By Geoff Zimmerman
The high school graduation rate is a key outcome indicator that we track as a region to understand how many students are successfully completing high school. It is a key outcome for the Strive Partnership and it is one of the three education focused Bold Goals for our region coming out of this recent comprehensive community engagement effort led by United Way.
It’s also one of the most confusing due to the numerous ways in which high school graduation rates can be calculated. As a result, local education, policy, and business leaders are left wondering which graduation rate statistics are the most credible. This post is an update of a brief that I put together a couple years ago intended to provide an overview of the different types of graduation rate calculations, some of the challenges associated with each method, and steps being taken to improve our understanding of graduation and dropout rates. It is critical that we have an accurate picture of how many students are completing high school.
Overview of High School Graduation Rates
The high school graduation rate is a measure of student success that gets much attention in the media and the community. It intends to capture the number of high school students who stay in school and graduate on time. Graduation rates can be calculated a number of different ways. For example, the graduation rates reported by the Ohio Department of Education and the Kentucky Department of Education are much higher than the rates reported by an alternative source, as illustrated below. Below are the graduation rates reported by the state departments of education and the publication Education Week. The data are for the year 2008, the most recent year that data is available from Education Week.
The biggest questions this table raise are “which number is right?” and “how can the numbers be so far apart”? The reality is that neither of these calculation methods is perfect and the true high school graduation rate is likely somewhere in between the two.
State Methods of Calculating Graduation Rates
A common misconception is that school districts determine how their own graduation rate is calculated. The truth is that under the No Child Left Behind Act, the state determines the method and calculation. Until this year, the State of Ohio and the State of Kentucky required that public schools use a calculation method known as a “leaver rate” that relies on dropout data to calculate the graduation rate. This is the official graduation rate used on the report card for each school and school district, and it is used for the purposes of meeting No Child Left Behind Act requirements. It is defined as the percentage of students who entered high school that received a diploma four years later (including summer graduates and returning withdrawals). Although they both use the same rate, the way the denominator is calculated can vary slightly. This calculation method is reflected in the table above.
Disadvantage: This graduation rate likely underestimates the number of students who drop out. To code a student as a transfer (and not be counted as a dropout), Ohio requires a school to have a transcript request on file. However, the student may not actually enroll in the receiving district, even though the district has requested the transcript. In Kentucky, a student is coded as a transfer if the sending school gets a request for records from another school district. If the student does not actually enroll, he would not be counted as a dropout for either the sending or receiving school. Accurate tracking of student mobility is particularly challenging in urban areas, where students move frequently between schools and districts.
Moving Towards an Ideal Graduation Rate
The ideal method for calculating the graduation rate is to track cohorts of students from grade 9 (or earlier) to graduation and calculate the percent who graduate—a “four year adjusted cohort graduation rate.” In 2005, all 50 state governors made an unprecedented commitment to voluntarily implement such a common formula for calculating their states’ high school graduation rates by signing the National Governors’ Association (NGA) Graduation Counts Compact. On October 28, 2008, U.S. Education Secretary Margaret Spellings issued additional regulations under the No Child Left Behind Act. Under these regulations, all states are required to move to the same cohort rate, as described above, by the year 2011. The regulations also provide states an option to use an extended‐year rate for students who take more than four years to graduate as well as require reporting graduation rates by subgroup, reflecting achievement gaps of poor and minority students.
States that cannot meet the deadline can apply for an extension by providing evidence for why they cannot meet the deadline and a detailed plan and timeline for when it will be able to move to the new rate. The challenge with this lies in the ability to accurately track students. It requires states to implement better and more comprehensive data systems that track individual students over time across schools and school districts.
In Ohio: The state has been moving toward a four-year cohort rate over the last several years and just reported cohort data for the first time in August 2011. The state is publishing the new four-year rate this year, but the rate will not affect school report card ratings until next year, when it becomes the official graduation rate. The idea is to give school districts and the public time to get used to the new - often lower - graduation numbers.
There has been some skepticism and resistance. For example, according to the new on-time graduation rate, only 60 percent of Cincinnati Public's high school seniors graduated on time in 2009-10, the most recent year data is available. That rate is far lower than the district's official graduation rate of 81.9 percent. There are still challenges in coding and tracking students – especially in urban districts where students move in and out of schools often.
In Kentucky: Implementation of the cohort graduation rate formula in Kentucky is dependent on the full implementation of the “Infinite Campus” student support system, which will contain student tracking features. Based on the Infinite Campus implementation plan, all Kentucky school districts were online in March of 2009. Because it takes four full years of accurate student tracking data to generate the first report, Kentucky will not have its first high‐quality graduation rate data until after the completion of the 2012‐2013 school year.
