OEA Overview

Open Education Analytics (OEA) is a community of education systems and tech partners, coordinated by Microsoft Education, collaborating to develop open source data analytics solutions for the education sector built on Azure Synapse Analytics and PowerBI.

• OEA Website: https://openeducationanalytics.org
• OEA GitHub Repository: https://github.com/OpenEducationAnalytics/OpenEduAnalytics
• OEA Overview Video: OEA Overview on YouTube

OEA Communities Ignite and Expand

Our OEA communities are taking off. The OEA Global Community has now expanded to include an Australian OEA Community and a US OEA Community. In addition, OEA Partners are starting to plan OEA Communities in the UK and Brazil. The number of education systems actively participating in meetings has increased dramatically since our initial launch in the fall. In particular, the Australian OEA community is extremely active with 53 community members representing both K-12 and higher education systems from across Australia. The Community is inspired by the stories and insights of those systems sharing their OEA-based use cases. In recent calls we have seen:

• University of New South Wales’ predictive model of students at academic risk
• Pennsylvania State University System’s predictive model of enrollment patterns
• Catholic Diocese of Maitland Newcastle and Monash University (Australia) sharing their approaches to system-wide data governance
• Fresno Unified School District’s (California) predictive model of student chronic absenteeism
• Forsyth County School’s (Georgia) predictive model and automated notification system for students at risk for high school graduation
• Helsinki Education Division’s dashboards and modelling of student well-being:

Figure 1: Pedagogical well being PowerBI dashboard by Helsinki Education Division

All of these use case examples are being collected in the Global OEA Community Team, in a library of presentations and materials. We are working to make many of these use cases open-source packages in the OEA GitHub repository so that education systems can leverage them to jumpstart projects.

Learning Analytics Case Study with Lastinger Institute

A new OEA case study from the University of Florida’s Lastinger Institute was recently published.

The Institute is using OEA to provide analytics to researchers and school districts using Math Nation, an app used by over 3 million students in the US.

See the great story here to have an idea of how OEA can be used to provide a research solution for edtech efficacy.

Data Analytics and AI

Our OEA data science partner, Kwantum Analytics, recently developed a predictive model of chronic absenteeism with Fresno Unified School District in California. Education research shows that students who are chronically absent have lower educational and health outcomes later in life. With over 70,000 students coming from high poverty and rich ethnic diversity, solving the problem of chronic absenteeism through preventive and proactive interventions can ensure that every student has greater chances of success. Aside from its impact on the students and the schools in the district, chronic absenteeism also affects how schools are reimbursed, so even a 1% increase in absenteeism rates could lead to a loss of millions of dollars annually.

For our predictive modeling work, the focus was on students in kindergarten to third grade, where chronic absenteeism (defined as missing more than 10% of school days) is over 40%. The model development cycle started with clearly defining the use case using the OEA Use Case template, stakeholder meetings, identifying needed data and landing it in the OEA Azure Synapse environment, data transformation and exploration, building and testing the models, and creating PowerBI dashboards that present the results of the predictive model.

Figure 2: Fresno’s Chronic Absenteeism PowerBI dashboard by Kwantum Analytics

The data team worked in partnership with Fresno’s Department of Program and Intervention to understand the scope of the problem and what could be done with the predictive model. We are currently exploring how to automate interventions to support the small team of professionals responsible for addressing chronic absenteeism in Fresno and tracking the impact of those interventions. Kwantum Analytics presented this predictive model at the March US OEA Community, and it is viewable in the Global OEA Community Team.

OEA Partners Expanding the Program

OEA Partners are expanding the scale of OEA. Since the start of the OEA Partner Program in September 2021, over 30 technical partners have joined the “entry” level of the program, and 12 have become OEA Advanced partners who are qualified to deploy OEA-based analytics solutions. OEA partners are also expanding the reach of Microsoft Data and AI through OEA. At BETT in London Community Brands is initiating a new UK OEA Customer Community that will be focusing on Multi Academy Trust and Independent school networks, which make up almost half of the UK K-12 education sector. In addition, Community Brands is leading efforts to communicate about OEA through posts like this on LinkedIn that describe what is needed to advance analytics and AI in the UK education sector. Similarly, Analytikus hosted a regional OEA webinar on March 3rd in Spanish for the Latin American region.