Skip Navigation
small NCES header image

Chapter 1—What Is Data Governance and Why Does It Matter?

Data governance can be defined as both an organizational process and a structure; it establishes responsibility for data, organizing program area staff to collaboratively and continuously improve data quality through the systematic creation and enforcement of policies, roles, responsibilities, and procedures. As a structure, clear and specific roles and responsibilities are assigned and staff are held accountable for the quality of the data they manage. Ultimately though, data governance is not about who is in charge: it is about identifying existing or potential data problems and fixing them to prevent them from happening or recurring. As a continuous and iterative process, data governance is a systematic way of handling data throughout the information life cycle, from definition to retirement (see chapter 1 of Book Two: Planning and Developing an LDS).


dictionary

Data governance is both an organizational process and a structure; it establishes responsibility for data, organizing program area staff to collaboratively and continuously improve data quality through the systematic creation and enforcement of policies, roles, responsibilities, and procedures.


The process fosters coordinated responses to ongoing data quality issues and, eventually, a shift to proactive action to stem these problems before they occur. An environment is created in which technical and business data issues can be resolved and prevented in a collaborative, efficient, and transparent fashion. This coordination should extend beyond the compartmentalized program areas and across the business/technology divide, throughout the education agency and even to school districts and other organizations such as postsecondary institutions, the labor department, social services, and others with valuable data related to student histories and outcomes.

Benefits of Data Governance

In an education agency, a data governance initiative typically aims to improve at least three major areas: organizational coordination, data quality, and data use. Figure 2 presents the benefits strong data governance can provide. Operational improvements at the bottom of the chart lead upward to improve the agency’s data quality, which then facilitates more effective and widespread use of data to improve education.

Greater organizational coordination and collaboration

A more holistic, cooperative approach to handling data can be established through data governance. The decompartmentalization and coordination of enterprise-wide efforts can improve the culture of data collection, maintenance, use, and dissemination. More specifically, data governance accomplishes several goals.

key icon

Data governance goes beyond LDS

While data governance is a key factor in LDS success, it provides even broader benefits in terms of how an agency manages data, ensures data quality, and fosters effective use of those data. In effect, any data initiative will benefit from good data governance.


Establishing clear ownership and responsibilities

As the adage goes, when something is everyone’s responsibility, it is no one’s responsibility. Data governance assigns responsibility for each and every data element and deliverable to a single data steward, who becomes their “owner.” The roles of program area staff are specifically laid out to avoid confusion, and to ensure the necessary work is completed and only one person is accountable for a particular data problem. For example, if discipline data are requested, everyone should know which program area staff member is the discipline data steward and this person, therefore, should respond to the request regardless of who was initially called. In turn, if an issue arises with the discipline data in a legislative report, or on the annual report card, that same staff member is held accountable and is responsible for resolving it. In other words, data stewards are responsible for every aspect of the data they own, from collection to reporting to communication and so on. Thus, data governance helps agencies maintain an orderly operation in which every job is defined and every job is done. Again, responsibility for the data should be seated with program area staff, not the technology staff. The role of IT should be to support the agency’s business needs. Responsibility for the contents of those systems—the data—should rest firmly with program area staff.

Key images
IT staff “own” the infrastructure. Program area staff “own” the data.

Reducing or eliminating redundant efforts

With data governance, staff work to identify and eliminate collection redundancies wherever possible. As a result, data elements are shared by all appropriate program areas, but collected only once by a single area rather than multiple times within or across departments. Each element used for federal or state reporting or dissemination has a single authoritative source. Data elements are collected at the individual-record level, rather than in aggregate form, and all aggregate collections end. Ownership of, and responsibility for, all deliverables are clearly documented and communicated, avoiding duplication among multiple staff members. These activities may dramatically reduce the reporting burden of school districts that must report the data, as well as processing time for state staff. Ultimately, data governance will help the agency realize the “collect once, use many times” ideal, improving efficiency and effectiveness as well as the quality of the actual data.

Key images
Data governance helps agencies achieve the "collect once, use many times" ideal.

Facilitating more frequent, broader, and better quality communication and collaboration

Data governance forges lines of communication among a variety of stakeholders. The process is a mechanism for consistent transmission of expectations across program areas within the department, as well as externally with districts and other agencies and partners. Standards for data reporting and collection, and changes to those standards, are clearly broadcast. These include definitions, formats, business rules, responsibilities, and due dates. Beyond facilitating more effective communication, data governance brings affected stakeholders together to collaboratively plan their work and address data-related issues. Program area staff, IT, leadership, districts, and other relevant parties make decisions collectively rather than, for instance, the state agency staff making the decision and simply telling the districts what was decided. Districts and all other affected stakeholders are invited to weigh in on data issues that affect them, and to help create smarter solutions. They are involved in determining new requirements and making sure these can be met, and they also help find ways to make the data more relevant at the local level—a key to ensuring data quality. In short, with data governance, all decisions are made collaboratively and communicated effectively.


