Over the past two years, eight California State University campuses have been involved in a collaborative student analytics initiative involving Oracle’s Student Information Analytics (SIA) system. The intended outcome of this collaboration is a forecasting dashboard for college planners and department leaders to make smarter decision about intervention and service planning.
During the planning phase of this effort, a common question that was asked among the colleges participating was whether each institution was ready for analytics. To answer this question, I developed an Analytics Spectrum to give college and university leaders an easy way to assess whether an institution is indeed ready for analytics.
What analytics is in higher education sometimes depends on the vendor who is making a pitch. More often than not, administrators feel the Dashboard Dazzle of charts and graphs that seem to be bringing out complex relationships among student data that only an expensive product can produce. The harsh reality is this: third-party vendors have been raking in hundreds of millions of dollars to give public institutions the same visualizations that could be produced in Excel.
In Data Science for Higher Education, I defined analytics in higher education as the collection and analysis of data to produce information that reduces uncertainty about the future. The temporal component here is key in separating analytics from what Dashboard Dazzle is really doing: reporting. If institutions want to truly implement analytics and begin driving their forecasting and planning efforts with data, a clear understanding of where they are — and where they want to be — on the Analytics Spectrum is necessary.
The Analytics Spectrum
In 2009, the senior vice president and chief marketing officer of SAS Institute Jim Davis illustrated that analytics has eight steps:
- Standard reports. What happened? When did it happen? Example: monthly financial reports.
- Ad hoc reports. How many? How often? Where? Example: custom reports.
- Query drill down/OLAP. Where exactly is the problem? How do I find the answers? Example: data discovery about types of cell phone users and their calling behavior.
- Alerts. When should I react? What actions are needed now? Example: CPU utilization mentioned earlier.
- Statistical analysis. Why is this happening? What opportunities am I missing?Example: why are more bank customers refinancing their homes?
- Forecasting. What if these trends continue? How much is needed? When will it be needed? Example: retailers can predict demand for products from store to store.
- Predictive modeling. What will happen next? How will it affect my business? Example: casinos predict which VIP customers will be more interested in particular vacation packages.
- Optimization. How do we do things better? What is the best decision for a complex problem? Example: what is best way to optimize IT infrastructure given multiple, conflicting business and resource constraints?
These eight steps are accounted for in The Analytics Spectrum, which is broken up into three eras: the Era of Reports, the Era of Intelligence, and the Era of Optimization.
The Era of Reports
- What happened?
- How many? How often? Where?
- What exactly is the problem?
Methods: Standard reporting, ad hoc reporting, query/drill-down
In this Era, the primary focus of all analytics is descriptive analysis. It’s called the Era of Reports because the primary output is, well, reports. This is where majority of institutions are and will continue to be, even after investing in expensive third-party analytics suites. Dashboard Dazzle leads people to believe that evolving from this Era to the next is only a matter of buying the right product, when in fact you must have the right people and processes in place before any product is worth the investment and license costs.
To begin evolving from the Era of Reports, complete the Analytics Readiness Checklist for Higher Education.
The Era of Intelligence
- What could happen?
- What will happen next?
- What if these trends continue?
- What actions are needed?
Methods: Simulation, Forecasting, Predictive Modeling, Alerts
The Era of Intelligence is the sweet spot of where institutions should be. Here, the focus is on creating actionable intelligence, projecting outcomes based on certain constraints, and forecasting possibilities given variables that you can and cannot control. I argue that this is the ideal place for institutions to be, since getting any more rigorous involves the type of optimization techniques that might not be well suited for predicting the behavior of people.
THE ERA OF Optimization
- What is the best that can happen?
- What is the best that can happen given variability?
- How do we do things better?
- What is the best decision for a complex problem?
Methods: Machine learning, data science in higher education
The Era of Optimization comes about almost seamlessly once the Era of Intelligence has been fully digested across an institution. In this Era, computational tools are used to automate the processing of data into information and the dissemination of information so that it can be molded into actionable intelligence. Every institution should strive to reach the Era of Optimization because it will force planners to master the Era of Intelligence — which is where I think all institutions should get to and maintain.
What Era Are You In?
Are you still in the Era of Reports, and looking to evolve to the next Era? Perhaps you’re looking for a way to help guide the evolution of your institution toward a data-driven, efficient Era of Intelligence? Consider the Analytics Readiness Checklist for Higher Education (PDF), a free guide designed to provide leaders with decision support talking points for driving institutional efforts in analytics planning. It’s short, it’s free, and it’s got all the steps necessary to start the right conversations.