Education Tech Talk: How CourseSmart Analytics will change e-textbooks
Mobile devices and e-readers invaded new classrooms in 2012, and with voices as prominent as U.S. Secretary of Education Arne Duncan calling for an outright end to paper textbooks, publishers and schools alike want to know how to better use e-textbooks at all levels. This year, just in time for the the 2012 Educause annual conference, CourseSmart announced a pair of new initiatives to drill down into the behaviors and markets that will determine how the next generation of course readings are consumed.
Education Dive sat down with CourseSmart VP, Marketing Cindy Clarke at Educause to find out more about the company's new e-textbook pilot program, as well as CourseSmart Analytics, the program CourseSmart will use to measure student engagement with e-textbooks and help institutions to monitor how course materials are connecting with readers. Here's what she had to say:
[Editor's note: This is the third interview in Education Dive's "Education Tech Talk" series. Be sure to check out our chat with the University of Richmond's Matthew Trevett-Smith about his school's iPad program, and our discussion with Turnitin's Chris Harrick online grading and cheat controls.]
EDUCATION DIVE: What are the big questions you’re trying to answer with Analytics, and what do you see faculty members coming to you looking to know?
CINDY CLARKE: I think that there’s three key answers that Analytics addresses—or three key questions— and those align with the goals that you hear institutions are trying to achieve across the board. One is student retention, and we believe that providing insights into how students are using the book and how engaged they are with the textbook is an indicator of whether or not they’re going to stay in the class. It also seeks to improve student outcomes. Seeing how a student's engaging with a book, how much of the book they’re consuming, how they’re using the tools in the book—is an indication of how successful they’ll be.
Some metrics—if they are low compared to other students in the class or compared to CourseSmart overall for the use of that particular book—that might be an indication that the student needs some help. And so you can see where the instructor can evaluate metrics across their class, see how students are doing compared to each other. It’s not to be used in isolation, but also use in comparison with test scores and other information the instructor might have about their students to then decide who might need some extra help, who doesn’t need extra help, how they can help their students be more successful and how to make sure their student actually completes the class.
Are these analytics being used in traditional classroom environments as well as in online learning environments right now?
CLARKE: We just announced the beta, so we have three institutions who have agreed to take part in the beta—so it hasn’t formally been launched yet. The three are Villanova University, Rasmussen [College] and also Texas A&M [University]-San Antonio, [who] have agreed to take part, and we expect about seven others. We’re shooting for ten beta customers. We think that we'll be oversubscribed, but we’ll shoot for ten beta customers to take part, and we want to make sure we have representation across different types of institutions, because we think it has applications in not-for-profit, for-profit, online and traditional in-classroom teaching.
There are key metrics that are viewed in an easy-to-view dashboard that the instructor can access and apply filters to. The instructor can filter by date range, and it shows the average session length by students, the average pages viewed, the average number of highlights, notes and bookmarks used for a particular book or across a group of classes. "Books subscribed" would indicate that if the instructor has 30 students in their class and he sees 30 books are subscribed here, he knows everybody is subscribed to the book. If he’s seeing only 10 are subscribed and his class is 30, he knows he needs to follow up with two-thirds of the class.
Then, the engagement score is the pillar of what is very, very unique about all of this.
These scores are interesting. How did you come about devising them?
CLARKE: We tapped into expertise in technology and business—and also academia—to arrive at a proprietary algorithm that takes different metrics about how the student is using the book and consuming the book and calculates a score.
Is this at all influencing what publishers are having to do when they’re creating the books that are being used?
CLARKE: That is an excellent question. I’m really glad you asked that, because in parallel with this project we’re developing a dashboard as well for publishers to use that gives them aggregate data. They’re most interested in aggregate data, not specific student data, which is not very meaningful to them. But it will show them within their textbooks how that textbook is being consumed, and so if a publisher finds that nobody is using or there is very low engagement with chapter four in a biology book, they can assess what’s wrong with that chapter and figure out whether it’s needed, how they can improve it and make decisions about the specific course material. We think that this creates kind of a perfect storm of information and meaningful insight that can be used in the classroom, as well as by the publishers to make their materials much, much better.
This shows usage compared to the aggregate CourseSmart of the tools, we believe that usage of the tools is a high indicator of student engagement. So if I’m in my textbook and I’m highlighting and I’m taking notes, I’m probably pretty engaged with the material and consuming it in a way that will make me a successful student. And then the instructor also has the view to see which students in his class are most engaged with the materials and which are least engaged. He can look at all students by drilling down [with] the filters that are at the top. But a use case might be that he has a student who is clearly engaged in the book; she’s been using highlights and notes very frequently; she’s consuming the materials in a way that would indicate that she’s going to be successful in class.
