The Common Core of Data (CCD), administered by the National Center for Education Statistics (NCES), is an annual survey of the state-level education agencies in the 50 states, the District of Columbia, and seven other jurisdictions.1 Through this survey, statistical information is collected on all public school districts and their schools, staff, students, and finances. Information is not collected on private schools and their students, homeschoolers, individuals who never attended school in the United States, and those who have been out of a public school system for more than a year.
The dropout data collection was initiated with a set of instructions to state CCD coordinators in the summer of 1991. Those instructions specified the details of dropout data to be collected during the 1991–92 school year. Dropouts are reported for the preceding school year. The 1991–92 data were submitted to NCES as a component of the 1992–93 CCD data collection. The 2001–02 dropout data were submitted as a component of the 2002–03 CCD data collection. For the 2001–02 school year, a total of 49 states submitted dropout data to the CCD. Of these, 46 reported using agreed-upon reporting definitions. Those that did not were excluded from the CCD dropout data. Because of these exclusions, CCD data cannot be used to estimate a national-level dropout rate.
Data needed to estimate the averaged freshman graduation rate, specifically data on diploma awards and enrollment by grade, have traditionally been part of the CCD data collection. Like dropout data, diploma recipient reports are lagged a year (e.g., 2002–03 diploma counts are in the 2003–04 data files). All states reported diploma and enrollment data necessary for calculating the averaged freshman graduation rate, with the exception of diploma counts for 2003–04 for New York and Wisconsin.
The definition of “event dropout rates” that was agreed upon by NCES and the states was the following:
The denominator of the rate is the October 1st membership count for the state for the grades for which the dropout rate is being calculated. For example, the dropout rate for grades 9 through 12 would use a denominator that equals the October 1st enrollment count for grades 9 through 12.
The numerator (dropouts) is all individuals who
For the purpose of this definition
NCES is currently considering options for imputing missing dropout data. If implemented, the imputations may result in somewhat different estimates of dropout rates than presented in this report.
Data from the state nonfiscal CCD files are used to calculate averaged freshman graduation rates in this report. Graduates include only diploma recipients in this indicator. Other diploma recipients, such as those who earn a certificate of attendance, and those awarded high school equivalency credentials such as GEDs are not considered graduates. The purpose of these exclusions is to make the averaged freshman graduation rate as similar as possible conceptually to Adequate Yearly Progress provisions in the No Child Left Behind Act (NCLB) of 2001 (P.L. 107–110). These provisions require measurement of on-time graduation from public high schools, and explicitly exclude GEDs and other types of non-regular diplomas. Another reason for the exclusion of equivalency credentials in the averaged freshman graduation rate is that not all states report giving equivalency credentials, so comparable estimates across states would not be possible.
These are individuals who are awarded, in a given year, a high school diploma or a diploma that recognizes some higher level of academic achievement. They can be thought of as students who meet or exceed the coursework and performance standards for high school completion established by the state or other relevant authorities. State and local policies and data collection administration can have profound effects on the numbers of diploma recipients reported by a state. There are differences in what a high school diploma represents in different states. Some states award regular diplomas to all students who meet completion requirements, regardless of the extent to which these requirements address state or district academic standards. Other states award some form of alternative credential to students who meet some, but not all, requirements.
Exclusion of other high school completers
Other high school completers were excluded from the calculation of AFGR. These individuals receive a certificate of attendance or some other credential in lieu of a diploma. One example of such a credential is a certificate of attendance for special education students who do not address the regular academic curriculum. Students awarded this credential typically meet requirements that differ from those for a high school diploma. Some states do not issue an “other high school completion” type of certificate, but award all students who complete school a diploma regardless of what academic requirements the students have met.
Exclusion of high school equivalency recipients
High school equivalency recipients are awarded a credential certifying that they have met state or district requirements for high school completion by passing an examination or completing some other performance requirement. High school equivalency credentials, such as those earned by passing the GED test, are generally considered valid completion credentials, but recipients of such credentials are excluded from the averaged freshman graduation rate because No Child Left Behind called for only diploma recipients to be counted and because not all states report high school equivalency counts on the CCD.
