Last year, we released a solution in the US, known by the name of SIGMA – the Student Individualised Growth Model and Assessment tool. It allows educational institutions to use data more effectively to predict student outcomes by identifying at-risk students, and tracking their proactive management. It’s one answer to the question “Why is Business Intelligence in education so important?”. And also a powerful example of using learning analytics to support students.

The quote at the beginning of the SIGMA overview, from Carnegie Corporation of New York, explains why the issue of school drop out is so critical:

  Today, young people who leave high school without excellent and flexible reading and writing skills stand at a great disadvantage. In the past, those students who dropped out of high school could count on an array of options for establishing a productive and successful life. But in a society driven by knowledge and ever-accelerating demands for reading and writing skills, very few options exist for young people lacking a high school diploma.  

The decision to drop out of school results from a process of increasing disengagement that can begin as early as primary school. No single set or combination of generalised risk factors exist that will identify, with absolute certainty, whether a student will drop out. This is because Predictive Analysis—using historical data to anticipate future outcomes—is not an exact science. Research does suggest, however, there is a relationship between key early indicators that can help to identify which students are less likely to graduate on time or drop out altogether. The most common reasons for dropping out of school include:

  • Lack of educational support
  • Outside influence
  • Special needs
  • Financial problems

But knowing what the factors isn’t the same as being able to use that information to prevent drop outs. The SIGMA business intelligence solution, using factors identified through risk assessment, is able to create a series of reports. The example below shows an example of an early warning system ‘on-track indicator’. The graph consists of three areas - the student identifier, risk level, and visualisation tool. The Risk Level Summary consists of both a current school year’s risk index score and a longitudinal view of the student’s risk index, which examines the entirety of the student’s academic record.

SIGMA learning analytics report

There are two approaches to building solutions to identify at-risk students—business intelligence and predictive analytics:

  • Business intelligence is a forensic examination of historical data representing a ―”snap shot-in-time” view of the student, providing educators with insight into the student’s performance. It is the process of gathering, storing, analysing, and accessing targeted school and student data to aid stakeholders in making timely decisions based on the most up-to-date information.
  • Predictive analytics is a technique of applying statistical models to determine likely outcomes by examining historical records. The system assigns a mathematical index score to each student tracked by the system. Each model is unique in that it takes into account local factors found within the school system, the surrounding community, and other influences identified as relevant by the local education stakeholders. These models incorporate both protective and negative factors. An example of a protective factor would be the inclusion of classes and programs such as music, art, advanced placement, after school activities, or athletic programs that capture the interest of the student. Negative factors include poor attendance, low grades, or high levels of negative behaviour incidents.

Although the whitepaper on the SIGMA model is designed for an American audience, there are strong parallels to the Australian education system, and lessons that are applicable – and it is definitely worth a read if you want to explore more.

Learn More

Download the whitepaper on Microsoft’s Student Individualised Growth Model and Assessment (SIGMA)

 

For more on learning analytics in Australia, it’s worth looking into the Learning Analytics case study at John Paul College in Brisbane. I’m going up there at the end of the month, to talk at their ‘The Education Revolution in Action4’ conference, so I’m hoping to see it in action when I get there.