The Day the Big Data Hype Died

March 28, 2014 may well go down as the turning point where Big Data lost its placement as a silver bullet and came down to earth in a more productive manner. Triggered by a March 14 article in Science Magazine that identified “big data hubris” as one of the sources of the well-known failures of Google Flu Trends,[1] there were five significant articles in one day on the disillusionment with Big Data.

1. Google Flu Trends: The Limits of Big Data (New York Times)




2. Big data: are we making a big mistake? (Financial Times Magazine)




3. Recent Big-Data Struggles Are ‘Birthing Pains,’ Researchers Say (The Chronicle of Higher Education)




4. Analytics 3.0: Beyond Big Data (CMS Wire)




5. Doing Big Data and Analytics Right (Hapgood Blog)

The New York Times (NYT), Financial Times Magazine (FT Mag) and The Chronicle of Higher Education (Chronicle) articles mostly describe the Science Magazine article and the motivations of the research team.

They included this statement (from the NYT):

• The authors of the recent articles were partly motivated by the desire to puncture that notion. “Google Flu Trends became this paradigm that you just look at all this data, make correlations, and we don’t need anything else,” [Alessandro Vespignani, a professor at Northeastern University,] said.[2]

… and this one (FT Mag):

• Cheerleaders for big data have made four exciting claims, each one reflected in the success of Google Flu Trends: that data analysis produces uncannily accurate results; that every single data point can be captured, making old statistical sampling techniques obsolete; that it is passé to fret about what causes what, because statistical correlation tells us what we need to know; and that scientific or statistical models aren’t needed because, to quote “The End of Theory”, a provocative essay published in Wired in 2008, “with enough data, the numbers speak for themselves”.[3]

… and this one (Chronicle):

• The reaction, from some, boiled down to this: Aha! Big Data has been overhyped. It’s bunk.

• Not so, says [David Lazer, a professor at Northeastern University], who remains “hugely” bullish on Big Data. “I would be quite distressed if this resulted in less resources being invested in Big Data,” he says in an interview. Mr. Lazer calls the episode “a good moment for Big Data, because it reflects the fact that there’s some degree of maturing. Saying ‘Big Data’ isn’t enough. You gotta be about doing Big Data right.”[4]

Does this mean Big Data is over and that education will move past this over-hyped concept? Perhaps Mike Caulfield from the Hapgood Blog stated it best, including adding the education perspective:

• I don’t know if I have to sketch out the parallels in education, but just in case: we have two really unhelpful parties in learning analytics. We have the “it’s all bunk” crowd, and we have the evangelists. And I don’t know which is worse.[5]

• Here’s the thing — saying “Big Data is bunk” is pretty close in ridiculousness to saying “Oceanography is bunk”. Seventy percent of the planet is ocean. Likewise, the “data exhaust” we emit on a daily basis is growing exponentially. There is no future where the study of this data is not going to play a large role in the research we do and the solutions we create. None. Nada.

So what is the likely path for Big Data for education, and how can we best go about “doing Big Data right” while not falling into the bunk or evangelist camps? The key is found in one element of the CMS Wire article  —  looking to the democratization of Big Data.[6]

Too many initiatives get this process backward  — trying to find broad solutions based on big servers and black-box algorithms before we have enough meaningful local usage and learning opportunities.

Higher ed is still evolving the definitions of success; what learning outcomes are meaningful, and which ones can be measured with hard data (hint: not all of them). It’s hard to use data to analyze what’s happening and what impacts success when you have yet to agree on the definition of success. While it may be difficult to define success writ large across higher education, it is much easier to do so one class or project at a time.

There will be parallel efforts, and for limited cases with well-defined measurements, big-server algorithmic Big Data will make sense. There needs to be a shift in emphasis, however, toward the local.

If we focus on individual projects and classes, and look to consumer tools and local exploration, there will be more opportunities to learn and figure out how to benefit from the “data exhaust” inherent in all of the digital tools pervading the academy.

I think this is part of “doing Big Data right” within education, as I described in the Campus Technology issue from January of this year:

• I believe that the concept of combining data from multiple sources on a large scale to create unique insights will be very important for education in the long term. But right now the focus is too much on enterprise software solutions to vague problems with ill-defined data. The real potential in the short term is for consumer-driven tools to allow experimentation with new data, which will eventually lead to enterprise-class solutions.[7]

– – – –

References

[1] David Lazer, Ryan Kennedy, Gary King and Alessandro Vespignani, “The Parable of Google Flu: Traps in Big Data Analysis,” Science Magazine, Vol. 343, March 14, 2014. Accessed at http://gking.harvard.edu/files/gking/files/0314policyforumff.pdf

[2] Steve Lohr, “Google Flu Trends: The Limits of Big Data,” New York Times, March 28, 2014. Accessed at http://bits.blogs.nytimes.com/2014/03/28/google-flu-trends-the-limits-of-big-data/?_php=true&_type=blogs&_r=0

[3] Tim Hartford, “Big Data: Are We Making a Big Mistake?” Financial Times Magazine,  March 28, 2014. Accessed at http://www.ft.com/cms/s/2/21a6e7d8-b479-11e3-a09a-00144feabdc0.html#axzz2xOBqK8oq

[4] Marc Parry, “Recent Big-Data Struggles Are ‘Birthing Pains,’ Researchers Say,” The Chronicle of Higher Education, March 28, 2014. Accessed at http://chronicle.com/article/Recent-Big-Data-Struggles-Are/145625/

[5] Mike Caulfield, “Doing Big Data and Analytics Right,” Hapgood Blog, March 28, 2014. Accessed at http://hapgood.us/2014/03/28/doing-big-data-and-analytics-right/

[6] Virginia Backaitis, “Analytics 3.0: Beyond Big Data,” CMS Wire, March 28, 2014. Accessed at http://www.cmswire.com/cms/big-data/analytics-30-beyond-big-data-024672.php

[7] David Raths, “What’s Hot, What’s Not 2014,” Campus Technology, January 23, 2014. Accessed at http://campustechnology.com/Articles/2014/01/23/Whats-Hot-Whats-Not-2014.aspx?Page=4