If data lands in the ODS and no one uses it, does it empower educators?

Ed-Fi Community members are increasingly leveraging the Ed-Fi ODS as a source of data for business intelligence (BI) solutions, while many continue to develop it as a solution for compliance reporting. While the Ed-Fi vendor community provides many options for analytics based on the ODS data, some end-users wish to create their own custom reports and perform their own ad hoc analysis. The Analytics Middle Tier and the two Analytics Starter Kits recently published to the Ed-Fi Exchange aim to help this  group by simplifying the ODS data model, provisioning support for role-based data access, and providing sample visualizations. These solutions aim to empower those IT staff who are empowering their educators and administrators.

Analytics Middle Tier

The Analytics Middle Tier was announced in August. It is a collection of database views that mimic dimensional modeling. With much useful and encouraging feedback from a Special Interest Group that was convened to evaluate the proposed design, the Analytics Middle Tier installer is now available in the Ed-Fi Exchange.

What can we do with it? The Alliance has recreated its original Power BI Starter Kit using this new tool and a new Amazon QuickSight Starter Kit. We’ve also heard from one vendor who is already creating custom visualizations for an agency using the views.

PowerBI Starter Kit

Microsoft describes PowerBI as an “interactive data visualization BI (business intelligence) tool”. In 2017, the Alliance published a Starter Kit that produced Early Warning System dashboards based on ODS data. This kit included detailed deployment instructions and scripts for integrating Office 365 users and groups into the security. End users were encouraged to modify the solution for their own needs. However, the high degree of data normalization in the ODS was directly mirrored in the solution. Furthermore, the calculations in the data layer were correct but very confusing even for other experts. These were significant barriers to customization.

A second version of the PowerBI Starter Kit has now been posted to the Ed-Fi Exchange, utilizing the Analytics Middle Tier and rewritten / streamlined calculations. The end result is a set of dashboards that feel much more performant and should be easier to customize and/or learn from. As before, the project’s documentation in the Exchange fully describes the installation and customization process. The following screenshots demonstrate two of the of the reports available in the solution (the data comes from Ed-Fi’s “Glendale” set of semi-realistic-but-fake student data):

Analtyics Starter Kit Screenshot

Analtyics Starter Kit Screenshot

Amazon QuickSight

QuickSight is Amazon’s native business intelligence tool. While the user experience and reporting capabilities are less advanced than some tools, it has all of the elements needed to create easily-understood dashboards and is undergoing rapid innovation. AWS’s announcement of built-in machine learning through QuickSights ML Insights is particularly exciting, with the potential for helping data analysts quickly spot—and act on—anomalies in attendance, grades, etc.

At this time, QuickSight does not support portable assets for deploying dashboards. Each organization must build their dashboards from scratch. The QuickSight Starter Kit on the Exchange shows a series of sample visualizations, explains how the Analytics Middle Tier is used, and provides detailed formulas for recreating those visuals. The following screenshots show two demonstration dashboards, which include much of the same information shown in the PowerBI Starter Kit.

Amazon QuickSight Screenshot       Amazon QuickSight Screenshot

Usage and Feedback

To use any of these three tools, simply sign-in to the Ed-Fi Exchange and download the instructions.

These projects are provided as-is, but the Alliance welcomes feedback on additions or changes that would make these resources more user friendly. Feedback is best shared by raising a ticket on the Ed-Fi Tracker Exchange Contributions Project.