The Sloan Consortium’s Journal of Asynchronous Learning Networks Examines the Impact of Learning Analytics on Postsecondary Education
Special issue of JALN features papers on how learning analytics guide successful online programs.
Newburyport, MA (PRWEB) June 29, 2012
The effectiveness of teaching and learning is a critical accountability measure for higher education. In this month’s special issue of the Sloan Consortium’s (Sloan-C’s) Journal of Asynchronous Learning Networks, researchers share their methods for using learning analytics to guide long-term and just-in-time improvements.
“Advances in knowledge modeling and representation, data mining, and analytics are creating a foundation for new models of knowledge development and analysis,” said Karen Swan, Stukel Professor of Educational Leadership, University of Illinois Springfield (UIS); research associate, UIS Center for Online Learning, Research and Service; and editor of this special issue of JALN. “Perhaps nowhere are these new models more needed than in education. Because in online education it is possible to easily collect an enormous amount of information on individual students in real time, online learning seems to be on the frontlines of the learning analytics movement, and so online educators need to inform themselves about learning analytics.”
This special issue of JALN brings together scholars and practitioners who share practical advice, specific approaches, and illuminating examples about using analytics to help improve learning in postsecondary contexts. Contributors represent West Virginia University, Nova Southeastern University, University of Hawaii System, Rio Salado Community College, Colorado Community College System, The Apollo Group (University of Phoenix), American Public University System, Ellucian, My College Foundation, University of Wollongong, City University of New York, and University of Central Florida.
Highlights from This Issue
In “The Evolution of Big Data and Learning Analytics in American Higher Education,” Tony Picciano introduces the concepts of learning analytics, providing basic definitions, possible uses, and potential concerns about the uncritical application of analytic techniques.
Chuck Dziuban, Patsy Moskal, Tom Cavanaugh and Andre Watts provide concrete examples of how data can be used to inform action in higher education in, “Analytics that Inform the University: Using Data You Already Have,” with simultaneous “top-down” and “bottom-up” views of what is happening across the university.
In “Learning Analytics: A Case Study of the Process of the Design of Visualizations,” Martin Olmos and Linda Corrin, from the Graduate School of Medicine of University of Wollongong, Australia, discuss how visualizations of big data support human interpretation. The authors argue that finding the best way to represent data requires human interpretation.
Vernon Smith, Adam Lange and Dan Huston present a generalizable case study in “Predictive Modeling to Forecast Student Outcomes and Drive Effective Interventions in Online Community College Courses.” They show how postsecondary institutions can build statistical models that predict student outcomes and so inform interventions to support student success. They discuss how Rio Salado College is developing systems that combine data from student records with real time data to accurately predict which students are at high, medium, and low risk to succeed in individual courses.
In “The PAR Framework Proof of Concept: Initial Findings from a Multi-Institutional Analysis of Federated Postsecondary Data,” Phil Ice and colleagues describe their attempt to use learning analytics at an inter-institutional level to explore common factors contributing to student success and failure in postsecondary online courses and programs. Participants in the PAR (Predictive Analytics Reporting) project hope that once such common factors are identified, common interventions can be devised to reduce the unacceptably high numbers of students who enroll in such programs but never receive post-secondary degrees. In its initial, proof-of-concept phase, project participants managed to federate 661,705 student records from six very different institutions. They created common definitions of both predictor and outcome variables that will enable better strategic planning, decision-making and pedagogical research across online postsecondary contexts.