Caps, Gowns, and Games

High School Graduates and NCLB

Read complete document: PDF

PrintPrint this page
Share on Facebook Share on Twitter Share on LinkedIn Share on Digg Share on Reddit
| Email this pageE-mail
Document date: May 09, 2003
Released online: May 09, 2003

No. 1 in Series, "Learning Curve: Facts and Perspectives"

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

The No Child Left Behind Act (NCLB), enacted in 2002, requires that public elementary and secondary school systems be held accountable for achieving high levels of educational proficiency for all students. Most of the public concern about the mandated accountability has focused on the extensive student testing required and the high-stakes sanctions for schools that consistently fail to meet performance benchmarks.

Largely overlooked in these debates has been the fact that states must also include at least one other indicator of academic performance. At the high school level it must be the percentage of students graduating on time with a regular diploma. Holding schools and districts accountable for improving test scores and graduation rates is intended to safeguard against gaming strategies in which lower-performing students are pushed out to raise a school system's average test scores.

Accountability with Flexibility

NCLB typifies the Bush administration's policymaking philosophy: set high performance-based federal standards and give local actors significant flexibility in developing strategies for meeting them. In the case of graduation rates the law allows considerable leeway in how they are defined and measured.

Flexibility can be a double-edged sword. Permitting states latitude facilitates local implementation of federal regulations. However, since each state can pursue a distinct approach to measuring graduation rates, it may become difficult to hold states to the same standard. Moreover, because the stakes for low performance are so high, a state may be tempted to select the graduation indicator that casts its performance in the most favorable light.

Three Views of the Graduation Rate

To gauge the impact this choice might have, Urban Institute researchers applied three formulas to the class of 2000: the NCES method from the U.S. Department of Education's National Center for Education Statistics, the Cumulative Promotion Index (CPI) developed by the Urban Institute, and an Adjusted Completion Ratio (ACR).1 They calculated graduation rates using the most comprehensive source of data on student enrollments, graduation figures, and dropout counts: the Department of Education's Common Core of Data (CCD).

The three methods were chosen for several reasons. First, each method adheres to NCLB provisions and presents an alternative relevant to state decisionmakers. Second, each uses nonlongitudinal aggregated data available in most states. NCES relies on dropout counts to estimate the percentage of students who leave school with a diploma. CPI and ACR employ enrollment and graduation counts. Third, because the NCES formula was created by the Department of Education's statistical agency, states may believe it carries an implicit endorsement for use in NCLB accountability. In fact, over one-third of the states have proposed incorporating the NCES approach in their accountability systems.2

Choice of Method Matters

What do these measures tell us? In the case of NCES, the results are strikingly limited. Using this method, graduation rates could not be determined for 27 states (including Washington, D.C.). The reason: most states did not collect information on dropouts or failed to do so according to CCD standards. Because most of the largest states are missing dropout data, only 38 percent of public high school students nationwide are in districts for which an NCES rate can be calculated.

By contrast, CPI and ACR rates are available for 49 and 48 states and Washington, D.C., respectively, and cover 90 percent of high school students. The two generate nearly identical estimates of those in the class of 2000 who graduated on time with regular diplomas: 66.6 percent for CPI and 65.1 percent for ACR.

Although CPI and ACR employ distinct formulas, they produce very similar results. In fact, their rates differ from each other by less than 3 percent on average. In addition, there is no systematic pattern or bias in their estimates. Rates using CPI are somewhat higher in about half the states, while ACR rates are higher in the other half. In every state for which an NCES value can be calculated, this rate exceeds both CPI and ACR by over 10 percent on average.

Policy Implications

The consequences of choosing one formula over another can be dramatic. Suppose that states are required to adopt an on-time graduation rate of 75 percent as a goal. Among the 24 states for which the three rates are available, 20 would meet this performance standard if they used NCES. But with ACR or CPI only 8 or 9 states, respectively, would meet this standard.

These three measures represent only a subset of the approaches states have proposed for their NCLB accountability systems. Accordingly, the practical impact of a state's choice of graduation rate is likely to be even greater than related here.

The flexibility built into NCLB recognizes the legitimate authority states possess over their public education systems. But if accountability under NCLB is to have any chance of improving the nation's schools, the federal government will need to strengthen its leadership role in the federal-state partnership so that flexibility does not degenerate into chaos.

The Department of Education, for instance, could spearhead a scientifically rigorous effort to establish quality standards for measuring high school graduation rates. Only methods that meet these standards would be approved for state accountability systems. The department could also require that states adopt procedures to publicly certify the completeness and accuracy of their graduation and dropout data, since a graduation rate will only be as good as the data used to calculate it.

Taking these two steps alone would go a long way toward ensuring that no matter where a child lives, he or she will not be left behind.

Further Details

Swanson, Christopher, and Duncan Chaplin. 2003. "Counting High School Graduates when Graduates Count: Measuring Graduation Rates under the High Stakes of NCLB." Urban Institute,

Greene, Jay P. 2002. "Public School Graduation Rates in the United States." Civic Report No. 31. Manhattan Institute for Policy Research.


1. For details on each formula, see Swanson and Chaplin 2003. The ACR measure used here is closely modeled after Greene 2002.

2. Based on state plans publicly available as of April 2003.

Public High School Graduation Rates, 2000
  NCES Method CPI Method ACR Method

National Average —% 66.6% 65.1%
Alabama 78.5 61.3 60.0
Alaska 78.9 59.3 59.9
Arkansas 79.3 69.2 73.6
California 68.3 63.9
Colorado 70.3 65.7
Connecticut 85.4 76.3 70.0
Delaware 81.0 67.0 62.3
District of Columbia 53.5 51.0
Florida 49.9 49.0
Georgia 69.5 53.5 50.8
Hawaii 62.3 62.1
Idaho 74.7 75.8
Illinois 73.9 73.5
Indiana 70.8 72.8
Iowa 88.8 77.6 81.8
Kansas 73.3 71.7
Kentucky 63.7 66.6
Louisiana 62.9 59.5 58.6
Maine 86.0 72.5 71.7
Maryland 72.7 67.0
Massachusetts 84.5 75.5 72.1
Michigan 74.0 71.7
Minnesota 81.8 79.5 78.0
Mississippi 75.4 59.2 57.9
Missouri 80.3 71.3 71.2
Montana 83.3 76.5 77.3
Nebraska 85.0 77.7 80.6
Nevada 69.1 55.2 60.9
New Hampshire 72.8 67.8
New Jersey 81.6
New Mexico 71.4 60.1 59.0
New York 60.2 60.3
North Carolina 60.3 55.6
North Dakota 90.4 79.7 83.9
Ohio 81.6 70.7 72.4
Oklahoma 67.3 68.9
Oregon 62.6 64.9
Pennsylvania 84.5 75.2 73.7
Rhode Island 81.2 72.6 66.9
South Carolina 48.4 52.7
South Dakota 78.0 80.4
Tennessee 48.6 54.5
Texas 62.9 59.8
Utah 80.8 79.4 79.6
Vermont 72.9 74.3
Virginia 77.5 71.1
Washington 62.3 62.7
West Virginia 82.7 70.2 80.0
Wisconsin 76.6 77.1
Wyoming 81.1 74.7 79.3

Source: Common Core of Data, 1996-2000, National Center for Education Statistics.
— Indicator is not calculated due to insufficient data.

Topics/Tags: | Education | Governing

Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

Disclaimer: The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.

Source: The Urban Institute, © 2012 |