Predictive Analytics | Interview
PAR Framework Project Eyes Student Data Through a 'Multi-Institutional Lens'
A Q&A with WCET Executive Director Ellen Wagner
 WCET Executive Director Ellen Wagner |
WCET, the WICHE Cooperative for Educational Technologies,
has been awarded $2.56 million from the Bill & Melinda Gates Foundation to validate and extend the PAR Framework Project, ongoing research into the use of predictive analytics on a
large-scale federated database to support decision making that removes barriers
to student success.
For more than a year, WCET has directed the project, which
pools the efforts and contributed anonymized (de-identified) student data and
course records (now more that one million records) from currently 16
participating institutions in order to identify
trends and patterns that can predict student success. The new funding will
support the ongoing validation of the project's data definitions and the extension
of data model development and other related research and analysis protocols into
2014.
Six founding partner institutions contributed to the first year's work on the PAR Framework: American Public University System, Colorado Community College System, Rio Salado College, University of Hawaii System, University of Illinois Springfield, and the University of Phoenix. The already diverse profile of this group has been expanded since they have been joined by 10 new partners: Ashford University, Broward College, Capella University, Troy University, Lone Star College System, Penn State World Campus, Sinclair Community College, University of Central Florida, University of Maryland University College, and Western Governors University.
Ultimately, the PAR Project's ambitious work will benefit
educators more broadly by creating an easy-to-use shared resource that
leverages "big data" analytics to predict risks to student retention and degree
completion. Reports, dashboards, and decision support tools will ultimately inform
higher education and serve the broad goal of student success. Campus Technology
asked WCET Executive Director Ellen Wagner for comment on the
impact of the project and the significance of the extended research funding.
Mary Grush: Congratulations on your announcement of continued funding for the PAR Framework
project from the Bill & Melinda Gates Foundation.
Ellen Wagner: Thanks. We are pleased at the vote of confidence, and are looking forward to the next
18 months. As you know, we have been working on PAR for a little more than a
year now, and are very pleased that the foundation is interested in helping us
continue our explorations. We have much to learn, of course, but compared to
where we were a year ago, we have a far better sense of the true possibilities
of using predictive analytics to improve student success. For me that's as
exciting as anything could be.
Grush: Was the first year a kind of proof of concept?
Wagner: When we began, our Bill & Melinda Gates Foundation project officers challenged and encouraged us to see if we could find multiple institutions that would voluntarily contribute their anonymized student- and course-level records, and collectively help assemble a single federated
database to which one then could apply data mining techniques to look for
patterns of student loss and student momentum. We were able to do
that very thing. Additionally we built some predictive models that have already
yielded analyses that lead to actionable results. Our founding PAR partners are
committed to improving student success. And this is why they have been so willing to collaborate on building a resource that will help us identify the patterns that predict risks, so barriers can be
removed before they become problems.
Grush: Where are you now in pursuing this research?
Wagner: We now stand at 16 institutional partners. Our schools include representatives from the public, private, and for-profit sectors, including 2-year and 4-year institutions. Some are traditional, while some are progressive. Each institution
is committed to the success of their students. Each is engaged and on board to support the collective efforts to find new intelligence to help make better decisions about what keeps students in school, and what keeps them engaged in their learning experiences. PAR is an effort by educators, for educators, and uses a multi-institutional lens to look
for patterns of student loss and momentum. By looking across institutions, not
just within institutions, one has a chance to see some of the patterns of
student loss and momentum that are not visible when viewed through the lens of
a single institution.
Grush: Will this research take years to provide usable results or guidelines?
Wagner: This is not the kind of basic research project where the questions need to be examined for years before we see results, although we hope that
what we are doing will spur on many more research questions. However, our team
is not asking those kinds of research questions. We are not talking about proving or disproving null hypotheses, at least not in the conventional sense. Our work on the PAR Framework is focused on developing a resource that helps anticipate risks to student success. PAR gives institutional decision makers the opportunity to "see" relationships between and among different variables and assess their effect upon the instructional experiences of each student whose records are included in the massive data set. Several of our founding partners report that they have already been able to apply intelligence
gleaned in our "first findings" in their "real world" of academic advising and
student support.
Grush: What is the most exciting part for you now, at this milestone in your work with the PAR project?
Wagner: We know from experience that educators rarely get involved in creating the products that directly transform our own educational practices. That's just simply not what educators do. We believe that the PAR Framework gives educators a significant resource
for putting the power of learning analytics to work in the service of improving
student success, and that PAR will help shape the direction of the
solutions and interventions being developed to improve rates of student
retention. We understand that innovation is a shared responsibility. The PAR
partners are excited by our progress to date and are all looking forward to contributing
to shared understandings of student success.
