Enabling Predictive Analytics For Enrollment Management And Growth

Melody Childs, Associate Provost & CIO, University of Alabama in Huntsville
1229
2017
395
Melody Childs, Associate Provost & CIO, University of Alabama in Huntsville

Melody Childs, Associate Provost & CIO, University of Alabama in Huntsville

Role Of Big Data In The Education Space

Big data is top business priority as well as a top academic priority – but what does ‘big data’ really mean? The phrase ‘big data’ has become nearly cliché; however because there are so many different genotypes of data, applying what amounts to a taxonomy to the different kinds of ‘big data’ will serve individual institutions well when addressing challenges, formulating policy, and assessing priorities for investment.

For example, big data can be unstructured or structured. Examples of unstructured data include instructional and marketing videos, Learning Management System content, Facebook posts, Tweets, and other social media content, scholarly books, the text of magazine and journal articles, e-mail messages, word processing documents, and even campus maps. In other words, unstructured data emerges lacking a predetermined format that easily lends itself to statistical analysis, or data-mining. To get a sense of global scale, more than 30 billion pieces of (unstructured) content are shared on Facebook every month; while in the U.S. alone, more than 300,000 new books are published every year.

By contrast then, structured data is arranged in such a way (theoretically anyway) to make it easier to locate and extract useful information, perform sophisticated analytics, and develop research models or conclusions. In higher education, structured data comprises many forms including student records, donor prospect data, enrollment figures, and indicators related to student retention and success. Institutions also typically assemble another diverse set of structured data in the form of scholarly works such as dissertations, emergency contact information, student health records, and a whole host of other data that typically lives outside the confines of the ERP and even the enterprise data center. Finally, many if not most institutions collect or create research data including everything from peta bytes of satellite-generated weather data to vast amounts of human subjects material.

Understanding what data you have is a start, but what’s really important is understanding what you’re trying to achieve with all that big data to better serve institutional mission and goals, as well as meet the compliance requirements of the university. Although there are many more goals to be sure, a good start on a list of priorities for CIOs might be:
• Enable business analytics for improving efficiency and reducing the bottom line
• Satisfy compliance with federal, state, and regulatory mandates such as FERPA, HIPPA, ITAR, accreditation bodies, etc.
• Achieve competitive advantage in research by having a data repository that satisfies grant solicitation requirements for NOAA, NSF, NIH, and others
• Preserve the scholarly record

Management concerns for big data comprises another list in itself:

• Privacy and security
• Producing and cataloguing metadata
• Data retention and provenance policies
• Storage requirements
• Human capital and skill sets needed for all the above.

Read Also

Predictive Analytics Key Component of Customer Experience Management

Predictive Analytics Key Component of Customer Experience Management

Richard Barton, CIO, American Management Association
Predictive Analytics in Higher Education

Predictive Analytics in Higher Education

Linda Hartford, CIO, Northeast Wisconsin Technical College
Predictive Analytics And Child Protective Services

Predictive Analytics And Child Protective Services

David Elges, CIO, D.C. Government