A. Improvement in Policy and Assessment
There is an emerging consensus that an evaluation of public education should include multiple measures, not simply test results. Proposed indicators of effectiveness and improvement include an increased percentage of students earning a high school diploma, reductions in chronic absenteeism and grade retention, and an increasing number of students taking and passing advanced-level courses. *397 The frequency of disciplinary exclusion, however, is often considered only as an indicator of school order and safety--as if student discipline had little connection to overarching educational goals.
Even outside the context of an administrative law challenge, the disparate impact analysis can help policymakers see that frequent disciplinary removal is not likely educationally justifiable and is likely to have a negative impact on minority students and their families. Moreover, if policies concerning the assessment of schools took into account analysis of disciplinary data like that discussed here, it could help strengthen our measures of school effectiveness and positively influence achievement.
It is also necessary to acknowledge that current discipline trends are not occurring in a vacuum. Federal policy currently provides an incentive for school leaders to remove low-achieving students from the cohort of students used to evaluate school performance. These lower achievers are more likely to be disruptive (Kelly, 2010). No Child Left Behind has imposed accountability measures for schools based primarily on student test results--but only for the test scores of students who attend a school for a full academic year (Elementary and Secondary Education Act 2002, Public Law 107-110). There is, in fact, research supporting the possibility that frequent suspensions are used to avoid accountability for the test scores of lower achievers (Figlio, 2003), and civil rights advocates have expressed concern that test-driven accountability for schools encourages frequent suspension for minor offenses-- that “push-out” low-achieving students, especially students of color (Advancement Project, 2010). This suggests that the disparate impact analysis should also be used to evaluate accountability policies and practices, not to mention resource distribution (Losen, 2004).
