Changes planned, but will test problems be solved?
NOTE: This is the third of three articles on Flunking the Test: A look at 'No Child Left Behind'.
Sara Craft had only one thought as she sat down to take the "last-chance" math proficiency test in early May to determine whether she graduates Thursday with her Xenia High School class of 2004.
"I'm going to pass, I'm going to pass," Craft said.
"My whole family is praying for me. My whole church is, too."
Craft's final obstacle to commencement was the math portion of the Ohio Ninth Grade Proficiency Test, the state's high school exit exam that state officials say is based on what students should know at the end of the eighth grade. About 2 percent of Ohio students — 2,200 last year — fail one or more portions of the exam throughout their high school years and are blocked by state law from getting their diplomas. Those 12th graders who haven't passed their two previous attempts during their senior year can take the exam a third time in May to see whether they can obtain a diploma during their school's commencement.
This year's ninth graders who share the high school hallways with seniors such as Craft face a much stiffer challenge that will almost certainly block larger proportions of Ohio students from graduating in 2007 and beyond.
The new Ohio Graduation Test is based on academic "content standards" and "benchmarks" adopted by the state and will be about two grade levels more difficult than the current high school exit exam. Next year's sophomores will take the Ohio Graduation Test for the first time in March 2005.
How hard is it?
The first time the OGT math test was given to sophomores in March 2003, more than three out of four students — 76.9 percent — flunked statewide. Nearly 95 percent of Ohio's black students failed the exam.
A member of an Ohio Department of Education review committee of teachers and school administrators told test authors that the 2003 math test was "more difficult than the ACT," a college admissions exam. Another committee member said the test challenged even his school's brightest students.
"Others gave up," the educator said in a report compiled by Measurement Inc., the company that produced the exam. The report did not identify the committee members.
This year's sophomores who took a revised graduation math test in March performed better.
Statewide, 32.1 percent of the students failed, based on the scores of a sample of 75,000 students — about half of those who took the exam, according to a summary released to state school board members May 10. Scores of the remaining students are still being compiled.
But the racial gap persisted: 61 percent of the black students who took the March math test failed, compared to 26 percent of whites, according to state records. The Ohio Graduation Test was given to sophomores last year and this year as a "dry run" to help Ohio comply with the federal No Child Left Behind act and to assist school districts in finding weaknesses in their curriculum.
Although the scores appeared to improve considerably, this year's sophomores took a scaled-back test — and were graded more leniently.
The Ohio Department of Education dropped four of 36 multiple-choice questions and one of two "extended-response" questions, in which students show their work and explain their answers. And if the state board in early June adopts the grading scale recommended by a committee of teachers, parents, administrators and business representatives, students will be able to pass the exam by earning 19 out of the possible 46 points, or 41 percent. The grading scale for the 2003 math test set the proposed passing score at 31 of the possible 54 points, or 57 percent. Some questions are worth multiple points.
The distribution of scores shows that the 2004 math test "was a fairer test," said Bob Bowers, the Ohio Department of Education's deputy superintendent. But Bowers also suggested the test may return to a higher level of difficulty.
"We'll introduce more rigor over time so we give the test higher complexity," the deputy superintendent said.
Student performance on a 2003 Ohio Graduation Test reading exam was much higher than the math scores, with 66 percent of students passing. The passage rate rose to 77.7 percent in the 2004 reading test.
One Ohio educator who helped give the math part of the 2003 Ohio Graduation Test at his school found the exam's academic expectations wildly unrealistic, and a testing expert described Ohio's system as immoral.
A member of the Ohio Department of Education technical advisory committee told test author Measurement Inc., that the math standards on which the new graduation test are based are similar to those adopted by the National Council of Teachers of Mathematics. The standards "are overly ambitious and seem designed to make every high school student a mathematician . . . ," according to a report prepared by the testing company.
The dismal 2003 math results also appear to contradict the generally upward trend of Ohio high school students on other math measures. Two years ago, a record 99 percent of the class of 2002 passed the math portion of the Ohio Ninth Grade Proficiency Test, and Ohio students who took the SAT college entrance exam improved their math scores by 14 points in the past 10 years, from 527 in 1993 to 541 in 2003. In addition, Ohio's performance on the National Assessment of Educational Progress 2003 math test shows that the state's eighth-graders outperformed their fellow eighth-graders in all five neighboring states (Indiana, Pennsylvania, Michigan, Kentucky and West Virginia) and improved their scores between 2000 and 2003, according to the Ohio Department of Education.