Standardizing business processes over time

Data governance brings staff together to define enterprise-wide standards for each data element, and to normalize the procedures for data reporting and collection. Clearly defined, documented, and well-communicated policies and processes let everyone know what needs to happen, by whom, how, and when. These processes are documented and strictly followed, and each one has a single "owner." For example, while calculating the National Governors Association graduation rate* may involve multiple staff members, only one data steward should be responsible for ensuring that the rate is calculated on time, properly, and in a consistent manner. Guided by clear data management protocols (regarding collection, reporting, sharing, etc.) that are consistent within and across program areas, staff members no longer operate in their own fashion within their own "silos," and the same tasks are no longer performed differently by different people. Everyone, including districts, knows what to expect and what is expected each cycle. Documentation of these processes also helps to ensure sustainability over time despite staff turnover, as well as to increase transparency by detailing processes and clarifying the origins of the data.


dictionary

LDS Lore: Data quality from management simplicity

More and more work was being done, but the quality of the state agency’s data wasn’t improving. In fact, the staff’s lack of coordination was just making things worse. Cedric and Mark worked together to compile the data for a federal report, running some quality assurance tests on the district data, aggregating them, and building tables. Meanwhile, Amy, the staffer who worked on the prior year’s report, had some free time on her hands. Poking around the agency system, she found the new file and took a look. Some of the numbers seemed off, so she ran some code. Her results seemed better than the ones in the file, so she pasted them in…

Months later, districts started calling to say that some of the numbers in the file were off. Mark and Cedric reviewed and re-ran their code to identify the problem, but they couldn’t replicate the reported numbers. No one could figure out what had happened. Eventually, word got around and Amy came forward and admitted what she had done.

The message was clear: Without the assignment of clear roles and responsibilities, and the adherence to clear processes, confusion will result, work will be duplicated, errors will be difficult to trace and resolve, and time will be wasted. In this example, less is more when it comes to data quality, and quantity of work does not necessarily translate into quality of data.


Shifting operations from reactive to proactive mode

Naturally, the data governance process will begin as a reactive course of action that deals with fixing problems that already exist. But, as the pressing problems are solved, the focus will eventually shift to identifying areas that can be improved, and to preventing mistakes from arising in the first place. For instance, if a state report card lists two discrepant counts pulled from two different silos, the agency’s reactive response should be to bring the two data stewards together to figure out if there is a reason to use both sources. If not, the next step is to determine which system is the authoritative source for the requested information. A proactive process, on the other hand, would have identified the redundancy before report card season, when everyone worked together to catalog all of the agency’s collections and data elements.


Enhancing understanding of the organization’s data assets

Through the governance process, staff become more aware of the data the agency collects, and learn which data stewards are in charge of which elements. At a higher level, data governance may help spur a culture shift from viewing data as compliance-driven to viewing them as assets that can help improve their work and student performance. With student-level longitudinal data, agencies can do more than fulfill compliance requirements—they can help improve programs and policies. The data governance process helps stakeholders shape the system to better meet their needs and to expand their capabilities.


Higher quality data

Implementing a longitudinal data system does not in itself ensure higher data quality. However, it provides an opportunity to improve data quality by bringing errors and inconsistencies to light through the enterprise-wide integration of disparate silos. Without a systematic approach to governing data, the organization has no means of addressing these issues. As a result, the masses of information collected and maintained in the LDS will be questionable and may not meet stakeholder needs. Consequently, many agencies see improving data quality as the primary reason to focus on governance. More specifically, effective data governance will help in several ways.


Improving accuracy and reliability

Data governance serves to increase alignment among program areas, ensuring consistency in data and related management procedures. Data are more reliable when data managers come together to lay out clear definitions and other standards, and data stewards work to identify and correct any deviations from those standards. In addition, thorough and consistent validation procedures starting at the local level ensure data accuracy; and authoritative data sources are identified and redundancies eliminated, creating a single "truth." These processes ensure that the agency collects data elements only once, streamlining data reporting and making analyses more consistent.


Key images
Data governance is about people and policies. Technology supports the process but should never drive it.

Increasing data usefulness

Data will be more useful when they are aligned with the needs of program areas and other stakeholders, rather than driven by information technology. Once the data governance process helps staff identify, manage, and control all the data, and those data become trusted and reliable, then another level of data governance may manage the use of that information.


Providing timelier access

Data governance leads to timelier information by increasing the efficiency of data collecting and reporting. Furthermore, districts receive ample notice of changes and, thus, have more time to prepare and work out potential issues. Standard business processes at the state level may also save significant time. Streamlined data-sharing procedures, for instance, can improve how quickly requests are processed, giving users faster access to the information they need. Data governance also serves to standardize records processing from year to year, helping to eliminate time wasted figuring out who will do what and preventing the need to reinvent the wheel every collection cycle.


Improving security

Within a strong data governance system, staff from program areas and IT work together to determine the sensitivity of each data item and implement effective protections. Clear and consistent data-sharing processes streamline and coordinate agency efforts and help prevent improper release of sensitive data (see chapter 7 and chapter 8).


Increased use of data to improve education

While many see high-quality data as the primary goal of data governance, the ultimate benefit may be their increased use by legislators, administrators, and educators to improve education. For data to be used, they must be accurate and trusted, timely, and designed to meet stakeholder needs. Data governance helps agencies realize these goals by coordinating staff from across the enterprise to collectively solve data issues. Clear data ownership ensures that the right staff members are disseminating and answering questions, and better communication increases all users’ understanding of the information collected by the agency. Better quality data equip decisionmakers to improve resource allocation choices, and better student data enable educators to enhance their students’ instructional needs. In addition, data governance can culminate in smoother operations and better decisionmaking based on the foundation of timely, secure, high-quality data aligned with agency goals.

These benefits should far outweigh the costs of implementing data governance. If, in fact, your agency has the in-house capacity to implement a data governance process, the cost of a data governance program can be as low as zero. While some state or local agencies will decide to hire an outside consultant to drive the process, others will need to invest only their time. Though changing the way the organization operates and improving data quality will take time, persistence will eventually lead to greater efficiency; time and resource savings; better programs and policies; higher quality information; and, ultimately, better student outcomes.

Top


* The National Governors Association (NGA) graduation rate is a standard, four-year adjusted cohort graduation rate agreed upon by all 50 governors in 2005.

Would you like to help us improve our products and website by taking a short survey?

YES, I would like to take the survey

or

No Thanks

The survey consists of a few short questions and takes less than one minute to complete.