Group items tagged

"The Predictive Analytics Reporting (PAR) Framework is a multi-institutional data mining project that brings together 2 year, 4 year, public, proprietary, traditional, and progressive institutions to collaborate on identifying points of student loss and to find effective practices that improve student retention in U.S. higher education. Current efforts focus on removing barriers to student success in online and blended programs. With sixteen WCET member institutions, over 1,700,000 anonymized student records and 8,100,000 institutionally de-identified course level records, the PAR Framework offers educational stakeholders a unique multi-instituional lens for examining dimensions of student success from both unified and contextual perspectives. More benefits of PAR Framework participation."

"The big theme of Mr. Shirky's talk was openness. Taking advantage of technology, he argued, will require "doing more sharing than we're used to and then learning to live with results." He underscored the power of sharing with an anecdote about what happened when the Smithsonian Institution made a cache of photographs available on Flickr, the photo-sharing site. Users catalogued the archive with tags that reflected an unpredictable range of interests, including facial hair, the history of photography, and the fiction genre known as Steampunk. "There's all kinds of hidden value in our systems which you can't even understand until you open them up to see what people do with them," Mr. Shirky said. "The thing that drives me craziest in conversations with large institutions about large data sets is they want to know in advance what will happen. Why should we open up our data? To which the answer is, you open up your data to see where the value is. It's the value you can't even predict until you try it that you get back.""