The Community College Journey to Predictive Analytics
Published by: WCET | 6/5/2013
Tags: Collaboration/Community, Completion, Data And Analytics, Outcomes
Published by: WCET | 6/5/2013
Tags: Collaboration/Community, Completion, Data And Analytics, Outcomes
This week, Luzelma Canales, Executive Director, Resource Development & Administration, Lone Star College System brings us along the path that her system has followed to adopting predictive analytics. Luzelma and others will share more about their analytics journey at the WCET Leadership Summit: Under the Sword of Data. If you’re not able to join us, watch the Frontiers blog next week for a dispatch from Salt Lake.
If I ask ten people to define predictive analytics and how it is used on their respective campuses to improve student outcomes, I am likely to hear stories about how colleges are using their data systems to provide access to decision makers. Some may go on to describe the tools that they use to provide access to data to faculty, staff, and administers. Most often, however, the definition or stories are dependent on where colleges are in their journey on building a robust and rigorous culture of evidence to improve student outcomes.
Maya Angelou once said, “Wouldn’t take nothing for my journey now.” This quote got me to think about the journey that I’ve been on not only with the two colleges where I work but also the Achieving the Dream colleges that I provide technical assistance to as a data coach.
Shifting Our Thinking about Which Numbers to Track
As I reflect on the last seventeen years and how the use of data has evolved at community colleges, I think back to the early 1990s when our primary focus of data discussions was based on enrollment numbers. All of our questions where centered on whether we had sufficient sections to meet our enrollment targets. These discussions became even more critical during planning for a “base year,” which drives the funding formula for community colleges in Texas.
However in the late 1990s early 2000s, we began to hear more public criticism about graduation rates at community colleges and the push to move into performance based funding. It was during this time that I remember the dialogue at my college shifting to include more serious discussions about key student outcomes.
Nurturing an Analytical Culture
Achieving the Dream (ATD) hit our college at a time when we were ready to take our journey to the next level. When ATD announced that it would award a small grant to community colleges that committed to rethink their approach to student success by building a culture of evidence and engagement to improve student outcomes, my college jumped at the opportunity and was selected as one of the first twenty-seven colleges invited to participate in 2004. I believe that ATD gave us the framework that we needed to make the full transition from an access/enrollment driven college to a college maintaining its commitment to access while focusing on student success through a rigorous review of policies and practices.
It was during the planning year that I saw a total transformation in not only the data we requested but also the shift in how the data was collected, analyzed, presented, and discussed. For the first time, I saw institutional researchers emerge as leaders in building a student success agenda based on a basic form of analytics. Instead of walking into meetings with reports of raw data, institutional researchers brought forth synthesized reports with some analysis. I knew that we were moving in the right direction when our questions changed from how many to what impact.
A few years later (2011) we also had an opportunity participate in the planning year of Completion by Design, an initiative funded by the Bill & Melinda Gates Foundation. Completion by Design introduced our college to the Preventing Loss, Creating Momentum Framework. This framework asked colleges to identify the critical points where we were losing students and to develop strategies to improve student outcomes by understanding student progression from Connection to Entry to Progress to Completion. This framework forces us to once again reconsider the data we were reviewing and the questions we were asking.
During our participation in Completion by Design, Jenkins and Cho (2012) introduced us to a method that community colleges can utilize to understand student progress and outcomes by analyzing student course-taking behaviors to determine when students enter a program of study. In this report, “Get with the program: Accelerating community colleges students’ entry into and completion of programs of study,” Jenkins and Cho proposed that identifying clusters of three credit bearing courses that lead to a program of study can assist colleges in redirecting support services to students as they make progress in entering and completing a program of study. They suggest that student course-taking behaviors a greater predictor of progress into a program of study than declaring a major or intent.
PAR Framework Takes Us to Deeper Understanding
It is only at this stage of the journey that I believe that we can take full advantage of the Predictive Analytics Reporting (PAR) Framework. This framework takes our understanding of the student experience to much deeper level. Helping us understand where we are most at risk of losing our students gives us an opportunity to take a critical look at how our policies and practices are supporting or hindering student success. More importantly the PAR Framework gives our colleges an opportunity to place the data at the hands of those who can make the most difference in student success, our faculty and staff. The Lone Star College System joined the PAR Framework in the summer of 2012 as an institutional partner. Early participation in the PAR Framework is giving our college the opportunity to accelerate our capacity to leverage predictive analytics in a meaningful way that maximizes the use of our limited resources.
Take some time to reflect on your college’s journey in using data to understand the student experience at your respective colleges and remember that sustainable change takes time. You have to know where you are starting to develop the roadmap to where you want to get.
Executive Director, Resource Development & Administration
Lone Star College System
5000 Research Forest Drive
The Woodlands, TX 77381-4356
Email: Luzelma.Canales@lonestar.edu
1 reply on “The Community College Journey to Predictive Analytics”
[…] its goals, while PAR data scientist, Jeff Grant shared what the tools being developed can do, and Luzelma Canales, Lone Star Community College System, shared the experience of an institution participating in […]