accomplish this. We therefore endorse the idea that accountability measures, such as NCLB, should require schools to report their graduation and dropout rates, both aggregated for the full population of students and disaggregated by race/ethnicity, gender, socioeconomic status, English language learner status, and disability status. We further think that schools and districts should report a number of different dropout and graduation rates and complementary indicators to provide a comprehensive and accurate understanding of the problem and their progress toward addressing it. Examples include graduation and completion rates for all entering ninth graders, regardless of whether they transfer, and grade 9 promotion rates.
Although public airing of rates brings attention to the problem, more is needed to enact the kinds of changes that need to occur. We think that schools should also be required to set progress goals for improving their completion rates and should be held accountable for achieving those goals.
The formulas that are adopted for these rates should be structured so that students who transfer from one school to another are included in the graduation and/or dropout rates for at least one school. The methods of including students in the formula should be designed to minimize the potential for introducing bias in the rates as a result of systematic transfers into or out of a school (an issue taken up in detail in Chapter 3).
We therefore recommend:
RECOMMENDATION 2-1: Federal and state accountability policy should require schools and districts to report a number of types of dropout, graduation, and completion rates: for all students and for students grouped by race/ethnicity, gender, socioeconomic status, English language learner status, and disability status. Furthermore, accountability policy should require schools and districts to set and meet meaningful progress goals for improving their graduation and dropout rates. Rates that are used for accountability should be carefully structured and reported in ways that minimize bias resulting from student mobility and subgroup definitions.
In calculating graduation and dropout rates for disaggregated groups of students, it is important to remember that trends in the rates are susceptible to change as a result of definitional criteria (an issue described further in Chapter 3). That is, the rates may fluctuate simply as a result of the definitions used or policy about who is included in the subgroup. To the extent possible, criteria for inclusion in subgroups should be made uniform across states and districts. To help users interpret these rates, information on the number of students in the subgroups should be included along with the reported rates.
As alluded to in this chapter, there are multiple ways to report these rates and a variety of decisions to make about how they are calculated. The remaining chapters of this report describe these issues.