New high school graduation rates are out

Kentucky actually shows slight improvement using credible formula

But, Kentucky’s current, non-credible formula shows decline

If the headlines confused you, don’t feel too bad.

Computing credible trend lines for high school graduation rates and dropout rates has been an issue for decades. I have been complaining about the problem for about 15 years.

The official high school graduation rate formula currently used by the Kentucky Department of Education in their new nonacademic data report was shown to be notably inflated back in 2006 in a major federal study. That study urged all states to adopt a new formula called the Averaged Freshman Graduation Rate (AFGR), which is explained here.

Sadly, Kentucky won’t finally come on board with the AFGR until next year, making it one of the last states to adopt a more credible graduation rate formula.

While we still must wait for better graduation rate data from the KDE, federal research shows an old formula called the “Freshman Graduation Rate” works far better than the “Leaver” rate the KDE is still using. Furthermore, the Freshman Graduation Rate can consistently be applied to Kentucky’s available data way back into the 1980s, as I have done in this graph.


By the way, I also ran an approximate calculation of the AFGR for 2009 using an average of 8th, 9th and 10th grade enrollment for the denominator and the total of NCLB creditable diplomas (those regular or governor’s diplomas earned within four years plus those regular diplomas earned by learning-disabled students within five years).

That AFGR for 2009 was 76.7 percent. This is still slightly high because I didn’t make another necessary correction for the number of disabled students who are not assigned to a specific grade. The federal formula actually requires that correction, which I estimate would drop the “on-time” AFGR to about 75.7 percent, within two points of the 2009 Freshman Graduation Rate shown in the graph above - and FAR lower than the 83.91 percent rate that the KDE just officially claimed.

Anyway, the good news is that using the fairly accurate Freshman Graduation Rate, our high school graduation rate has finally risen above the previous peak set back in 1993, shortly after KERA was enacted.

The bad news is that we are still losing about one in four students prior to graduation. After 20 years of very expensive education reform efforts, that is simply unacceptable.



The data I used to assemble the graph includes the ninth grade fall membership counts from annually produced reports (for older years, the SD-125R paper reports, for newer years, the Growth Factor/Ethnic Membership Reports, which can be accessed on line from here).

The graduation counts for older years came from old paper records at the KDE which I researched many years ago. More recent graduation counts used in this graph come from the graduation counts in the transition to adult life data from the KDE. Data back to 1993 is available on Page 15 in the new Nonacademic Data Briefing Packet that was released by the KDE on June 1, 2010.

One last note: You can find much rosier numbers over at the Prichard Committee’s web site. One reason those numbers look so good is that Prichard consistently fails to acknowledge the transfer of about 1,500 students from private eighth grade schools to the public system in ninth grade. Without a correction for that issue, Prichard’s use of eighth grade public school only enrollment in its formula results in somewhat inflated numbers.

Prichard’s calculation also includes diplomas awarded to non-learning-disabled students in over four years. No Child Left Behind does not permit that, probably reasoning that if students don’t have disabilities, a good high school program should enable them to graduate on time at lower expense to the taxpayer.

For unknown reasons, Prichard has consistently refused to acknowledge the federal research that has been done in this area and the research-based recommendation to switch to the AFGR pending a state’s developing a truly accurate student tracking system that is essential to enabling truly high quality graduation rate reporting.

Prichard does get some things right. Like us, they are upset that the KDE has never released disaggregated graduation rate information at the district and school level. That makes it harder to estimate graduation rates for the minorities, a data failure that simply is inexcusable. On that, both Prichard and the Bluegrass Institute agree.

education