Data has this marvelous capacity to show patterns of human behavior, tell stories and even predict what we are going to do next. It’s the premise of “Dataclysm”—the stories data tells about what we value, how we think and act. I chose the book as one of my 2015 reads because of the big mountain of data that education institutions are collecting; I wanted to get a glimpse into how data predicts behavior, to learn about privacy boundaries, and was hoping to get a glimpse into how data might help us design, develop and deliver better learning experiences for students. A tall order. Not surprisingly I didn’t find answers; but I did learn about the power that data holds and discovered a good report by EDUCAUSE that does have some of the answers I was looking for.
The biggest takeaway from “Dataclysm” is the incredible potential data holds, which can translate to education sector. On its own data has no value, but with the right software data can inform, support, predict and help. Most organizations including education institutions collect mounds of data. Some is put to good use though according to the EDUCAUSE report the majority of data is used to satisfy credentialing or reporting requirements rather than to address strategic questions. And much of the data collected is not used at all (Bichsel, 2012). Education data is of abundance. Students generate a significant chunk. Every time a student logs-on to the LMS, school portal, or uses school software, printers, e-books, etc. data is collected. Every click, key stroke, time on web pages, links clicked are recorded.
I definitely think it’s good. … All of this data — everything in the book and generally anything you read online about people’s behavior on sites — is aggregated and anonymous. Nobody’s looking at your personal account. But when you put all this stuff together, you’re able to look at people in a way that people have never been able to look at people before. — Christian Rudder, Author of “Dataclysm: Who We Are” NPR Interview
Rudder, author of the book and quote above, is also co-founder of the dating site OKCupid. He gets most of the content for his book from data on his site though he also draws from Twitter and Facebook. Rudder describes how he takes data, without identifying details such as user names, and analyzes it to create narratives that describe human behaviors. The book is full of stories the data tells about race, gender and politics, which at times was disturbing. Not the writing, which is witty and entertaining, but the results of his analyses. Rudder calls his work more of a ‘sociological experiment’, examining human behavior, values, even biases by looking at (online) actions, words, choices, link clicks, and ratings.
‘Dataclysm’ was interesting—not instructive but insightful. Since finishing the book I’ve recognized how, what many label as disruptive services, are data-driven. Uber for instance, the new taxi service. Its business model rests entirely upon big data. Uber uses complex algorithms to aggregate data into actionable info that quite literally drives the business (Marr, 2014). Another—a new email program by Google, SmartReply, can write email responses for us by using machine learning to ‘work on a data set that they cannot read’ (Corrado, 2015). Whatever that means. But the gist is, its BIG data behind it.
Big Data and Education
EDUCUASE defines analytics as “a tool used in addressing strategic problems or questions”. Analytics are typically applied to institutional data and learning or academic data. Yet it appears that the potential of big data in education is yet to be tapped. The field is broad, complex and there are numerous barriers as described by Bichsel.
One of the major barriers to analytics in higher education is cost. Many institutions view analytics as an expensive endeavor rather than as an investment. Much of the concern around affordability centers on the perceived need for expensive tools or data collection methods. What is needed most, however, is investment in analytics professionals who can contribute to the entire process, from defining the key questions todeveloping data models to designing and delivering alerts, dashboards, recommendations and reports.
Though there are many institutions working extensively in learning analytics with the goal of helping students succeed and improving outcomes. One is University of Michigan who have helped create a standard that ensures third-party vendors (e.g. LMS providers) provide institutions with access to data generated by their students—not to withhold the data which can be critical for schools looking to use it to support and inform student success (Mathewson, 2015). Another is Purdue University who has done extensive work in academic analytics with its LMS program Course Signals (Research on Course Signals, n.d.).
Rudder states in his book that we are on the ‘brink of a revolution—a data revolution‘. I think he may be right. The education sector may take some time to figure it out, but for those that get it right, it will be revolutionary.
- What Universities Can Learn from Big Data – Higher Education Analytics, Bill Schmarzo
- Analytics in Higher Education, Jacqueline Bichsel
- Keeping Up With…Learning Analytics, Association of College and Research Libraries
- Students Speak Up About Learning Analytics, Laura Deveney
- Uber as Boss: The Employer as an Algorithm, Elizabeth Dwoskin, Wall Street Journal
- OKCupid’s Unblushing Analyst of Attraction, Natasha Singer, New York Times
Bichsel, Jacqueline. Analytics in Higher Education: Benefits, Barriers, Progress, and Recommendations (Research Report). Louisville, CO: EDUCAUSE Center for Applied Research, August 2012, Retrieved from http://net.educause.edu/ir/library/pdf/ers1207/ers1207.pdf
Corrado, G. (2015, November 3). Computer respond to this email [Web log post]. Retrieved from http://googleresearch.blogspot.com/2015/11/computer-respond-to-this-email.html
Marr, B. (2015, May 7). The amazing ways Uber is using big data [Web log post]. Retrieved from http://www.datasciencecentral.com/m/blogpost?id=6448529%3ABlogPost%3A274404
- Research on course signals and academic analytics. (n.d.). Retrieved from https://www.itap.purdue.edu/learning/research/