differentiated accountability proposals

In response to Secretary Spellings’ March 2008 call for differentiated accountability proposals, 17 states submitted plans. The Department forwarded all 17 plans to a peer review committee, which will comment on them in mid-June before a final decision by the Secretary. Up to ten state plans may be approved. I've had a chance to review all 17 plans, and they vary in quality and method.

Ultimately, there is evidence within the applications to support both sides of the debate. Those who want to preserve NCLB as originally intended will see the lessened importance of disaggregation and specific subgroups as wrong and as a blatant disregard for the law as written. Others who want more state flexibility, a little more lenience, and the recognition that not all failing schools are equivalent will also find evidence to support their cause. Altogether, here are the lessons learned:

• At least four states (Georgia, Indiana, Louisiana, and Virginia) used this opportunity to ask to change the emphasis of Public School Choice (PSC) and Supplemental Educational Services (SES). In the case of Virginia, they repeated this request throughout the application. In fact, it appeared the state had no other plans for differentiation, as it did not introduce new categories or cut lines to demarcate them.
  

• Two states (South Carolina and Tennessee) acknowledged a change in the number of students who would be eligible for PSC and SES and provided an estimate for the change. South Carolina estimated a 7% decline in the number of students eligible for choice and a 22% increase in students eligible for SES. Tennessee’s numbers were dramatic. They estimated 47% fewer students (-29,333) would be eligible for choice and 73% fewer (-35,551) for SES. Other states declared the eligibility rules would not change in the interim, but they did not address how the differentiated accountability would affect the provisions in the future.

• Despite the Secretary indicating that priority would be given to states with at least 20% of their Title I schools identified as in need of improvement, only three of the 17 met this criteria. About half addressed why, while not meeting the 20% threshold, they should still be considered, but the remainder of states (6/17) ignored the requirement altogether.

• The methods of identifying the new categories varied in style and depth. Many simply divided schools and districts that previously did not meet Adequate Yearly Progress (AYP) by the percentage of Annual Measurable Objectives (AMOs) met. For example, South Carolina had three tiers. Schools and districts meeting 90-99% of their AYP objectives were slotted for “limited” intervention. Schools and districts meeting 60-89% would bear “targeted” intervention, and schools and districts below 60% would fall into “comprehensive” support. Twelve states featured this type of division of categories, albeit with different specifics.

• Two states (Illinois and Maryland) have created new categories depending on whether the school and/or district fail in the “all students” category or for single or multiple sub-groups. While Maryland featured a chart that justifies the validity behind this tactic, it also violates the central tenets of disaggregation in NCLB.

• Maryland also included a chart that broke down which schools and districts would be most affected by the changes. As we might expect, the changes give more flexibility to suburban schools that are largely failing because of only one subgroup, compared to urban schools that are struggling across the board.

Arkansas’ application states what many skeptics of the differentiated accountability plan suspect, saying, "In contrast to much of the more national rhetoric of schools missing performance goals for NCLB due to one or two subgroup, the Arkansas data models suggest that missing for one or two subgroups is a more isolated situation…A much more common occurrence is schools missing NCLB due to several subgroups in both reading and math."

New Jersey’s application contains the following alarming passage, which seems to suggest that, even if a school reaches restructuring, it would only have to do it for the failing category that’s to blame:

Our system [of differentiated accountability] does not result in schools with low performing subgroups such as Students with disabilities or English language learners remaining in the least intensive phase of intervention. A school that continues to miss AYP solely on the basis of the performance of one subgroup will continue to move through the phases and will ultimately reach Phase III: Restructuring. However, within this phase of improvement, the intervention will necessarily be focused on improving the subgroup’s performance.

At what point is this just a legitimization of under-the-table negotiations between states and districts over schools that REALLY fail and those that only sort of fail? Surely states already make distinctions at least in terms of oversight and aggressiveness; these plans just codify them.

Arguing for differentiated accountability would have been a lot more defensible if more applications included statements like this one:

PDE’s [Pennsylvania Department of Education’s] proposed differentiated accountability model will allow Pennsylvania to restructure our interventions so that communities with the fewest resources are receiving the most assistance.

Since these sentiments were rare, the cynic in me says these proposals are mostly about avoiding strong consequences for schools and districts with a lot of political power.