W. James Popham, professor emeritus at UCLA, former president of the American Educational Research Association, said the math test of the Ohio Graduation Test is based on too many academic content standards and benchmarks.
"There are just too many targets, and they overwhelm teachers," said Popham, who owned an education testing company that provided questions for former Ohio proficiency tests of the early 1990s. "It is almost immoral to throw 55 mathematics benchmarks into the air and ask teachers to hit them in any instructionally sensible manner . . . When you use a test such as this to deny diplomas to children, then you must toss out the 'almost' modifier. This is an unvarnished instance of testing immorality."
Mitchell Chester, the assistant state school superintendent who oversees Ohio's testing programs, suggested Popham has an anti-testing bias.
"You could predict what Jim Popham would say about anybody's test," the assistant state superintendent said.
Chester and his state education department colleagues vigorously defended the math exam. Ohio's deputy school superintendent, Bob Bowers, said a study of six states' high school exit tests by Washington, D.C.-based Achieve Inc. "will show our test is a fair test . . . . We're not the hardest and we're not the easiest." The math test in particular, Bowers said, is slightly easier than those developed by the other states in the study, which Achieve will release in early June.
Achieve Inc.'s Joseph Garcia, spokesman for the non-profit organization that promotes school accountability through standards and testing, said he could not discuss the study's results, but he described Bowers' comments as "an accurate characterization" of the organization's analysis of Ohio's graduation test.
Still, the 2003 math test's dismal passage rates did not come as a complete surprise to state education department officials, Chester said. Some students did not take the test seriously since it carried no penalties for doing poorly. And, Chester said, "We knew this test did not reflect the current math curriculum."
When the Ohio Ninth Grade Proficiency Test was first administered in 1990, only about 33 percent of students passed all portions of the exam the first time and thus met what would become the graduation requirement, said Jan Crandell, director of the Ohio education department's office of assessment. But by the time passage of the test became a requirement for graduation in 1994, the passage rate rose to 97 percent.
"That 33 percent was a wake-up call," Crandell said.
Education department officials are also treating the performance gap between black students and their classmates as a wake-up call, taking steps to narrow the "achievement gap" among ethnic groups and between rich and poor students.
The education department has launched a "Schools of Promise" program that identifies high-achieving schools — those with strong test scores despite high poverty levels, for example — and seeks to share strategies and the most promising practices with schools that are struggling, said Joseph Johnson, the Ohio Department of Education's special assistant to the superintendent.
"We need to show communities that this can be done, that mortals can do this," Johnson said.
Social consequences too great
The stakes are high, especially since Ohio students face the requirement of passing not just the math and reading graduation tests that just completed a two-year "trial run," but also three other brand new tests in writing, science and social studies. The students' diplomas — and their futures — hinge on the newly developed exams.
Some of the educators who reviewed the 2004 Ohio Graduation Test results for two education department oversight committees expressed concern over the significant changes the state was forced to make following the 2003 math test results, noting that there will be no similar opportunity to fine tune the social studies, writing and science tests. Those exams will be given for the first time next year to the sophomores who must pass the exams to graduate.
Bowers said state officials learned from the math exam experience, which he said "gives us two years of data and two years more experience in test development."
Educators hope Bowers is right — and worry about the impact if he's wrong.
"The state can't have the situation where huge numbers of kids are not able to get high school diplomas," said Thomas Lasley II, dean of the University of Dayton's School of Education and Allied Professions. "The social consequences are too great."
Lasley, along with UD's former president, Brother Raymond Fitz, helped write the final recommendations of the Governor's Commission on Student Success in late 2000 that led to an overhaul of Ohio's testing and school accountability programs and to the new graduation test.
The UD dean expects the Ohio Graduation Test math scores will improve once the exam carries the high stakes of denying diplomas and when school curricula are aligned to the state's standards. But dramatic gains often take time and require that students have had adequate exposure to the content before they are tested, he said.
Randy Overbeck, director of instructional services for the Xenia City Schools, said he is concerned about how students will react if they get a low score the first time they take the math part of the Ohio Graduation Test.
"If they see this as an impossible task — if they just get killed on this test — I'm really worried about losing a lot of those kids" to dropout, said Overbeck, who serves as an adjunct professor at the University of Dayton and Wright State University, and who has participated in state education department testing committees that reviewed Ohio proficiency tests.
