Data mining higher education records in search of improved outcomes

The Bill and Melinda Gates Foundation has just funded a potentially very interesting and useful study at WCET. Unfortunately, I have to go through a few explanatory layers of acronyms in order to explain that sentence, but stick with me. WCET stands for the WICHE Cooperative for Educational Technologies, while WICHE stands for The Western Interstate Commission for Higher Education. WICHE was created to facilitate sharing of resources among the higher education systems of the West, and has 15 member states.  

The grant focuses on identifying variables that impact student retention and progression.  The study itself will involve data mining of student records from an interesting mixture of instituions: American Public University System, Colorado Community College System, Rio Salado College, University of Hawaii System, University of Illinois Springfield, and University of Phoenix.  The aggregate data to be analyzed contains records of over 400,000 students:

Each of these postsecondary entities has performed extensive research on their own datasets.  The combination of large data sets across a wide variety of types of institutions (public/private, two-year/four-year, traditional/non-traditional) will seek to find patterns identifiable only through analyzing large data sets representing a widely diverse set of students.

The collection and mining of data relating to learning with the goal of understanding and improving outcomes is sometimes called Learning Analytics. Ellen Wagner, WCET Executive Director, discusses the aims of the grant and the general approach of Learning Analytics in a recent blog. As she indicates, there is now a huge amount of data in the online learning realm that almost certainly contains many important insights into the interactions of students with various aspects of that approach. Uncovering these insights should enable improvements in online learning that will lead to significantly improved learning, retention, and graduation outcomes.  

Through large n studies such as these, we can greatly improve our chances to uncover some of those critical insights. WCET and the institutions that have agreed to participate should be congratulated. Hopefully, this study will serve as a stimulus for other large n, multiple institution studies. 

 

Comments

[Mike]

The topic of Learning Analytics is interesting. I think that it is good to collect data to determine the appropriate interactions to engage students appropriately. It should include information that helps the institutions and students. More colleges and universities are moving away from traditional classes. This seems to be a beneficial way to find ways to develop it more.

[Janet]

A study of this magnitude is important for the information it will provide educators. Moreover, to the extent the US federal government is increasingly involved in trying to mandate accountability for institutions of higher education, it represents educators' focus on evaluating their own impacts.