Tracking students shows higher graduation rate
PUBLIC school teacher Thomas E. Stuart (Letters, June 15
) would profit from less effort crafting his polemics against the state Department of Education and more on researching the facts. Certainly the DOE has many problems, but deceiving itself about its graduation rates is not one of them.
In 2002, just after passage of the No Child Left Behind Act, I was involved in developing the DOE's measurement of the actual four-year graduation rate for Hawaii's public high school students. Contrary to the report published by Education Week, on which Stuart was commenting, we used actual tracking of individual students to calculate the percentage of entering ninth-grade students who graduated four years later. The rates we calculated then are consistent with those quoted by the DOE in the Star-Bulletin's June 12 story on the Education Week magazine report. About 80 percent of the entering ninth-graders graduate within four years.
THE DIFFERENCE between these data and the seemingly shocking figures from Education Week stem from Education Week's reliance on weak methods and assumptions. The journal does not have access to individual student data, as the DOE does, so it can't track students. Instead, it used a method similar to that used by Jay Greene, a researcher at the right-wing Manhattan Institute, who published a study in 2002, which reported a similarly low graduation rate for Hawaii -- in the range of 65 percent.
Greene compared the count of ninth-graders in one year to that of graduates four years later, assuming that the difference represented drop-outs. That is not so. First, the count of ninth-graders is inflated by hold-overs, students who failed courses or didn't complete enough credits the previous year to be promoted. Many of these catch up with their class before leaving school. There also are transfers -- students who move to private schools or schools in other states.
GREENE ASSUMED both these phenomena to be nil, but they aren't. The ninth-grade enrollment bulge is substantial. When every student is counted only once, in the class with which he or she entered ninth grade, and transfers are subtracted from the starting count, you get the figures that the DOE published, not those put out by Greene or Education Week.
Those are the facts. Can we now focus our attention on problems that are real, and not on manufactured ones?
Thomas Graham Gans is a retired evaluation specialist for the state Department of Education.