There are five types of learning analytics: [hindsight] descriptive &  diagnostic; [insight] discovery; and [foresight] predictive and prescriptive. Each of these types of analytics answers a different question. Descriptive analytics answers the question: What is happening? Diagnostic analytics answers the question: Why did it happen? Predictive analytics answers the question: What is likely to happen? and Prescriptive analytics answers the question: What should I do about it?

When descriptive analytics merges with diagnostic analytics the result is real time information on demand with greater interactivity (e.g., student success dashboards!) When this type of real-time information is extended to the student themselves, counselors, advisors, and faculty … timely meaningful feedback is possible. Real-time data and meaningful student feedback is incredibly important.

Discovery and predictive analytics detect trends, clusters and exceptions. Prescriptive analytics simulates multiple potential courses of action in order to answer the question What is the best course of action? Each of these analytics have distinct tools and analytic methods. With increasing depth of analytics, structured data can yield reports, info-documents; info-apps and dashboard data (real-time information), performance metrics, as well as ad hoc query and analysis, data discovery and predictive analysis; whereas unstructured data can be used for search-based applications, search query; text/word analytics; and sentiment analysis.


Corcoran (M). (n.d.). The Five Types of Analytics. Retrieved from:

Schmarzo, W., (2014) Business Analytics: moving from descriptive to predictive analytics. Retrieved from:

#580EdTech; #Learninganalytics