In the meantime, the state of Kentucky will be reporting the Averaged Freshman Graduation Rate (AFGR) for three years before moving to the new cohort rate. The AFGR was developed by the U.S. Department of Education, and more information on it can be found below. This rate also isn’t without criticism. If a student transfers out of the state and graduates elsewhere, they are counted as a drop out by the formula. Additionally, if a student takes longer than four years to graduate, they count as a drop out, even if their delay was due to medical conditions or other outside circumstances.
Current Local Graduation Rate Data
Kentucky released their graduation rates in the beginning of August. A statewide map of current graduation rates by district is below. The entire dataset can be found on the Kentucky Department of Education website.
Ohio released their latest graduation rates with the school report cards in late August. Both the current calculation method and future adjusted cohort rate can be found on the report cards.
Alternative Calculation Methods and Definitions
The following lists some of the most well known alternative methods for calculating graduation rates. They generally use administrative data from the National Center for Education Statistics (including data sent from state education departments) and the U.S. Census Bureau. The alternatives fall into several broad catagories.
The Straight Diploma
• The Boston‐area researchers, including Walt Haney, Gary Orfield, and Jing Miao, use a straight diploma method. To determine what percentage of students graduate, you must know how many start high school. The straight diploma method simply counts the number of students in either eighth or ninth grade and compares that to how many students graduated.
Advantage: It is not reliant on school dropout data.
Disadvantage: This method doesn’t account for ninth grade retention and ignores student mobility. It is less accurate at the district level than the state level.
Who uses it: Other researchers have built on this method.
• John Robert Warren uses the eighth grade to diploma rate described above and adds in a migration/mortality correction (using Census bureau state population estimates by age).
Advantage: This is a better method than the Boston‐area researchers because it has a way to estimate the effect of student mobility.
Disadvantage: This method is also less accurate at the district level than the state level because of the need for some data for the migration/mortality adjustment, which is an estimate at best (Mortality is relatively low for teens and not a serious concern, but it gets mentioned for completeness.).
Who uses it: Other researchers have built on this method.
The Averaged Freshman Graduation Rate
• The U.S. Department of Education has an "average freshman graduation rate" that is similar to the straight diploma method described above except that it averages the eighth, ninth, and tenth grade enrollments to determine the starting point. For high schools that do not have an eighth grade, the formula averages ninth and tenth grade enrollment.
Advantage: This is an attempt to address the enrollment “bulge” that is a result of ninth grade retention. The grade nine bulge is due to restricting promotion from grade nine to grade ten because some students do not have enough credits to advance to their sophomore year, creating much larger ninth grade class.
Disadvantage: This method ignores student mobility and is less accurate at the district level than the state level.
Who uses it: Schott Report, State of Kentucky (temporarily)
• Jay Greene and Marcus Winters use the U.S. Department of Education averaged‐freshmen rate plus a migration/mortality adjustment that is almost identical to Warren's.
Advantage: It is not reliant on school dropout data.
Disadvantage: This method does not account for ninth grade retention and ignores student mobility. It is less accurate at the local level.
Who uses it: Manhattan Institute
The Cumulative Promotion Index
• Christopher Swanson's method is a prediction, based on two years of school data, of the ratio between the number of ninth graders and the number of students who graduate four years later.
Advantage: It is not reliant on school dropout data
Disadvantage: It is uncorrected for grade‐retention and migration/mortality issues.
Who uses it: Education Week in the Diplomas Count report, Urban Institute.
The Mishel and Roy Method
• Lawrence Mishel and Joydeep Roy created a relatively new and entirely different method than those above. It uses another U.S. Department of Education longitudinal dataset that follows student experiences and combines this with results of the Current Population Survey from the U.S. Census Bureau to estimate graduation rate. The result is much higher than the previous methods and closer to what the state methods calculate.
Advantages: It is not reliant on school dropout data or mobility of students.
Disadvantages: Critics of this method would say that this method is too dependent on the U.S. Department of Education longitudinal dataset which is a representative sample and the census surveys that depend on people telling the truth about their success in school. Furthermore, this study was really a one‐time event because the data used is from a nationwide sample that started with the eighth grade class of 1988 and has not been reproduced since then.
Who uses it: Economic Policy Institute
Are your comments limited to Kentucky and Ohio? Do you have thoughts/perpective on Indiana?
Posted by: Laura | September 22, 2011 at 09:38 AM
Laura,
Thanks for your question. I have not dug into the Indiana high school graduation rates as much as I have for Ohio and Kentucky. However, I can tell you that Indiana has been a leader in this efforts and was actually one of the first states to move to the new cohort graduation rate in 2005-06. More information on Indiana data and the history of how they have calculated their rate can be found at: http://www.doe.in.gov/gradrate/
-Geoff
Posted by: Geoff | September 26, 2011 at 06:49 AM