Interventions for at-risk students are being transformed through advanced technology, especially in the identification process for which students might be at-risk.
Traditionally, to identify which students are at-risk, the district or school sets standards in areas like attendance, grades, assessment scores, behavior incidents, etc. When students cross a certain threshold on any of these data points, they are considered for intervention.
However, the problem with this process is two-fold. Firstly, these thresholds are often educated guesses of what students need to achieve. It can be difficult to know if it is actually an indication of a student being at-risk of falling behind or dropping out in the long term. Secondly, usually only a single data point—not the wholistic view of the student—is used, which may be an anomaly.
The trouble—and the opportunity—is the overwhelming amount of information available on each student. It feels nearly impossible to think about all of these data points at once. We need to use this often-scattered data to make informed decisions that are more cohesive and indicative of true risk. With finite resources, better at-risk identification helps us provide resources more equitably.
How can technology empower this improved decision making? I’ll dive into four areas that are transforming at-risk student identification.
One of the biggest and most important hurdles districts need to overcome is integrating disparate data sources. We collect information on students everywhere, from student information systems (SIS) and state assessment systems to finance, transportation, and HR systems, not to mention sources of information like social & emotional learning (SEL) survey data or statewide longitudinal systems. Just think about all the sources of information out there.
Data management technology enables districts to create a data warehouse with integrated data connectors so that all of this information flows into a single source of truth; transforming the process of at-risk student identification.
With access to more than just a single assessment score or a student’s recent attendance records, we can not only better identify which students need intervention, but we can also better identify which areas they might need additional help with (what specifically are they struggling with?) and the best strategy to help them (did they score low because they have moved 3 times in the last year?).
Traditionally, if we want to use more than just a single data point, a comprehensive view of the student needs to be created. This time-consuming task is almost always out of date as soon as it is created. This is especially true if the information needs to be analyzed or put into graphs or charts to more easily identify trends. I have seen school districts compile this information into thick binders, using processes that can only be done annually because of the time it takes. To really understand the current level of each student’s achievement, we need ongoing access to the most recent information.
Furthermore, near real-time access to data makes flexible grouping in RTI or MTSS an actual possibility, not just a theoretical best practice. As soon as students start showing progress, it can be systematically identified and reported out on.
A cutting-edge topic in education, predictive analytics allow teachers and leaders to more effectively blend the science of data with the art decision-making. This enables school districts to redefine their thresholds of success to be more accurate. Rather than making an educated guess that 90% attendance is required, predictive analytics might identify that in fact students with 85% attendance are successful.
This is what is really powerful about predictive analytics—it uses your specific data to identify students at risk. It enables us to really focus on those students that need our attention most. Closing the gap on over- or under-identifying students enables us to more equitably provide the right resources to the right students.
Beyond showing students “in the green” or “in the red,” predictive analytics can provide shades of red (who needs to most help and who is just on the edge of green). Giving us a tool to answer the tough questions such as prioritizing the use of limited resources.
Student data wall showing risk indicators
More than just access to a wholistic view of students in a timely fashion, data visualizations and role-based dashboards empower educators to take informed action quickly. For example, a quick view at a color-coded chart can easily tell a classroom teacher which students might need extra attention in class today. Trend charts can give a principal quick insight into the overall direction of student achievement. And patterns can be quickly spotted by an intervention specialist to create skill-based small groups.
Additionally, the ability to create cohorts of students and drill down to specific students enables us to see patterns as well as discern individual traits. When we are empowered with this insight at a glance or in just a couple of clicks, a transformation of how we spend our time takes place. Rather than spending time on an analysis of students, we spend our time focused on the actual instruction.
The process of at-risk identification is transforming. Through data management, predictive analytics, and data visualization we are empowered to make better, data-driven decisions. Hoonuit partners with districts to put these systems in place to support your initiatives and goals.
Learn more about the intervention transformation in our white paper, Facilitating Effective Interventions: Using Data-Driven Insights to Transform Your Processes.