Statistical software analyzes student achievement growth

Texas’ Plano Independent School District is using performance management software to provide immediate student achievement results that help teachers provide targeted instruction tailored to individual needs.

Associate Superintendent Jim Hirsch said the district started looking at software several years ago because Plano “struggled with providing meaningful and actionable information on student achievement to teachers, parents, and the students themselves.”

In 2003-04, the district adopted the Measures of Academic Progress, an achievement measurement system produced by the Northwest Evaluation Association. This computer adaptive test took the place of all other nationally normed assessments and provided results to teachers within 24 hours.

“For the first time, Plano students in grades K-10 could participate in an online assessment environment that provided a common scale so individual student achievement growth could be measured and predicted in a statistically accurate model,” Hirsch said at a Monday workshop.

The district also used the Cognitive Ability Test (CogAT) from Riverside Publishing for students in grades 3, 5, 7, and 9. “In conjunction with our state-administered high-stakes standards test, our students’ projected achievement growth trajectory could now be deduced with a high degree of reliability,” he said.

Adaptive achievement measures, cognitive ability, and proficiency on state standards was modeled along with other important variables affecting student achievement, such as socioeconomic status, limited English proficiency status, and mobility. The goal was to develop a predictable learning model for each student.

However, Hirsch said, Plano found “the limitations of the existing web-based software data system were delaying the efforts of the district to move beyond autopsy-style results and enter into the arena of predictive modeling.”

After reviewing various options, the district selected SAS Institute’s Enterprise Intelligence Suite to provide teachers and principals with information on likely student performance in upcoming assessments.

SAS’ analytical and modeling tools allowed the district to design multivariate algorithms that were run against student assessment information in near real-time fashion.  That allowed educators to implement individualized interventions early in the school year, which could lead to better test scores on the state assessments administered in the spring.

According to Hirsch, Plano’s framework measures growth in a way that is fair to all students, regardless of ability; fair to all teachers, regardless of the diversity of students assigned to them; and fair to all campuses regardless of their concentrations of diverse populations.