New test scores; old concerns
by Alexa Aguilar and Georgina Gustin; Data analysis by Jaimi Dowdell
Annette Hoffeditz remembers the day last year when she learned that the standardized test scores of her third-grade class had dropped by 30 percent.
"I was disappointed," she recalled. "But not shocked."
This year was a different story. The 2005 results of the Illinois Standards Achievement Test, released by the state today, show that Hoffeditz's class at Franklin Elementary in Belleville scored more than 40 percent higher.
"Everything I put in front of them, they just got it," she said, snapping her fingers.
Same teacher, same principal, same school - but a different group of third-graders.
Locally, dozens of schools posted gains or dips of 20 percent to 40 percent, compared to the previous school year.
"You're comparing last year's third grade to this one," said Tim Barth, curriculum director for the Belleville High School District. "It may have nothing to do with the school's performance, and everything to do with the mix of students. And in smaller schools, a few students have more impact."
Four years into the No Child Left Behind Act - the federal legislation mandating that schools show annual gains - the emphasis on standardized tests is greater than ever. Yet, educators say, results can be misleading until long-term trends can paint a more complete picture. Scores also will become more meaningful this spring when the state will begin testing students in every grade.
Still, school officials pore over the scores, searching to see which groups of students failed to make "adequate yearly progress." That's the term for achievement goals the state must set each year, by federal law. If schools don't meet those goals, they must offer tutoring and the option to transfer to other schools. Ultimately, a school could be taken over or have its federal funding withheld.
This year, Illinois required 47.5 percent of students tested to score proficient or better for a school to make "adequate yearly progress."
Locally, 73 percent of 81 schools in Madison County made adequate progress. In St. Clair County, 69 percent of 94 schools hit the mark.
Some of the area's most well regarded schools, such as O'Fallon, Edwardsville, Highland and Triad, once again outscored statewide averages. But every school has the potential to fall short under the federal law.
Jim Rosborg, a retired school superintendent who teaches at McKendree College, said even if some of the kinks are worked out of the state's testing system, all schools will eventually fail, because the law requires 100 percent of students to meet standards by 2014. That's not going to happen, he said.
Changes are in place
Improvements are planned for the state tests next year. Rosborg said teachers finally will be able to see what questions the greatest number of students failed to answer correctly and focus their attention on those skills the next year.
And the state will start testing every grade, so that students' performance in third-grade can be compared to how they fare in fourth-grade, and so on.
That makes more sense to Linda Held, a third-grade teacher at Kreitner Elementary in Collinsville. Held has taught for 25 years and said she feels more than ever an enormous pressure for her students to perform well on tests.
At Kreitner, the percentage of third-graders who passed the math test jumped to 81.3 percent this year from 61.5 percent last year. But so many factors play into how that happens, Held said. Students' home lives, how well Hispanic children have learned English, what kind of background and life experiences they bring to the table - all affect test performance, she said. Even the way questions are worded or the way they are presented in the test booklet can affect scores.
"A couple of students can pull your scores up or down," she said.
Many in this year's class are reading below grade level, Held said. She knows they are all making gains, but the state's reading test doesn't always prove it.
Monty Neill, director of the National Center for Fair and Open Testing, an organization that advocates less emphasis on standardized tests, agrees that the makeup of each class determines big jumps up or down, rather than a change of program.
"Taken by itself, parents shouldn't put a lot of stock in a test score," Neill said. "But it's a snapshot indicator" that can signal an area where a school needs to improve.
Subgroups are a key
That's why the law's emphasis on the performance of smaller groups of children is so important, say proponents of No Child Left Behind.
The law requires that various "subgroups" of students - white, black, low-income, special education and nonnative English speakers, for example - must meet the same standards as the overall student body. If one of those groups fails to meet the goal, the whole school fails.
In Illinois, there must be 45 students in any of these subsets to be considered a subgroup, which means larger, more diverse schools usually have a greater burden. If, for example, a school has 44 special education students, that group will not have to meet the targets.
In the Metro East area, 14 schools fell short of standards because of the performance of just one subgroup.
One of those schools - Belleville East High School - has failed to hit targets for the past four years, despite the fact that it's among the highest performing high schools in the area. Belleville East's special education students failed to meet this year's goal, so the whole school is sanctioned.
"The whole thing about NCLB is the fine print," said Barth, the district's curriculum director. "It's really about whether the community understands what it means."
Alexa Aguilar and Georgina Gustin
St. Louis Post-Dispatch
INDEX OF NCLB OUTRAGES