One in two new college graduates are either jobless or underemployed, reported the Associated Press in April. With millions of college graduates struggling to get a toehold in the job market, the pressure is mounting for universities to justify the high price tag of a bachelor’s degree.

Some universities are turning to the hard, cold numbers of predictive analytics to help students get the most out of their college education. In a fascinating article, The New York Times and The Chronicle of Higher Education teamed up to provide glimpses into the ways that Arizona State University, Austin Peay State, and other universities are using analytics software to predict how well students are likely to perform in a class or major, their progress in completing assignments, and to recommend future classes.

At Rio Salado, a community college with about 70,000 students, an employee of the college who was also a computer science major at nearby Arizona State developed software that mined the behavioral data of students:

Mr. Lange and his colleagues had found that by the eighth day of class they could predict, with 70 percent accuracy, whether a student would score a “C” or better. Mr. Lange built a system, rolled out in 2009, that sent professors frequently updated alerts about how well each student was predicted to do, based on their course performance and online behavior.

In another example, students at a midsize university near Nashville receive a Netflix-like list of recommended courses via an algorithm that takes into account the students’ transcripts, along with past students’ grades and standardized test scores to make suggestions for classes they are likely to excel at.

If it wasn’t already obvious that higher education is a multibillion dollar industry, these examples and others in the article make it abundantly clear: like retailers, colleges can now personalize their offerings for their customers students.

This might not necessarily be bad news. Just like marketers, professors benefit from the greater insights that data mining provides about the people they serve. A professor who understands where his or her students are struggling with the assignments, for example, can address those issues before students bomb the midterm exam.

The downside is the risk of reducing a college education into a series of numbers and statistics. It is also not clear if an education dictated by predictive analytics will increase a graduate’s chances of getting a job. Hopefully, university administrators and professors will be smart enough to realize that analyzing big data is only part of the answer.