State Education Data Systems and the Stimulus

The American Recovery and Reinvestment Act (ARRA) provided new or additional funds for several federal education programs. Title I and IDEA, programs that received more than $10 billion each in the stimulus legislation, have gotten significant coverage in the media. But several smaller programs have been mostly overlooked. Statewide Longitudinal Data Systems (SLDS), an Institute of Education Sciences (IES) competitive grant programs that was allocated $250 million in the stimulus, could play a major role in improving state data capabilities and meeting reform goals in the future.

SLDS was authorized by the Educational Technical Assistance Act of 2002 to aid states as they create and implement longitudinal data systems. Longitudinal data systems typically track individual student records, including demographics, achievement on state tests, and sometimes transcripts, over time and across schools and school districts. Strong systems can link students to teachers or include data on district finances such as per pupil spending on specific programs.

These data systems can improve a state's ability to organize and analyze student-level education data including achievement outcomes. Similarly, strong systems can help districts, schools, and teachers improve student learning using up-to-date information on student progress.

So far, 27 states have received three-year grants from SLDS ranging between $500,000 and $6 million. Specific state goals and purposes for the use of grant funds vary widely. Some states, such as Arkansas and Florida, already have extensive data systems up and running. These states use the grant funds for higher level functions like creating electronic student transcripts and sharing their experiences with other states.

Many states, however, are just beginning to build good systems including Minnesota, Indiana, and Wisconsin. These states plan to use funds for more basic purposes like creating unique teacher identifiers, build student demographic data sets, and standardize their data dictionaries (which define each data point in a system).

In fiscal year 2008, SLDS was appropriated $48.3 million. That amount increased significantly to $65 million in the recently passed fiscal year 2009 appropriations bill. Between the $65 million in regular 2009 appropriations and the $250 million from the stimulus bill (expected to be distributed in the fall of 2009), SLDS is expected to expand significantly. This could mean large grants to a few selected states or medium sized grants to as many as all 50 states and the District of Columbia.

Regardless of how the new SLDS funds are distributed, it is clear that they will play a significant role in achieving some of the school reform goals outlined by Secretary Duncan in his guidelines on use of stimulus funds. Specifically, the guidelines encourage states to build Pre-K through college and career data systems to "track progress and foster continuous improvement."

Additionally, the stimulus legislation pushes states to rethink teacher compensation including programs that link teacher pay to student achievement. These types of programs require extensive and sometimes complex longitudinal data systems that link student performance data (such as standardized test scores) to specific teachers while accounting for student demographics and teacher qualifications. Few states currently have this capability or the staff to implement such complex modeling. SLDS funds could be the extra support states need to make these programs possible.

Beyond the stimulus, the increase in SLDS funds could have an important effect on education reform moving forward. Longitudinal data systems are vital to the use of growth models in assessing student achievement. Although current NCLB accountability structures are based on achievement levels, not individual student growth, both the Secretary and education pundits on both sides of the aisle have supported a move to growth models. These models would base accountability scores on how much students improve over a year of instruction, rather than how many students meet a certain threshold of proficiency in a given year. Such a system would require a continuous data set containing a student's previous and current achievement - something that many state systems do not currently have.

It is clear that well designed and implemented longitudinal data systems are likely to play a major role in education reform moving forward. The additional stimulus funds made available for SLDS grants could bring our state data usage into the 21^st century and beyond. Once the systems are built, it is up to the states to figure out how to use the new data to significantly rethink teaching and learning.