Feature                                                       Pages 16-21


Data-Savvy School Systems    

Five districts where collecting sound data on students was just the starting point for improving instruction and raising academics


A team of 4th-grade teachers in California’s Sacramento City Unified School District collaboratively examines student writing products using a rubric that outlines elements in high-quality persuasive writing. The teachers analyze how they taught and assigned the written work in relationship to what the students produced. They focus on what a few of the teachers did differently to yield high-quality papers.

Then the teachers develop a short reteach lesson to take students through the task of revising their writing.

This meeting of the 4th-grade team is part of an instructional “problem of practice” identified by teachers based on weaknesses they discovered through three data sources — formative and summative tests and student work.

While school districts are improving their use of data to identify academic intervention needs, few are able to connect data to teaching and learning similar to the teachers in Sacramento. One common element found across high-improvement school districts is their ability to leverage the right information to better all decision-making processes. Of course, data alone cannot spark improvement. The keys to their success are changes in practice and culture resulting in a system that wants, values, trusts and uses multiple sources of data to affect teaching and learning.

Through my research and work with more than a dozen high-improvement school districts, I’ve identified four essential elements of strong performance management systems that, leveraged in concert, can improve wholesale student instruction and move the needle in places where it has been chronically stuck. 

Heather Zavadsky
Heather Zavadsky

Basic Components
Performance management refers to the systemic use of all types of data and includes much more than just monitoring student achievement. It means obtaining information to improve district practices at all levels, instructional and operational. Many great resources describe the systems, tools and practices that work best to build an effective performance management system.

In brief, the basic components include:

COMPREHENSIVE DATA SYSTEMS. A user-friendly one-stop shop provides detailed information on the instructional and operational needs of students, teachers, schools and central office. The best data systems are easy to query and allow users to connect summative and formative assessment data and student record information, such as transcripts, absences and discipline data. They also connect to curricula or standards, with some that even provide sample lesson plans.

CONTINUOUS MONITORING SYSTEMS AND TOOLS. Good formative assessments provide at least quarterly diagnostic information on what was taught. Additionally, structured walk-through processes conducted by a team with rubrics targeting specific instructional objectives or approaches help monitor curriculum implementation and support needs.

STRUCTURED TIME AND TOOLS TO EXAMINE AND DISCUSS DATA. Monitoring progress of stated goals and strategies in the district strategic plan and using scorecards and/or dashboards to examine multiple data sources with one tool helps structure and organize progress monitoring. Providing scheduled time and specific processes to review and discuss data at all system levels also must be in place.

CULTURE OF DATA TRUST AND TRANSPARENCY. Data are not useful without a culture that trusts, desires, shares and displays data. This culture must be nurtured at all levels and reflected in conversations and actions. Data should be used as a flashlight, not a hammer.

The key to success is using all the elements together to raise the quality of instruction for students. Having a great data system but no time to review and discuss data likely will not work. All the pieces need to be in place and used across the entire system. Even actions that seem unrelated to instruction, such as creating bus schedules, allocating resources and organizing central-office structures, have an impact on the classroom.

Yielding Proof
Several school districts have pulled together these elements effectively and incorporated them across the central-office, school-leadership and classroom levels to improve teaching and learning. The proof is in their outcome data — they all have yielded increased student achievement, even in chronically low-performing turnaround schools.

IMPROVING SCHOOLS THROUGH AN ACCOUNTABILITY FRAMEWORK. In 2009, Denver Public Schools launched an innovative improvement/turnaround strategy that yielded systemwide achievement increases, including the movement of six turnaround schools out of the state’s lowest performance category (“Does Not Meet Standards”). An important driver of Denver’s success was the district’s creation and use of the School Performance Framework, which pulls together multiple data sources to assess the organizational health and effectiveness of each school.


Four Reasons Data are Used Inconsistently in Classrooms (Zavadsky)

Denver’s leaders developed the framework to gain more nuanced information about school performance beyond what they received from the state to better inform their intervention process for improving chronically struggling schools. The framework has undergone several improvements. In 2011-12, the indicators used for each school included student progress over time; student achievement (percentage proficient or above); postsecondary readiness (on-time graduation rates); student engagement and satisfaction (attendance and satisfaction survey data); re-enrollment; and parent satisfaction.

The School Performance Framework variables are rolled up into color-coded ratings that range from “Distinguished” (blue) to “Accredited on Probation” (red). A red rating means the school needs some type of support and triggers an external diagnostic review and then intervention planning by a committee that examines the framework and reviews diagnostic data to identify problems and solutions.

Color-coded maps are posted everywhere in the district so that school performance within the five-level coding system can be grasped in a glance. The result of this system is a unified and clear definition of the established performance bands and desired targets, performance transparency and the understanding that multiple data points should be examined together to fully understand school performance.

MANAGING PERFORMANCE THROUGH A STATE-LEVEL PILOT. The Pharr-San Juan-Alamo Independent School District is a small, rural district at the southernmost tip of Texas in the Rio Grande Valley. Despite high rates of poverty and a large English as a second language population, the 31,000-student district has dramatically increased its graduation rate, reduced the dropout rate, raised success in advanced courses and increased the number of students who are enrolled dually in college and high school by carefully tracking those data elements from their current system.

Pharr-San Juan-Alamo fully takes advantage of opportunities to strengthen its systems and practices. Taking data use to the next level, the district became a participant in the Texas Student Data System pilot in 2011-12. The pilot data system consists of dashboards designed in Texas by educators and experts to provide timely actionable student data to appropriate stakeholders at the district, campus, classroom and student levels. Sample data include attendance and discipline information; formative and summative assessments and grades; student potential for advanced courses and college readiness; and student transcript data. Negative trends are highlighted by red colors and flags.