So he sees she’s highly engaged in the book but she did poorly on the exam, and so he could suggest to her or help her with her test preparation, or she might have test anxiety. She’s obviously consuming the materials but something is not connecting to the test and so he would how to intervene in that way.
Or, if you have a student who you see, say it’s one month into the course and you look at your student engagement list and you see that there are a couple of students who haven’t even cracked open the book and we know that if somebody starts late into learning the materials in a class they’re very likely to drop out and be completely unsuccessful. And so he can use that to intervene and help them first make sure that they have the course materials, make sure that they understand what they need to prepare for their test and just reinforce that they need to be reading their materials. So those are two examples on how this would be used at the instructor level.
Is there any application above that at the administrative level?
CLARKE: In the administrative level, the numbers are much bigger. The administration level can drill down to instructors, to courses, so if they want to look at only biology compared to some other course, they can do so. They can look at specific books, books by specific publishers if they want to look at one publisher against another. If they’re using different publishers’ textbooks for a similar course, that would be insightful for them, and they get the same metrics as well as the engagement score overall.
So, they do not get the by-student view, we don’t necessarily think that’s meaningful or actionable at the administrator level. They’re going to be looking at a much higher level—how different books are being used across their institution. So once again, you get the same engagement score compared to CourseSmart overall for that particular title.
And what is CourseSmart learning from this information?
CLARKE: So as I mentioned, this is a beta. CourseSmart has a company spirit of making—when we do new products or new services—it’s getting to a point where we feel like we gathered enough information and worked somewhat in isolation, and you know [when you] get it out into the market and get it into the hands of people who are really going to use it. So we can use real use cases to inform how we iterate on the product for a full-scale launch. Likewise, with the subscription pack pilot that we also announced about a week ago, we’re partnering with Internet2 to provide students at the institutions that take part a great value [and a] very affordable option.
And there are what, two levels they can subscribe at?
CLARKE: There are two different levels. The institution is selecting the level, so either 100 students or 200 students will take part. Basically, the student will have a bookshelf of twelve slots, and they can choose e-textbooks from across the CourseSmart catalog and add those books to their bookshelf for one price. So the twelve slots—we don’t believe that an individual student uses twelve books in a term—but what that does is allow for an add/drop period. So you might be in a cycle where you’re not sure what classes you’re going to end up with, you need to add books so you can keep up, but you’re going to end up dropping them, so you have twelve slots where you can add books.
And you can exchange one for another?
CLARKE: You can, yes. From Student Monitor we know that students are shopping everywhere for their course materials now. They’re combining digital with print; they’re buying used; they’re renting.
They're using different devices to view the material.
CLARKE: Exactly. The way that the subscription pack will be very valuable to them is it’s a great price for them to access all their course materials. It’s very efficient; they can get their course materials immediately; they don’t have to wait for them to be shipped; they have one format in one bookshelf; and, to your point about access, they can access their books from any Web-enabled device.
We find through our own research that students use on average three devices a day. They’re using a laptop to take notes in class. They might use an iPad in a different instance. They also use iPhone when they’re on-the-fly. Students expect their life to come to them in many ways. Education falls within that as well. With CourseSmart they can access their reading material anytime they want and on any Web-enabled device.
What do you hope to find out looking ahead?
CLARKE: We did qualitative research leading up to this in-market study that we’re doing with Internet2. That reinforced that both faculty and students were interested in that as a concept and they thought it was very affordable and efficient. And so I expect that we’ll—through the market research study—find ways to fine-tune it and make it relevant to students across the board and institutions across the board.
I expect that our name will not end up being "Subscription Pack." That means different things to different people. It’s our working name for the moment. We’ll find out if 12 books is the right number, or if it’s less or more. We’ll find out about the price point and whether or not that works across all disciplines. When you look at textbooks for science and engineering, they’re very, very expensive. Textbooks in other disciplines are not so expensive, and so we’ll get insights on whether the pricing level we have set, based on internal evaluation of average price point across textbooks, is right or will work across disciplines.
Application-wise, I note you said "Web-enabled device," so there is a Web app. Is that the only way to gain access from specific devices? What does the landscape of that look like?
CLARKE: There are dedicated apps as well. CourseSmart was the first e-textbook provider to launch an iPhone app. Also the iPad app. I mean others have obviously gotten on board after us. We also have a dedicated Android app.
Do you see these two programs informing each other as you get maybe a year or two years down the road?
CLARKE: I wanted to mention that those that take part in the subscription pilot—or market research pilot—will also have access to the analytics package. That’s a different model for delivering textbooks and will also be able to assess these metrics within a subscription pack product and determine whether or not we’re seeing different engagement results, or similar engagement results, to institutions that aren’t necessarily taking part in subscription pack product.
Would you like to see more education news like this in your inbox on a daily basis? Subscribe to our Education Dive email newsletter! You may also want to read Education Dive's list of 10 ed tech edupreneurs who made waves in education in 2012.