't mean studying. I think I never studied once in the library in my whole college career. I liked to check stuff out and go to my room. But the algorithm is tailored to the culture. It would have to be tuned to other cultures which is both a drawback and a good thing. It makes it harder to release today to the whole world, but it means it is more likely to be based on reality.
vnewman2
50%
50%
vnewman2,
User Rank: Ninja
5/29/2015 | 6:30:19 PM
Re: Dartmouth Predicts GPA Based on Phone Tracking App
@David - I have the perfect solution.  The app should pari up with the yet-to-be-released-new-and-improved google glass and actually process what your eyes are doing.  The eyes don't lie!  Now we're talking!
David Wagner
50%
50%
David Wagner,
User Rank: Strategist
5/31/2015 | 11:55:01 AM
Re: Dartmouth Predicts GPA Based on Phone Tracking App
@vnewman2- I think you'd break the Google Glasses app because it would get tired of constantly reporting that the kids were just "checking out the hot classmates." :)
vnewman2
50%
50%
vnewman2,
User Rank: Ninja
5/31/2015 | 5:15:11 PM
Re: Dartmouth Predicts GPA Based on Phone Tracking App
@david: you read my mind! You must have your google ESP headphones on today!
SunitaT0
50%
50%
SunitaT0,
User Rank: Ninja
5/29/2015 | 12:25:04 PM
Re: Dartmouth Predicts GPA Based on Phone Tracking App
I think privacy was always the main problem in user information based apps, because right now anything with so much as a net connection, isn't safe anymore. For students this would be even more risky because it basically computes the data based on "where" the student went and "what" the student does. These "where" and "what" may be hacked. 
David Wagner
50%
50%
David Wagner,
User Rank: Strategist
5/29/2015 | 1:09:29 PM
Re: Dartmouth Predicts GPA Based on Phone Tracking App
@Zerox203- I think things like conscientiousness are boiled into multiple aspects of the app. For instance, showing up at class or going to sleep on time. Things like that. The video, I believe, is to try more to help students succeed than it is to tell people exactly how the app works. They have really baked a lot of what makes a student succeed deeply into the app, but then the app can say to them, "hey, maybe you're skipping too many classes" or whatever and put it into plain english.
SunitaT0
50%
50%
SunitaT0,
User Rank: Ninja
5/29/2015 | 12:21:15 PM
Really confused here
Students who follow a set of daily activities except studying can check this app and see their expected GPAs and therefore reschedule their life activities in order to increase their GPAs. For example, a student who parties 2 hours daily finds his GPA to be 6.5 but then he reduces partying to one hour and sees his GPA improving, then that could motivate anyone to work hard. Also this app could host IQ, memory and logic based games to check whether it calculated the right GPA.
asksqn
100%
0%
asksqn,
User Rank: Ninja
5/29/2015 | 2:53:54 PM
So much hot air so little accuracy from Dartmouth
So an app that assumes simply because you are not in a library that you aren't studying.  This app would not yield anything accurate if I were to use it.  I'm guessing I'm not the only student on the planet who hasn't used the library exclusively to study/do homework.  Also, as any overworked, over tired student can attest, just because you make a physical appearance in class does not necessarily indicate you're paying attention and/or taking notes.  If this app is used as most tech is -a panacea- to predict certain outcomes, e.g. work performance, then its creator(s) and the dopey businesses that use it are in for a tremendous surprise.  This app MAY work accurately on those under age 18 but that's about the sum total of the demographic that will lead to anything useful.
progman2000
50%
50%
progman2000,
User Rank: Ninja
5/30/2015 | 7:09:37 AM
Re: So much hot air so little accuracy from Dartmouth
@asksqn - yep, I'm with you - sounds like a load of crap. And the video says it relies on a bunch of student reported data as well as what the phone collects. So it sounds like the phone makes assumptions on study/party habits based on where you've been and asks you to input a bunch of questions that if factors in. Plus they don't say what the sample range of their respondants were. If they used a bunch of enthusiastic students it's not hard to see that the entire sample might not have deviated much more that .17 of a point.

