Does Wisconsin’s method inflate graduation rate?

Original URL:
From the Milwaukee Journal Sentinel
Does state’s method inflate graduation rate?
Wisconsin says 92% finish high school; report estimates 78% do

Posted: June 23, 2005
A new report lambastes states across the country for using flawed, and even “irrational,” methods of calculating graduation rates that ultimately dupe the public.
The report does not criticize Wisconsin as harshly as a few other states, such as North Carolina, but it does offer an alternative method of estimating graduation rates that would put Wisconsin’s rate at 78% for the 2000-’01 school year, 14 percentage points lower than the 92% rate reported for the 2002-’03 school year.
“Every year (states) report these literally preposterous numbers,” said Kati Haycock, director of the Education Trust, a Washington, D.C.-based organization that advocates for disadvantaged students and released the report.
The report suggests that Wisconsin and many other states measure graduation rates in a manner that gives an overly rosy, distorted picture of the number of students who are actually finishing high school in the United States.

“The states that are reporting inaccurate graduation-rate data are doing themselves a huge disservice,” wrote Daria Hall, the author of the report. “They’re depriving educators, policy-makers and advocates of crucial information necessary to create a sense of urgency for high school improvement. And they’re leaving educators vulnerable to accusations of dishonesty.”
The authors of the report acknowledge that comparing rates from two different school years is not ideal. But they add: “While a better match of years would of course be preferable, state-level graduation rates do not change so much from year to year that it would preclude this comparison.”
For the 2002-’03 school year, Wisconsin determined the graduation rate by taking the number of students who graduated and dividing it by the number of graduates added to the number of dropouts over four years. So if a school had 100 graduates and 25 dropouts, the graduation rate would be 80%.
“It’s an estimate, just like ours is an estimate,” said Tony Evers, deputy state superintendent for schools, referring to the 78% figure. He added that the state hopes to have a more reliable system – which would track individual students by assigning identification numbers – in place by fall.
“When authors use inflammatory language when none is really needed, I always question their motives,” Evers said, in response to the Education Trust report. “But there are things we can learn from this article. We absolutely do need to have an integrated, statewide data system and . . . once we have it in place, we will be in a better position to have much more complete data.”
Methods, numbers vary
The Education Trust used a method created by Christopher Swanson at the Urban Institute to reach the 78% estimate for Wisconsin. Swanson compares the number of 10th-graders in one year with the number of ninth-graders in the previous year to estimate the percentage of ninth-graders who were promoted.
He then makes the same estimate for the other high school grades, and multiplies the different ratios to arrive at an estimated graduation rate.
Much of the discrepancy between state results and Swanson’s results stems from which students are counted as graduates and dropouts. Wisconsin, for instance, in the past has counted many students who obtain GEDs as graduates, not dropouts. But Joe Donovan, a spokesman for the state’s Department of Public Instruction, said Wisconsin is moving toward a more stringent definition in which more students would be counted as dropouts.
Walter Secada, a professor of teaching and learning at the University of Miami who wrote a study on Hispanic graduation rates, said he “would probably trust the Urban Institute’s way (of estimating graduation rates) because, essentially, they have no ax to grind. They are a third party that is looking at it in a way that lets the chips fall where they may.”
“I don’t think states do things maliciously,” he added. “But it’s important to keep pressure on states to say, ‘You really need to improve the methods you are using.’ ”
Evers said the new tracking system in Wisconsin will allow the state to distinguish between students who transfer to different schools or states as opposed to those who drop out of school. “This is a critical issue, and what causes such variability in numbers,” he said.
The Education Trust report also criticizes states that have set low targets for improvement in graduation rates under the federal No Child Left Behind Act. The report points out that Wisconsin, as well as 33 other states, actually set goals that are lower than their own reported graduation rates for 2002-’03.
Further, the report notes that a majority of states, including Wisconsin, do not report graduation rates specifically for low-income students or those with disabilities.
Evers said attacking states for setting low progress goals for federal reporting misses the point.
At the local level, he said, people are well aware of the need to improve the graduation rate for Milwaukee Public Schools.
“To say they are not being held accountable is bogus,” Evers said.

One thought on “Does Wisconsin’s method inflate graduation rate?”

  1. This is really a non-issue. Getting huffy about which summary statistic to use takes us off what ought to be our concern—individual students. Here is the important point: What is the school doing to keep students learning?
    As a rule, the pattern of skipping class seems to start in ninth grade, and escalates to dropping out by or during the tenth grade. Generally, if a student makes it to junior year with a good attendence record (and no pregnancy), he or she is not going to drop out.
    Many ninth graders can’t handle the freedom of high school, especially open campus. They skip out at lunch, flunk classes, and end up losing credits. This puts them behind their peers, and they simply give up (at least for a time). Some drop back in, usually to an alternative program, if the school district has one, and end up graduating from that program or getting a GED. If you attend graduation ceremonies at one of these programs you will be so impressed by the kids that you won’t care what summary statistic on drop outs is used. I would choose one that recognizes these kids for struggling to overcome their problems.
    So, the real issue is what can be done to improve the chances that kids won’t drop out. Some schools close campus for freshman and sophomore years. Some front-load (before lunch) the tougher required classes (math, English, science) and back-load (after lunch) the fun classes and electives for freshmen and sophomores.
    Another way to handle this is to “core” freshman year for at-risk students, so students in a big “anonymous” high school have classes with the same group of students.

Comments are closed.