Association for Public Policy Analysis & Management

The last decade has seen a great deal of research on the complex accountability policy rules of No Child Left Behind and their effects on student achievement. One of these rules, the participation rate mandate (i.e., that 95% of students in tested grades must be included in performance calculations), has received only limited attention. The participation rate rule is intended to ensure that 1) schools focus on all students and 2) schools cannot make performance appear higher by excluding the lowest-performing students from performance measures. However, only students who are labeled as full-academic year (FAY) enrollments within a school year are included in the 95% participation rate regulation. Each state determines the FAY cutoff date by which a student must enroll to “count” for accountability. As such, states that set restrictive FAY cutoffs (i.e., earlier dates) can make it easier for schools to exclude large proportions of mobile students, undermining the validity of school performance scores.

Mobile students are frequently considered an at-risk population, associated with lower achievement, increased disciplinary and absentee rates, and higher rates of dropping out of school. Large proportions of students are mobile each year (e.g., 34% of urban students in North Carolina move annually). As a result, the limited existing evidence suggests that as much as 22% of an urban district’s enrollment is excluded from accountability calculations because of mobility. Policy research has yet to investigate what proportion of students is excluded from school-level performance measures annually, who these excluded students are, and how the exclusion of mobile students influences school performance measures. Leveraging restricted-use Washington state administrative data as well as publicly available data from 2010 through 2014, this paper describes the ways existing school performance measures fail to account for mobile students, jeopardizing the inferences made about schools.

I investigate five possible approaches to including mobile students in school performance calculations. This work begins with recalculations of school proficiency rates for all public schools in Washington following all state-established policy rules. Using descriptive statistics, correlation analysis, and ordinary least squares regression models, I present how these five approaches to counting mobile students alter the proportions of students included in accountability measures, school performance outcomes, and the number and types of schools identified as low-performing. These analyses consider both overall enrollment and subgroup performance measures. Finally, I compare the schools identified as low-performing on the newly calculated proficiency rates to a school value-added measure to examine any unintended consequences of including mobile students in performance measures.

This research highlights a commonly ignored stipulation of school accountability policy, providing in-depth analysis of the relationship between the FAY cutoff regulation and subsequent school performance. Preliminary results suggest that more than half of mobile students in Washington, enrolled in tested grades, are excluded from school accountability measures annually. Excluded students are disproportionately concentrated in low-income and urban schools. The results of this work provide evidence to policymakers regarding the implications of future accountability system design.