Education Letter to the Editor

Graduation-Rate Tool and ‘Oversimplified Estimates’

June 11, 2007 1 min read

To the Editor:

In the description of the online graduation-rate mapping tool produced by the Editorial Projects in Education Research Center, a corporate sibling of Education Week, the claim is made that “for the first time, comparable, reliable data on graduation rates will be readily available for every school district in the country” (“New Graduation Rate Resource,” www.edweek.org, April 25, 2007). The Cumulative Promotion Index has serious flaws that make it anything but comparable or reliable, however.

First, it completely ignores the effect of student transfers and population changes in a district. For example, a student who transfers to another school and graduates from there is treated the same as a dropout for the school that the student transferred from, and a school that receives a lot of transfers could have a graduation rate of over 100 percent, regardless of its dropout rate.

Second, the Cumulative Promotion Index method itself doesn’t calculate what it claims to, which is “the probability that a student in the 9th grade will complete high school on time with a regular diploma.” To correctly make this calculation, you would need to follow the same student from 9th grade through graduation, but the index’s oversimplified estimate just compares current-year populations with prior-year populations across grade levels. Moreover, data taken from only two years are highly susceptible to year-to-year population changes that have nothing to do with graduation rates. For small schools and regions where many families move to new school districts every few years, this percentage can be many times larger than the dropout rate.

Finally, and most importantly for many school districts, the index doesn’t address demographic differences. Schools with a high percentage of limited-English-proficient students have a significant disadvantage in their overall on-time graduation rates when compared with schools that are almost entirely composed of students who have grown up speaking English.

I urge the EPE Research Center to present meaningful calculations of graduation rates instead of, or at least along with, its Cumulative Promotion Index.

Jason Sowalla

Manassas Park, Va.