Students who do not get a diploma face significant obstacles in getting ahead in education or careers, said Vicki Huff, senior guidance counselor to half of Xenia High School. Some universities, academic programs and jobs require a high school degree and reject applicants even if they have obtained a GED, the equivalent of a diploma. On average, the long-term earning potential of those without diplomas is greatly reduced, Huff said.
The Washington, D.C.-based Center on Education Policy, which annually tracks the impact of high school exit exams in the United States, has noted that states with high dropout rates have high school exit exams, while few or none of the states with the lowest dropout rates require passage or a statewide test for graduation. In a report entitled "State High School Exit Exams: Put to the Test," the center's associate director, Keith Gayler, said, "While we cannot yet directly link exit exams to higher dropout rates, there is enough evidence to suggest a relationship between the two."
A report released this month by the New York City-based Manhattan Institute disputes the link, concluding that adopting a high school exit exam "has no effect on a state's graduation rate."
Ohio Department of Education spokesman J.C. Benton said the department has kept statewide graduation rates only since 1997-98 — four years after the Ohio Ninth Grade Proficiency Test became a graduation requirement. Between 1998 and 2003, Ohio's overall graduation rate has risen from 80.2 percent to 83.9 percent, Benton said.
Deputy state school superintendent Bowers said members of the class of 2007 and all who follow will have at least five chances to take the Ohio Graduation Test exams before graduation — and as many as seven if they take special summer exams.
"I'm optimistic they can do it," Bowers said.
Greg Bernhardt, dean of Wright State's college of education, agreed: "As we get closer to 2007, I think people will get this message and work very hard on it."
Search for alternatives
Ohio educators don't yet have an answer to a question crucial to the future of the Ohio Graduation Test: How do you make the test fairer without compromising a commitment to high standards?
A March 29 "working draft" of a report by the state board of education's "Task Force on Quality High Schools for a Lifetime of Opportunities" suggested that Ohio should move away from the high-stakes Ohio Graduation Test toward end-of-course exams that measure knowledge gleaned in a particular subject after students complete a class. That was a recommendation of an earlier governor's task force that legislators ultimately rejected.
The draft report also suggested that Ohio consider using a college entrance exam such as the ACT or SAT as a substitute for the Ohio Graduation Test, or consider adopting an appeals system under which students who have fulfilled the requirements of a rigorous curriculum, but have not passed all portions of the Ohio Graduation Test, might still obtain a diploma.
Ohio legislators already have created an alternative route to a diploma for students who pass four of the five Ohio Graduation Test subject areas in 2007 and thereafter, but they set demanding criteria. Students must come close to a passing score on the failed test; earn a 97 percent attendance rate through all four years of high school; carry a grade-point average of at least 2.5 out of a 4.0 in the subject area missed; participate with a 97 percent attendance rate in any intervention programs offered by the school and obtain letters recommending graduation from teachers and the high school principal. Educators doubt that many students will qualify for this alternative route to a diploma.
Carl Wick, a state board of education member from Centerville who serves as co-chairman of the quality high schools task force, said the group's members are looking for a way to recommend an alternative to the current system that places so much emphasis on the Ohio Graduation Test — "bold measures that still do not back off or dumb down state standards."
The Center on Education Policy study of high school exit exams nationwide reported that some states responded to initially high failure rates such as Ohio's by postponing punitive measures, adjusting passing scores or allowing alternative measurements. The report summarized a battle that Ohio may soon face, as the Ohio Graduation Test's full impact approaches:
"In several cases, debates about whether to amend exit exam requirements were contentious and highly politicized. State leaders who had advocated exit exams as a necessary remedy for inferior schools and apathetic students searched for ways to keep graduation rates at an acceptable level and avoid massive backlash, while not appearing to back down from their commitment to rigorous standards."
Wick said Ohio's current testing requirements "might be overly harsh in certain instances," and acknowledged that he and his fellow state school board members will analyze the most recent Ohio Graduation Test results with great interest.
UD education dean Lasley said any resolution to the graduation-test debate will be difficult educationally, politically and socially. But he said state officials and educators owe it to students to attack — and solve — the problem.
"A lot of kids are putting their lives into the hands of a lot of adults, and hoping those adults make good decisions," Lasley said.
One of those kids, Xenia High School student Sara Craft, found out Monday afternoon that she passed the math proficiency test that had threatened to block her from graduating. Today she's making plans to march in Thursday's commencement.
"It was a really nerve-wracking week last week — I was freaking out every day," Craft said. "I feel like a big burden is completely off my back. It's a big relief."
Mark Fisher and Scott Elliott
Dayton Daily News
INDEX OF NCLB OUTRAGES