Academics across the country see an increasing need to collect and use data to make decisions as they work to help students. The epidemic severely disrupted our schools, and many district students used any data to identify who is struggling and how best to provide assistance.
As districts respond to the growing epidemic situation, it is important to understand the factors that contribute the most to student success by making the most of all available data to improve student outcomes.
The ability to use data is limitless. Used properly, it can help districts make important decisions about setting goals and providing targeted support for students. Whether you are new or experienced in K-12 data analytics, here are three practical ways to apply data for better student outcomes.
1. Use data to see an overall picture to identify and support at-risk students
Academics can and should use data to gain an overall perspective on each student. A data point from a single observation never tells the whole story of a student. Capturing a student’s academic, behavioral, attendance, and engagement data can provide a deep, informed understanding of who the student is, where they are succeeding, and where development is needed. Dashboarding data from a variety of areas of interest can often illuminate trends and early warning signs, lending information to identify which students may need assistance.
A Mississippi high school tried to visualize the data based on their indigenous risk model consisting of three departments: attendance, discipline and grade. Each department had its own risk score from zero to three. Combining the three categories creates a total potential risk score from zero to nine. See chart Custom risk criteria Below for reference. For attendance, five or six days of absence from school will create a two-person attendance risk, a tendency towards high risk for absenteeism. Assuming the same student never missed an extra day at school, had no disciplinary events, and had all grades over 70, their total risk score would be two.
Specifying a unique and multi-layered rubric for each risk segment provides a natural way to analyze and analyze large amounts of information and data. In this example, the school administration discovered that chronic absences are responsible for the highest risk in their student population, with at least one risk point responsible for 97% of student absences. Disciplinary incidents were negligible overall, with some overall risk points coming from this category. Risk based on low performance in the classroom revealed an interesting but problematic pattern. Although some students were at risk for having low grade grades, most students in this group had an overall high-risk score (average six). In addition, the data revealed that students who failed in one classroom usually failed in at least one other subject.
|# Absence||# Violation||Grade below # 60||Score|
|0 – 1||0||0||0|
|2 – 4||1 – 2||1||1|
|5 – 6||3||2||2|
|7 or more||4 or more||3 or more||3|
|0 – 3||0 – 3||0 – 3||0 – 9|
Filtering and comparing results by grade level and other demographic factors allows educators to see if there is a difference based on students’ current situation (e.g., facing homelessness or being in a program after school). In other words, this information indicated that some students, more than others, were more often observed as overall high-risk or high-risk by certain departments.