Predictive Analytics

Early Warning for Students Showing Signs of Risk

early-warning-badge-line

Dropping out of school is a process not an event. Students who eventually drop out or have a major negative event happen typically show signs of risk for years before the actual event occurs.

mariam-icon
~ Dr. Mariam Azin

Are my students on track for graduating from high school?

Which schools are succeeding at graduating students from high school?

How do we ensure all schools are identifying students in need of support?

Are at-risk students getting the interventions and support they need?

Four Domains,
Supported by 24 Success Indicators

We gather data from different buckets—attendance, academics, behavior, and demographics—and analyze this data on a real-time basis simultaneously so as to connect students to the support they need early on.

Early-Warning-Framework

Timely identification of students in need of support – when signs of risk first show up.

EARLY

Other Early Warning approaches start at 8th or 9th grade—that’s not early enough. We identify students starting to show signs of risk as early as 1st grade—because intervention needs to start now.

ACCURATE

Other Early Warning approaches start at 8th or 9th grade—that’s not early enough. We identify students starting to show signs of risk as early as 1st grade—because intervention needs to start now.

Graph_1

CUSTOMIZED

One size does not fit all. Our predictive engine tunes insights to grade levels, schools, and students. 

When we look at patterns exhibited amongst students who eventually drop out, it is clear that risk is not a one-size-fits-all phenomenon. What is risky for a 3rd grader looks very different, and is more subtle, than what is risky for a 9th grader. If unaddressed, however, these risk factors can spread in breadth, frequency, and severity over time.

~Dr. Mariam Azin

TIMELY

Real-time data analytics for effective, targeted support today.

The Mazin Early Warning Framework simultaneously looks at dozens of data points spanning academics, behavior, attendance, and demographics on a real-time basis—and detects historically risky factors when they first start to emerge.

SENSITIVE

Nuanced information provided to educators to inform support and practice.

Educators operate under limited resources, they need information at a level of detail that allows them to make informed decisions regarding student needs and support.

Mazin Early Warning Framework, built by educational researchers with decades of experience, displays risk as a continuum, on multiple levels and across multiple domains.

Risk is not a dichotomous yes/no phenomenon. A student does not become “at-risk” from one day to the next when they hit a certain “threshold". The Mazin Early Warning Framework displays risk as a continuum, across multiple domains and on multiple levels. This makes the system extremely sensitive to detecting signs of risk when they first start to emerge—and provides nuanced information to help educators provide targeted support.

Mazin Education is the Industry Leader in Predictive Analytics for K-12 Education.

We are the only ones doing this to scale, on a real-time basis with quality, accuracy, and maximum usefulness assured.

The Mazin Early Warning Framework and Predictive Analytics Engine
powers the Clarity® Early Warning Platform

In use with nearly a million students nationwide

Algorithm is not destiny—just as the presence of factors that have historically been associated with greater risk of dropping out can increase the likelihood that negative student outcomes may happen, the embedding of protective factors can decrease that likelihood. This is why it is critically important to get accurate information to educators EARLY on —so that negative trajectories that are starting to emerge can be changed for the better.
~ Dr. Mariam Azin