Using Predictive Analytics to Improve Student Interventions

Posted by Courtney Stevens on Jul 12, 2018 1:15:57 PM

Using Predictive Analytics to Improve Student Interventions

 

Two decades of wide-ranging experience have shown me the importance of data-driven decision-making.


When I moved from Seattle, Washington, to Phoenix, Arizona, in 2001, I was well-prepared for the 23-degree jump in average daily temperature — much of my childhood had been spent following my parents around tropical locales in Africa and Southeast Asia. What I wasn’t expecting, however, was unexpected changes that were in store for my professional life.


I began my career as a mental health counselor working closely — often even one-on-one — with emotionally disabled students. This work was as challenging as it was fulfilling, and I enjoyed having the opportunity to help kids with their mental and emotional wellbeing. But when I arrived in Phoenix, the demand for teachers was so high that I soon found myself in a classroom environment. I learned the ins and outs of classroom teaching on the fly, but in doing so, I realized the incredible value of experiencing the public education system from a different perspective.


A Return to My Roots


While I ended up teaching for a number of years, my heart always remained in student support and working with student success which led me back into a counselor role, but this time as a high school guidance counselor. As such, when University High School — a college prep school catering to low-income minority students in Tolleson, Arizona — opened in 2006, I quickly jumped on the opportunity to build it’s student support structures and college and career readiness curriculum as its guidance counselor. My role at University High convinced me that I could make a difference from a leadership level, pushing me back to school to get my administrative degree.  This allowed me to step into the Principal position in 2010 — giving me yet another valuable perspective on the the public education system.


Four years later, one of my former mentors approached me with an unexpected — but very intriguing — opportunity. The Washington Elementary School District in North Phoenix/East Glendale was in the process of adopting an educational data platform, and was looking to hire someone to facilitate implementation and, in the longer term, build a data-driven culture. I’ve spent the last four years working to expand the district’s use of data and integrate a range of student-level technologies into its accountability infrastructure.


Discovering the Power of Data


My work in the Washington Elementary system opened my eyes to the true power of data in education. As a leading member of the district’s data initiative, I was given the leeway to experiment with different approaches, learn from mistakes, and develop a deep empirical knowledge of how data can — and should — be leveraged at the district, school, and classroom level.


This process was profoundly transformational for the district as a whole, and enabled us to discern both what we were doing well and where we needed to reevaluate our practices in order to achieve our big picture goals moving forward.


Witnessing the countless ways in which educational data drives results in the modern classroom prepared me to tackle my latest professional endeavor: joining Hoonuit’s team as the Vice President of Education Research.


Taking Data Utilization to the Next Level


By the end of my tenure with Washington Elementary, we had integrated a number of Hoonuit’s capabilities into our data infrastructure to drive better decision-making — including early warning with a plan to continuing implementing additional metrics.  


It has been fascinating to learn more about other districts working with Hoonuit, such as Elgin Public Schools, the second largest school district in Illinois. Instead of relying on established and often ineffective warning system “triggers” — flagging any student whose attendance falls below 95 percent, for instance — predictive analytics is allowing them to harness the power of machine learning to account for numerous interrelated variables, all of which play a part in determining whether a student is at risk of a bad outcome.


Using sophisticated machine learning algorithms, Hoonuit’s predictive analytics tool is able to execute remarkably comprehensive analyses of student performance, painting a picture of what high-risk students look like before traditional warning signs that they are  falling behind. This empowers educators to deliver additional support to students earlier — and in a more precise way — than previously possible.

 

College Readiness Predictions and Measurements


That said, a predictive analytics tool like Hoonuit doesn’t replace the need for human judgment — far from it. Ideally, educators should use predictive analytics as a way of processing volumes of data with an eye toward uncovering correlations that would otherwise go unnoticed.


Building Roster

Example student roster used to identify correlation between — and pockets of overlap among — students with missing credits, low test scores, low attendance, and high mobility to more accurately predict likelihood of on-time graduation.


As an educator with two decades of wide-ranging experience in the field, I’ve seen firsthand the value of such targeted, impactful data-driven insights. Interested in learning more? Click here to find out how predictive analytics can empower educators. Then visit our data analytics solution page to learn how Hoonuit empowers educators with multi-sourced, actionable data.

 

Learn More

Subscribe

Suscribe to our Blog

Stay up to date on the lastest stories from hoonuit


Topics : Data & Analytics

UPDATES FROM THE HOONUIT BLOG