Averaged freshman graduation rate
The averaged freshman graduation rate provides an estimate of the percentage of high school students who graduate on time. The rate uses aggregate student enrollment data to estimate the size of an incoming freshman class and aggregate counts of the number of diplomas awarded 4 years later. The incoming freshman class size is estimated by summing the enrollment in 8th grade in one year, 9th grade for the next year, and 10th grade for year after and then dividing by 3. The averaging is intended to account for higher grade retentions in the 9th grade. Although not as accurate as an on-time graduation rate computed from a cohort of students using student record data, this estimate of an on-time graduation rate can be computed with currently available data. The AFGR was selected from a number of alternative estimates that can be calculated using cross-sectional data based on a technical review and analysis of a set of alternative estimates (Seastrom et al. 2006b). The rate for the class of 2003–04 was calculated in the following manner:
Although enrollments are reported by grade, some states report ungraded students. To adjust for this, an allocation procedure used in the Common Core of Data “Local Education Agency Universe Survey Dropout and Completion Data” file was applied. Through this process the data for ungraded enrollment counts were redistributed across grades in proportion to the graded enrollment of the state, and the resulting estimates for grades 8, 9, and 10 were added to the reported enrollment counts for those grades. For the 2003–04 school year, the averaged freshman graduation rate for public schools in the United States for the 48 reporting states and the District of Columbia is based on the 2,548,128 diploma recipients reported for school year 2003–04, divided by the average of the 3,261,969 8th-grade student enrollment reported for October 1999–2000, the 3,669,077 9th-grade student enrollment reported for October 2000–2001, and the 3,259,701 10th-grade student enrollment reported for October 2001–02. The 2,548,128 public school diploma recipients divided by the 3,396,916 averaged number of public school freshmen, multiplied by 100, results in a 2003–04 public school graduation rate for the United States of 75.0 percent. The same formula is applied to compute state-level AFGR in 2003–04, and for national and state rates in 2001–02 and 2002–03. To produce an adjusted 2003–04 national rate using estimates for the two states missing diploma counts for that year (New York and Wisconsin), the 2002–03 AFGR rates for these states were used to estimate the number of diplomas received in 2003–04 in these states. More demographic information about students is available on the district-level nonfiscal CCD data files. However, the district level data are incomplete, so additional demographic information about graduates is not shown in this report. NCES is evaluating different options for imputing these missing data so that more detailed analyses by demographic characteristics can be undertaken. Once imputations are complete, state level totals from the imputed data may differ somewhat from rates based on the state level data shown here.
Note that the rate is not the same as a true cohort graduation rate that shows the proportion of actual first-time 9th-grade students who graduated within 4 years of starting 9th grade. A true cohort rate requires data that track a given set of students over time. The CCD data used for the averaged freshman graduation rate are collected using repeating cross-sectional surveys. Individual students are not followed from year to year. Although the averaged freshman graduation rate was selected as the best of the available alternatives, there are several factors that make it fall short of a true on-time graduation rate. First, the averaged freshman class is, at best, an approximation of the actual number of first-time freshmen. To the extent that the averaging differs from actual net transfers into and out of a class, and to the extent that it does not accurately capture grade retention and dropout rates across all 4 years of a given freshman class’s expected high school stay, the estimate will be wrong.
Second, by including all graduates in a specific year, the graduates may include students who repeated a grade in high school or completed high school early and, thus, are not on-time graduates in that year.
Taking these factors one at a time, it is possible that more high school students will move out of a given jurisdiction than move into it during the 4 years between the beginning of 9th grade and the expected graduation date. The averaged freshman count would overestimate the size of the actual cohort and thus underestimate the graduation rate. On the other hand, if more high school students moved into a jurisdiction than moved out during this 4-year period, the averaged freshman count would underestimate the size of the cohort and thus overestimate the graduation rate. Similarly, the use of 8th-, 9th-, and 10th-grade enrollment counts to estimate a first-time freshman class may not work as intended in many situations. Using 8th- and 9th-grade enrollment counts can be inaccurate to the extent that they do not adequately account for grade retention at 9th grade. Retention rates at 9th grade tend to be relatively large. While adding 8th-grade enrollments to the average may help diminish this problem, it is likely that in many cases it will not wholly adjust for actual 9th-grade retention rates, thus overestimating the first-time freshman count and underestimating the graduation rate. Using 9th- and 10th-grade enrollment numbers can be inaccurate to the extent that the 10th-grade counts exclude 9th-graders who dropped out from the previous year (effectively underestimating the cohort) or include students retained in 10th grade (effectively overestimating the cohort).
The inclusion of graduates who spent more or less than 4 years in high school increases the number of graduates in the numerator and yields a higher estimated rate than would be the case if only on-time graduates were included in the numerator. On the other hand, not recording early graduates with their actual cohort decreases the graduation rate for a class.
As a universe data collection, the CCD does not have sampling errors (the difference between an estimate based on a sample and the estimate based on an entire population). However, there are potential sources for nonsampling errors in universe data collections, including inability to get information about all cases (i.e., nonresponse), definitional difficulties, respondent inability to provide correct information, and errors made in recording, coding, and processing data.