The recent FTC workshop “Big Data: A Tool for Inclusion or Exclusion?” posed important questions on whether and how analytics could be used to restrict life chances for people rather than create economic and social opportunity. The answer lies in the hands of the user of the technology, not in the technology itself. The critical question is how people use, implement or otherwise act on the discoveries – the indicators, insights and evidence – that data analytics can uncover or reveal.
As with all knowledge, the value of the insights made possible by research and science depend in large part on the purposes they are used to advance and the environment in which they are deployed. Data analytics in education is a good example.
A study by Johns Hopkins University research professor Robert Balfanz shows that most students who eventually drop out can be identified as early as the sixth grade by their attendance, behavior and course performance. Using those indicators, it is possible to identify by the middle of ninth grade virtually everyone who will drop out. As Professor Balfanz put it recently, “These young men are waving their hands early and often to say they need help, but our educational and student-support systems aren’t organized to recognize and respond to their distress signals.”
What should be done with this knowledge? While one could worry that some might take an exclusionary path to perhaps stop wasting resources on students who are predicted to fail, the popular consensus created by recent educational innovators, such as Joel Klein, CEO of Amplify and former Chancellor of the New York City Department of Education, is to use these potential drop-out indicator as discoveries to develop a path of inclusion to take corrective actions early for these students, thereby promoting social and economic opportunity in education:
“Using data to help identify these students and give them meaningful supports and interventions as early as possible would have a significant impact on the number of students that graduate ready for success in either college or career. This isn’t the stuff of science fiction. These are actionable steps we can take right now, thanks to the power of technology.”
Which path is chosen is a matter for educators, not a matter of data analytics. It would be disheartening, to say the least, if policymakers, fearful of data analytics used as a possible tool of exclusion, were to, in essence, call for educators to put their heads in the sand – not to see, and worse, not to use for good, the indicators that data analytics discovers. Would anyone really argue that schools around the country would be better off not knowing the determinants of student failure, since it is possible that such knowledge might be used to discriminate against at-risk students?
In fact, many schools throughout the country are raising their heads, and hands, to apply data analytics to help their students succeed.
- Research shows that attendance, behavior and course performance can assess dropout risk in a way that allows schools to design early intervention systems to support students. Miami Carol City Senior High in Florida designed an intervention system based on this kind of data. In 2013, one-third of students flagged for missing school got back on track to graduation. Two-thirds of the students who were having behavioral problems made a turnaround.
- In Hamilton County Board of Education in Tennessee, early student interventions led to increased graduation rates by more than 8 percentage points and standardized test scores in math and reading by more than 10 percent thanks to applying predictive analytics.
- In Mobile County, Alabama the dropout rate has been nudged downward by three percent since the application of data analytics on a broad range of factors including demographic variable to identify at risk students.
As J. Alvin Wilbanks, CEO and superintendent of Gwinnett County Public Schools, the 14th largest school district in the U.S., put it, “data analytics can help us both identify the child and create a better picture of who they are, what areas they’re deficient in, and point to things we can do differently. As we perfect our use of analytics, I think we can even get to the point where it’ll suggest, this student is weak in fractions; here are some activities that can help improve that.”
These examples and countless more show that educators are putting discoveries from data analytics to use for the good of all students, using indicators not to exclude students but rather to develop personalized courses of study for each one to succeed.