High school graduation and dropout rates have long been used as indicators of educational system productivity and effectiveness and of social and economic well-being.
While determining these rates may seem like a straightforward task, their calculation is in fact quite complicated. How does one count a student who leaves a regular high school but later obtains a General Educational Development (GED) credential? How does one count a student who spends most of his/her high school years at one school and then transfers to another? If the student graduates, which school should receive credit? If the student drops out, which school should take responsibility?
The Committee on Improved Measurement of High School Dropout and Completion Rates was asked to address these issues and to examine (1) the strengths, limitations, accuracy, and utility of the available dropout and completion measures; (2) the state of the art with respect to longitudinal data systems; and (3) ways that dropout and completion rates can be used to improve policy and practice.
THE RATES AND HOW THEY ARE CALCULATED
In their simplest sense, graduation rates reflect the percentage of students who earned a regular high school diploma, and dropout rates reflect the percentage of students who did not. However, the requirements for earning a regular diploma vary widely among states and districts, and there are multiple means of completing school besides earning a regular diploma after attending
high school for four years. Students may obtain a GED, a certificate of attendance, or another alternative type of diploma. They may take longer than the typical 4 years before completing high school and may transfer across schools or districts before graduating or dropping out. There are a variety of strategies for accounting for these factors in calculating dropout, completion, and graduation rates; different strategies will affect the appropriateness of the rate for a given purpose. Thus, decisions about how to handle these factors should be consistent with the purpose for calculating the rate.
For example, if the purpose is to describe the level of education of the population, what matters is people’s eventual level of education, not what kind of diploma they received or how long it took them to earn it. But if the purpose is to evaluate a school’s effectiveness in graduating students in 4 years, those factors are of critical importance. All methods for calculating the rates require decisions about who to include in the numerator and denominator of the rate and how to handle certain groups of students, such as those who receive a GED or who take longer than four years to graduate.
We recommend that analysts and users keep their purpose in mind when selecting from among the various kinds of rates and choose the indicator best suited to that purpose (Recommendation 4-11). To help users draw sound conclusions, analysts should document the limitations of the rate and the decisions that went into calculating it (Recommendations 3-1 and 3-2). When the limitations are made explicit, alternative rates can be calculated to verify any conclusions drawn from the statistic (Recommendations 3-3 and 3-4).
The most accurate rates are those based on longitudinal data that track students over the course of their schooling, and we recommend that dropout and completion rates be based on individual student-level data whenever possible. This will allow for the greatest flexibility and transparency with respect to how analysts handle methodological issues that arise in defining the numerator and denominator of the rates (Recommendation 4-2).
BUILDING DATA SYSTEMS
Calculating rates based on individual data requires that states have a system for tracking students over time. At a minimum, such a system needs unique student identifiers as well as complete information on students’ enrollment status throughout high school. However, a more comprehensive system would incorporate data elements that allow school systems to monitor students’ progress, identify students at risk of dropping out, and evaluate the effectiveness of programs to reduce dropping out. To perform these functions, data systems require detailed longitudinal data (Recommendation 6-1).
Recommendation 4-1 is the first recommendation in Chapter 4. Other recommendations are numbered accordingly.
Producing accurate rates requires that states and districts adopt procedures to ensure the quality of their data; we, therefore, recommend that all states and districts maintain written documentation of their processes, procedures, and results. The documentation should be updated annually and should include a process for adding elements or making changes to the system (Recommendation 6-2).
Because the quality of the data begins at the point when data are collected and entered into the system, it is important that training be provided for those who carry out these tasks. We recommend that all states and districts implement a system of extensive and on-going staff training that addresses procedures for collection, storage, analysis, and use of the data (Recommendation 6-3) and conduct regular audits to verify data quality (Recommendation 6-4).
HOW DATA SYSTEMS CAN IMPROVE POLICY AND PRACTICE
Improving graduation rates in this country requires more than simply reporting accurate rates. To truly improve outcomes for students, data systems need to incorporate information that enables early identification of at-risk students. Research suggests a number of factors associated with dropping out: frequent absences, failing grades in reading or math, poor behavior, being over age for grade, having a low grade 9 grade-point average (GPA), failing grade 9, or having a record of frequent transfers. These findings suggest that states and districts should build data systems that incorporate documented early indicators of the risk of dropping out. At the same time, they should also conduct their own studies to determine the factors associated with dropping out from their school systems. Once determined, measures of these factors should be incorporated into the data system so at-risk students can be identified in time to intervene (Recommendation 5-1).
Finally, the federal government should play an active role in this area by collecting data on these early indicators. These indicators should be collected by grade level and should include variables such as the number of students missing a month or more of school, average number of days absent, average number of course failures, number of students failing one course or more, mean GPA, and indicators of behavior problems. Collecting these data would allow for indications of progress toward graduation at the national level and enable comparative studies on early indicators of dropout across states and localities (Recommendation 7-4).
As educational accountability focuses increasingly on the successful completion of high school, appropriate, relevant, and understandable measures of high school dropout and completion are becoming more important as indicators of the functioning of schools and of students’ preparation for college and work. The findings and recommendations of this report are provided to guide the creation of such indicators at the local, state, and national levels.