Will Obama’s Plan to Use Big Data to Reduce College Costs Work?
No stranger to the benefits of big data analytics, the Whitehouse recently announced a plan aimed at using big data to create market aids targeted to help students to be better informed shoppers of colleges and universities. The plan aims to reform the higher education, and how it gets funded. Will it work?
The plan comes as part of the president’s plan to make college more affordable. The administration cites College Board and Census data that says the average tuition at a public four-year college has increased by more than 250 percent over the past three decades, while incomes for typical families have grown by only 16 percent. Last month, the president announced a plan that they hope will eliminate barriers that they say stands in the way of competition and innovation.
The new program involves the use big data technologies and techniques to develop a ranking system by 2015 that can be used to tie financial aid to college performance, and challenges states to fund public colleges based on that ranking system.
“The federal government provides over $150 billion each year in student financial aid, while states collectively invest over $70 billion in public colleges and universities,” said the administration in a recent release. “Almost all of these resources are allocated among colleges based on the number of students who enroll, not the number who earn degrees or what they learn. President Obama’s plan will connect student aid to outcomes, which will in turn drive a better, more affordable education for all students.”
To make these connections, the plan calls for the Department of Education to develop a new ratings system that they say will help students compare the value offered by colleges, and provide competitive encouragement for colleges to improve. “These ratings will compare colleges with similar missions and identify colleges that do the most to help students from disadvantaged backgrounds as well as colleges that are improving their performance,” the plan says.
In order to accomplish this, they intend to develop a model using a wide range of data points measuring such things as:
- Access, such as percentage of students receiving Pell grants;
- Affordability, such as average tuition, scholarships, and loan debt; and
- Outcomes, such as graduation and transfer rates, graduate earnings, and advanced degrees of college graduates.
“The Department will develop these ratings through public hearings around the country to gather the input of students and parents, state leaders, college presidents, and others with ideas on how to publish excellent ratings that put a fundamental premium on measuring value and ensure that access for those with economic or other disadvantages are encouraged, not discouraged,” says the report.
Will it work? Is this type of analysis valuable? Not everyone thinks so. Aside from the political questions involved, there is the question of whether a model that ranks such a subjective thing as higher education could possibly be valid. According to a recent article in the Washington Post, many colleges don’t believe so.
“With a powerful lobby in Washington, colleges have so far beat back such efforts, contending that no accountability system can effectively rate performance for 5,000 degree-granting schools that encompass two-year community colleges; the Ivy League; research universities; regional state schools; private, nonprofit liberal arts colleges; and for-profit institutions,” wrote Nick Anderson and Phillip Rucker in the Post. “Exactly how to compare colleges and judge outcomes is a matter of fierce debate within academia. Many colleges can’t even agree on which schools are their peers. There are major questions about how to calculate graduation rates and measure the earnings potential of graduates.”
It’s well documented that the Obama administration is no stranger to the big data trend, largely crediting the president’s most recent electoral win to their use of predictive analytics. However, as notable political statistician, Nate Silver argued earlier this year, an over-abundance of data can be dangerous and counter-productive if managed improperly.
Whether this is a good use of data remains to be seen, but it does demonstrate the magical allure that big data analytics holds over people in planning positions. Is gathering data and manufacturing a value ranking a plan worth following? The U.S. appears on the road to find out.