Will the winds of change bring a growth model?

One of the top questions about the No Child Left Behind Act, both before and after its passage, was "why isn't adequate yearly progress based on growth rather than just on year-to-year snapshots?" Now, three years into the implementation of this sweeping federal education law, the winds of change are blowing in Washington, and the use of growth models in AYP is more and more possible.

Call it longitudinal progress, value-added or a growth model, the central issue is looking at student progress from year to year. A growth model would address many concerns school administrators and researchers have expressed about AYP, which uses a snapshot approach by comparing the number of students proficient on a state exam to a preset annual target.

This snapshot approach fails to account for differences between cohorts of students. Research indicates that cohort differences account for considerable variation from year to year. In addition, the current system serves more as a proxy for a school's demographics than as an indicator of academic performance. By comparing groups of students to a target, AYP also fails to provide diagnostic information about why a group of children might lag behind peers. Did students with limited English proficiency miss the target because of a lack of fluency or because most of the children happened to be poor and were hungry on exam day?

To close the achievement gap under NCLB by targeting resources to the schools and students in greatest need will require an accountability system that reflects those goals. The current AYP system does not differentiate between a school whose growth has stagnated but hovers above the target from a school that has been improving by leaps and bounds but remains below the target. Such a school might benefit from AYP's "safe harbor" provision, which allows a school that has decreased the number of nonproficient students by 10 percent to make AYP.

Yet even that measure compares different cohorts of students--for example, a decrease in the number of 3rd grade students scoring below proficient from one year to the next. This may reflect improvement in the school but is not a true measure of growth.

Various Advantages

While no one knows yet what sort of model might eventually become allowable for AYP, it is clear that holding schools accountable for growth of students would better identify which schools truly need the most help. Some projections show 80-95 percent of schools failing to make AYP by 2014. That would make it even more difficult to target scarce technical assistance dollars. But if the system differentiated between a school that was making steady growth and one that was not, resources could be better targeted and schools making headway would be less likely to drop successful initiatives in the face of federal sanctions.

Growth models also may create conditions where teachers can address individual student needs to reach proficiency. By focusing on the growth of individual students, teachers can use their own judgment, based on the more effective use of formative, classroom-based assessments. Such assessments allow teachers to adjust instruction continuously and thus better teach each child.

The conversation about growth models has been going on for as long as NCLB has existed, even before it was signed. During many debates during reauthorization, AASA was at the table calling for the use of a growth model in AYP. Once implementation began to take shape and the law's impact was felt, the debate intensified. Recently, a task force of the National Conference of State Legislatures and a policy statement signed by more than 50 education and human service organizations, including AASA, have called for growth models. Many state associations of school administrators have released their own statements calling for the use of growth models, and 19 of them signed on to AASA's recent letter to Congress calling for improvements to NCLB.

Even back in March 2004, 14 chief state school officers sent a letter to the U.S. Department of Education expressing the need to track student growth. Then-Education Secretary Rod Paige responded by stating that No Child Left Behind "must be given a chance to work." He added: "Regrettably, there are some who would prefer to weaken accountability standards, regardless of the children who will be left behind as a result. Let me be very clear, changing the law to satisfy the concerns of the system at the expense of children learning is misguided and wrong."

New Openness

About a year later, policy watchers and educators everywhere were somewhat surprised by the change in tone. Education Secretary Margaret Spellings expressed a new openness to the possibility of growth models, saying, "We are in the process of convening a working group to find appropriate ways that growth models ... might be used to measure academic achievement. We are open to suggestions on how we can better understand how a state can use a growth model to meet the guiding principles of NCLB."

As of early summer, the membership, operations and transparency of this working group remain foggy. However, superintendents can begin thinking now about what might need to happen in their own district should growth models become an option. To what extent are formative, as opposed to summative, assessments in use--the kind that can help teachers help individual students? It would be wise to keep the public up-to-date and informed about assessment strategies and while public opinion on statewide tests is mixed, progress and growth do resonate with parents and non-parents alike.

The winds of change are blowing. What will result is not clear. AASA will continue to monitor this activity and keep you posted on what's next for AYP.

Terri Duggon Schworlzbeck is a policy analyst at AASA. E-mail: [e-mail]