How to incorporate mobile students, who enter schools/classrooms after the start of the school year, into educational performance evaluations remains to be a challenge. As mandated by the No Child Left Behind Act of 2001 (NCLB), all states currently require that a school is accountable only if the student has been enrolled in the school for a full academic year. This paper investigates the school response to this eligibility requirement using regression-discontinuity framework. I find that schools that face accountability pressure behave strategically in an attempt to boost the "assessed" student performance, creating significant achievement gaps between eligible and ineligible mobile students. The findings also suggest that these achievement gaps are primarily driven by the strategic classification of students by failing schools to alter the eligible test-taker pool. I propose an alternative approach to mobile students in educational performance evaluations that eliminates this undesired incentive, which ironically affects students whom accountability systems specifically aspire not to leave behind. Appended are: (1) Full Academic Year Definitions in 50 States and the District of Columbia as of 2009; and (2) Grading Formula in Florida: 2008-2009 School Year. (Contains 9 figures, 13 tables, and 15 footnotes.)