Guest post from Nicole C. Alioto, Ph.D., IBM SPSS Predictive Analytics Solutions Architect
Follow Nicole on Twitter: @NYDatagirl
Now that the concept of “data-driven decision-making” has infiltrated the K12 environment, particularly around identifying students at risk for dropping out, the amount of data to examine and the time available for teachers and administrators to review data have not increased at the same rate.
Let’s look at a typical district data team’s time allocation. Figure 1 (left) illustrates that most of the team’s time is spent reviewing data from different systems, different reports, and different people – at different points in time.
When there are new systems providing data and staff turnover, the time required for deep data discussion increases at the expense of time with students. If data teams meet infrequently, the review level of the pyramid could take several months, leading to a delay in providing appropriate interventions.
At the next level of Figure 1, the data team has identified skills deficits and instructional priorities and now can determine the best intervention strategies to improve student learning outcomes. If the data team has the time, the group can come to consensus on strategies and move to the next level. If meeting time is limited or there exists a lack of agreement on intervention implementation, this level of the pyramid could take weeks (or longer).
The last level of the pyramid illustrates the lack of time available to work with students when too much time is spent in data review and intervention planning discussions.
The concept of data-driven instruction relies on getting information about student needs into the classroom quickly and efficiently so that corrective measures can be put in place, and student outcomes can improve.
What if data teams could reduce their meeting time and teachers could increase the amount of time they spend working with students at risk? With predictive analytics embedded into the data review and implementation process, districts can flip the pyramid around (see image to the right) and teachers can devote more time to working with students – which is the ultimate goal.
Predictive analytics allows district staff to uncover patterns in data, and when combined with business rules and scoring and optimization techniques, decisions can be enhanced through improved precision and expanded by getting more students help they need much earlier in the year.
Imagine a data team armed with the list of students at risk based on predictive models that run and deploy automatically AND the list of the best interventions for EACH student based on his/her achievement profile.
Imagine if teachers are certain the instructional strategies they implement with students will improve their likelihood of academic success BEFORE the first day of school.
Imagine students getting the appropriate support they need in the areas that will have the greatest impact on their student learning outcomes.
Imagine if a new student enrolls in the district and based on his/her academic and demographic profile, the staff can identify needs and strategies in real-time.
With predictive analytics at the core of data team efforts, districts will not just imagine, but realize these possibilities and transform the way education operates.
Stay tuned for my next post when I explore the concept of big data and its relationship to K12 education.
For more information:
· Read the success stories and learn how Gwinnett County Public Schools and Hamilton County Department of Education changed the education conversation
· Watch a demo to see how to improve student performance with IBM predictive analytics