Developing Strategies for Implementing Analytics

In their discussions over the past year, the EDUCAUSE IT Issues Panel touched several times on the issue of analytics and strategies on how best to implement analytics in higher education. Discussions ranged from implementing analytics to improve core administrative and business functions to how best to implement analytics in a predictive sense to ensure student success in academic programs.

In answering five questions, the following IT Issues panelists provided their thoughts on how analytics can improve higher education and how best to implement analytics:

• Jenny Crisp, QEP Director and Assistant Professor of English, Dalton State College
• Patrick Cronin, Vice President of Information Technology, Bridgewater State University
• Lisa M. Davis, Vice President for Information Services and Chief Information Officer, Georgetown University
• James Kulich, Vice President and Chief Information Officer, Elmhurst College
• Celeste M. Schwartz, Vice President for Information Technology and College Services, Montgomery County Community College

How can analytics help higher education?

Celeste Schwartz: Analytics are being used in administrative and academic areas in colleges and universities. Typically in higher education analytics are in one of three broad categories: institutional business, student engagement analytics, and student learning analytics. Analytics help institutions better identify opportunities and gaps, and areas for improvement.

James Kulich: The primary benefit modern analytics brings to our world of higher education is the potential these tools offer to improve the experiences of students, faculty, staff, and others whom we look to influence. Technologies like customer relationship management (CRM) allow us, via embedded analytics, to more personally and more persuasively engage prospective students who have the best potential to fit well within our institutions. Analytics are becoming an increasingly valuable arrow in our quiver for addressing, as an industry, public pressure to deliver greater value at lower cost.

Jenny Crisp: Analytics can help those of us with a primary teaching mission serve our students better. We can better advise them when we have broader data about which students do well in our various programs. We also can use that broad data to help us improve our programs and our courses. On a smaller scale, we can use data to help us identify students in need of intervention or remediation.

Patrick Cronin: Analytics tools offer a competitive advantage in an ever-increasing marketplace — they can inform and provide insight into institutional trends and changes. The risk for institutions is that their competitors are using analytics to better recruit, retain, and graduate students. Better retention and graduation rates mean a better brand and more attractiveness for prospective students.

What risks (or opportunities) do higher education institutions face as they develop their analytics strategies?

Cronin: One big risk is that useful analytics require accurate data. Using analytics forces institutions not only to clean and validate their data, but also to agree on data definitions and processes. This is actually an opportunity to provide transparency and unite a campus.

Lisa Davis: Higher education institutions face numerous risks as they develop their analytics strategies. A short list includes addressing organizational structure issues to produce the best analytics possible; ensuring data quality in order to produce reliable analytics; and constantly moving goalposts that force analytics to be as agile as possible. At the CIO level, these risks could be considered big opportunities if positioned and advanced in the appropriate ways. To succeed, analytics efforts become a broadly collaborative effort that is led by top executives, funded appropriately, and governed effectively.

Kulich: Some of the thornier challenges with the use of analytics regard ethics and perception: What about privacy? What dissonance exists between our sense of appropriate use of student-generated data and public perception of appropriate boundaries? How do we ensure that our use of analytics always puts the interests of students first? We have to answer these questions in honest and convincing ways to take advantage of the real potential analytics bring to strengthen student experience and institutional performance.

What kinds of data do institutions have that can contribute to analytics? Where should institutions start in considering analytics?

Davis: Data is everywhere. In the traditional sense, we have core administrative and student systems that can provide useful data (e.g., financial, advancement, human resources, and facilities systems; and admissions systems, student information systems, and learning management systems). However, this data really isn’t valuable unless it is actionable. The primary goal of any institution is to understand this and then further structure the data from multiple sources for effective use. Then, for each core system identified, higher education institutions should define a core set of dimensions and measures that would best understand and improve those areas.

Crisp: Data to support student success and improvement can come from a number of areas. For advising purposes, admissions scores and student success data over the long term can help point students toward courses and majors in which others like them have been successful. Within courses, particularly those with an online component, individual assignment data can be used to route students to suggested or required review material. For purposes of course and program design, student outcome data collected for assessment purposes can be invaluable. More ambitiously, student records data may be used in conjunction with admissions or financial aid data to help identify student populations that might benefit from institution-wide, focused student success programming.

Schwartz: While many institutions have implemented analytics in budget development, purchasing, financial management, personal management, and fundraising, the current primary area of focus is students. Data that may be useful for student-focused analytics are student demographic data, course progression, grading, financial aid information, student organization participation, placement testing information, program of study, educational goal, use of the library and tutoring center, interactions with advisors, information from early alert systems, course attendance, and course participation/interaction information. A starting point is looking deeply at information in your student information systems.

How do you strike the right balance between using the data that is available to an institution and providing meaningful and actionable analytics?

Davis: Many changes are happening in the higher education space, and many experiments are quickly occurring. Keeping up with new systems, new data, and ultimately new requirements is a constant struggle. In order to develop an analytics strategy, you need to understand the current state of your data and core applications. Then you need to determine the important reporting and analytic requirements of each core area. Finally, communicating the benefits of any analytics project is key to success.

Schwartz: You start with the questions you are trying to answer and then based on the questions you build the actionable analytics.  Actionable analytics will help to identify areas where the institution can develop or improve processes that support improvement in institutional offerings and services resulting in increases in student success.

Kulich: Effective use of analytics is a team sport. Good choices for analytics projects begin with conversations on campus about pain points, which good information might help address, and about opportunities where well-presented information might make an experience better. By adopting a design view, a good balance will be struck between the necessary day-to-day use of data to support normal campus rhythms and deeper analytics efforts that have the potential to be transformational.