Data management for statewide longitudinal data systems
Improving how we manage and learn from data can positively impact our educational system. Master data management plays a key role.
If there’s one major advancement that can change the current realities of education as well as the future of how children learn, it is data. Schools are overflowing with data – attendance records, achievement data, even logs from mobile devices – and statewide longitudinal data systems (SLDS) are emerging as a solution to storing, managing and using this data to drive decisions. Yet even though these systems are proving invaluable, there are many ways to make them more efficient.
An important first step in leveraging this resource is establishing sound data management practices. In Implement, Improve and Expand your Statewide Longitudinal Data System: Creating a Culture of Data in Education, by Jamie McQuiggan and Armistead Sapp, master data management (MDM) is discussed as it pertains to the volumes of educational longitudinal data in an SLDS. MDM is defined as synchronized, defined and uniformed business data processes that facilitate consistent, reliable data to feed every operational and decision-support application, giving stakeholders access to accurate data.
Data quality and a single version of the truth
Accurate data and the concept of one version of the truth are the goals of MDM; they’re the ultimate guiding principles of using MDM in educational databases. If users can’t trust the output of the system and create actions surrounding it, it’s not worth much to anyone.
In SLDSs, databases are, by definition, linked and records merged. When a student transitions from elementary to middle school, or high school to college, there should be a clear, simple way for her record to be merged without duplication or error. MDM leads to less redundancy in efforts among stake holding agencies. It also leads to high levels of data quality and a more agile database in the face of policy change.
More context through metadata management
Metadata management is a component of MDM, referring broadly to data about data. Metadata management associates clear and universal definitions with data fields across contributing databases; it gives context and reliability to data. In an SLDS, it might be that each school has the same definition for a field, and limits input in the same way (i.e. text only in “name” field.) Metadata management leads to more useful and contextualized data.
Architectures to store, merge and make data available
System architecture is another important component to MDM. SLDSs are almost always constructed as a centralized data warehouse, in which all contributing sources send data to one place where data is stored, merged and available to the group (based on a tiered security access protocol), or a federated data system.
A federated data system provides for the necessities of a common database while still allowing each agency to maintain ownership of (and responsibility for maintaining) its data. When data is requested from another agency, a virtual data broker compiles the data from each partner agency and reports it back. These two architectures are the most common and best suited for SLDSs.
Using MDM as the data management principles for an SLDS gives the system a strong framework to ensure that student data is fully utilized and secure. MDM is a key concept that should inform a state’s planning and organizational decisions, providing guidance on dealing with data quality metadata management and system architecture.
- Once a data management system is in place, it has to be governed. Learn how to get started by downloading the white paper, Data Governance for Master Data Management and Beyond.
- More data management insights.