Yawn - have them give us the next Angry Birds and then we'll talk.
impactnow
50%
50%
impactnow,
User Rank: Ninja
5/30/2015 | 11:33:57 PM
Re: So much hot air so little accuracy from Dartmouth

I think the app would have to be tested in a broader sample group at multiple institutions to assure its accuracy. Once student entered data is used in conjunction with gps data the results get muddy. In the workplace I think it would be a night mare people have different working habits based on their jobs and their working style. I sometimes got more done talking to someone in the hallway than I ever could have done on the phone or email. There is a factor of efficiency that can't be measure with linear technology.

David Wagner
50%
50%
David Wagner,
User Rank: Strategist
5/31/2015 | 11:59:16 AM
Re: So much hot air so little accuracy from Dartmouth
@impactnow- Well, this isn't linear. In fact, it measures a lot of factors that have nothing to do with sitting in one place (sleep, fitness) and yes, I agree certain jobs need to stay more connected. So an app would have to be altered for different jobs. 

No one is saying this is perfect now. But it represents an interesting step forward in course correcting and intervening with people when the advice counts as opposed to when it is too late.
Kelly22
50%
50%
Kelly22,
User Rank: Strategist
6/1/2015 | 1:13:50 PM
Re: So much hot air so little accuracy from Dartmouth
While not fully developed, the idea behind the app has potential. I think to apply it to businesses, we'd need to consider each person and their working preferences. Some employees might need to interact with colleagues more to be productive while others can stay at their desks. I like working in coffee shops, for example, but doubt an app would say my 6 hours in Starbucks were well-spent.
David Wagner
50%
50%
David Wagner,
User Rank: Strategist
6/1/2015 | 5:35:37 PM
Re: So much hot air so little accuracy from Dartmouth
@Kelly22- Well, that's the beauty of the quantified life. Eventually the app would realize that was time well-spent and adapt to your needs.
tzubair
50%
50%
tzubair,
User Rank: Ninja
5/31/2015 | 8:32:31 AM
Re: So much hot air so little accuracy from Dartmouth
"If they used a bunch of enthusiastic students it's not hard to see that the entire sample might not have deviated much more that .17 of a point."

@progman2000: I agree. The data gathered can be very skewed and misleading. The sample has to take into account a lot of factors such as time, location, nature of students, circumstances etc. Only then you can base an argument. I don't think there'd be any real co-relation though.
David Wagner
50%
50%
David Wagner,
User Rank: Strategist
5/31/2015 | 11:53:10 AM
Re: So much hot air so little accuracy from Dartmouth
@asksqn- It isn't that simple. It can work out what you are doing in a lot of places. Besides, it isn't just study time in the library it is using for measures of success. It measures sleep time, social time, physical activity, stress and a bunch of other factors for success. For instance, one of the interesting aspects of the research behind th decision is a stress curve. Successful students see a rise in stress early and a decline as the semester goes on (I assume that is because they find their footing but it doesn't say way). If the app doesn't see a stress curve liek that in the student, it knows it is time to intervene before it is too late.

In my experience, the number one reason people fail is that they don't correct their behavior in time. In a college student's case, I can imagine that means not going to class and then as the semester ends trying to "pull it all together." Often, we're not aware of the little things that add up to make us fail. We are going along just fine in our minds and we don't realize that we're just spending a little less time doing what we should or taking care of ourselves and by the time we realize it is too late. 


tzubair
50%
50%
tzubair,
User Rank: Ninja
5/31/2015 | 10:23:48 AM
Application on the enterprise
So imagine if some business leaders start believing that this process is really credible and they implement it in the organization, what havoc it'd create. You'd start to be monitored on the number of hours you spend on the computer and what time at each file. How many restroom breaks you take and how long do you have your lunch for. That'd really make the idea of look-busy-do-nothing be useful in this case.
2017 State of IT Report
2017 State of IT Report
In today's technology-driven world, "innovation" has become a basic expectation. IT leaders are tasked with making technical magic, improving customer experience, and boosting the bottom line -- yet often without any increase to the IT budget. How are organizations striking the balance between new initiatives and cost control? Download our report to learn about the biggest challenges and how savvy IT executives are overcoming them.
Register for InformationWeek Newsletters
White Papers
Current Issue
Digital Transformation Myths & Truths
Transformation is on every IT organization's to-do list, but effectively transforming IT means a major shift in technology as well as business models and culture. In this IT Trend Report, we examine some of the misconceptions of digital transformation and look at steps you can take to succeed technically and culturally.
Video
Slideshows
Twitter Feed
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.
Flash Poll