Although Pharr-San Juan-Alamo is just in its second year of implementation, the district has found the dashboards to be effective for obtaining weekly information on student and teacher progress, which has helped them focus more on prevention rather than intervention. With the dashboards, the district can determine if a student is experiencing a short-term performance blip or something more extensive that educators will address through an individualized intervention plan. The district’s leaders believe the system enables them to look at multiple triggers that can affect students in one tool and on a much timelier basis than previously possible.

USING DATA TO CREATE AND MONITOR LARGE-SCALE INTERVENTIONS. Long Beach, Calif., Unified School District is known for long-term stability in leadership and performance. Already quite data savvy, the 84,000-student district continues to improve its data systems and data use to push those schools that perform below the district average. When Long Beach leaders found a uniform performance drop between 5th and 6th grades in its eight targeted, low-performing K-8 schools, district leaders moved all 6th graders from those schools (approximately 2,000 kids) into a self-contained 6th-grade academy.

This unique approach put the students under one roof and provided flexibility to address student needs all day long. With no bells or defined class periods, students could spend a half day or full day in reading with a literacy specialist if needed.

Using data to refine student interventions, academy leaders used formative assessment results to match student needs with teacher strengths. If a student were struggling with fractions, he or she was paired with a teacher whose classroom work yielded high assessment scores in fractions. After a few weeks, students were reassessed and regrouped as needed.

In 2010-11, the math proficiency rate of 6th graders in Long Beach’s target schools moved from 32 percent to 46 percent, just a few percentage points below the district and state averages. With the success of this intervention, the district’s leaders are considering which aspects of the model might be scaled to other schools.

USING DATA TO CHANGE INSTRUCTION. Many school districts now use formative assessments to better target instruction and interventions. However, few get down to the level of analyzing student work as does Sacramento, Calif., City Unified School District. In addition to creating strong data and accountability systems and practices, the 44,000-student district adopted a data inquiry model similar to the Data Wise process developed at Harvard by Richard Murnane, Elizabeth City and Kathryn Parker Boudett.

Sacramento’s data inquiry model incorporates the four essential performance management elements referenced above but differs through the addition of several follow-up steps. In this process, teachers and principals are trained to use growth data at all system levels to identify instructional weaknesses in six chronically low-performing schools called Priority Schools.

Sacramento leaders and teachers connect the growth data to instruction by creating a data overview that combines various data sources; identifying problem areas; examining student work in collaborative groups; and comparing student work to what teachers identified as “high-quality” products. Discussions between instructional team members focus on distilling instructional assignments into a discrete set of tasks for students. The result is a more concise understanding of how to design, assign and teach instructional tasks, increased expectations for student work and a common understanding of what high-quality teaching and learning look like.

The change in instruction is easily seen in notebooks that document improvements in student work and is evidenced in the state assessment results of Sacramento’s Priority Schools. During the first implementation year, all six Priority Schools made gains in composite math and English language scores. Now the district is focusing on scaling the practice districtwide.

USING DATA FOR COURSE-PLACEMENT INTERVENTION. The Garden Grove, Calif., Unified School District, the 2004 recipient of the Broad Prize for Urban Education, has shown steady performance increases for several years. Still, district leaders continue to examine all types of data.

For the past few years, the 48,500-student district, which is located near Anaheim, has focused on better preparing students for college and skilled careers. In addition to raising instructional rigor throughout its preK-12 curriculum, the district also focused on correct course placement for students, a process that began by eliminating many remedial courses and creating bridge-building supports to help students succeed with more difficult work.

Data has played an important role. It was used to identify the overall problem and potential solutions and to help the district better match student academic needs to courses. One approach was to have counselors and teachers more closely examine formative assessment results and daily grades to determine which students could succeed in more advanced courses with supports in place.

“It’s not just the quiet kids that sit in the front row that should be eligible for those courses,” stated one Garden Grove leader. “Counselors and teachers need to use multiple sources of evidence to identify students that, with some supports, can succeed in advanced courses.”

During that process, discrepancies were found between course grades and formative assessments, making it clear grades often do not reflect achievement. Subsequently, the district invested significant time unpacking the issues and setting more uniform grading standards. The result was a better match between the two data sources and more reliable course placement data.

In addition to placing students in courses, the district also uses flexible grouping to match student needs to teacher strengths, similar to Long Beach. For example, using the district’s Schoolwide Intervention Model, struggling readers are grouped by reading level for intensive two-hour instructional blocks. Students are assessed frequently, and instruction is adjusted accordingly.

These two of many data uses employed by Garden Grove have contributed to the district’s steady improvement under the California assessment system.

Multiple Sources
Data-based decision making is effective when applied systemically using the essential elements of performance management. Teaching and learning cannot improve without multiple data sources to better explain the needs of students, teachers, schools and districts. Once those data sources are acquired, it is important to have user-friendly tools and time to examine data and devise appropriate and timely interventions.

Experience and lessons learned from others can help uncover the many important uses of data systems, tools and practices. Denver, Garden Grove, Sacramento, Long Beach and Pharr-San Juan-Alamo represent data-savvy districts that continue to expand their ability to use data to have an impact on students systemwide.

Heather Zavadsky is an educational consultant in Austin, Texas, and the author of School Turnarounds: The Essential Role of Districts.
E-mail: zavadskyh@yahoo.com


Give your feedback

Share this article

Order this issue