About a quarter of Chicago’s elementary schools boosted student reading scores significantly over the last seven years, with the rest showing little or no significant improvement, according to a study released in late October by the school advocacy and research group Designs for Change.
The study’s author, Designs Executive Director Donald Moore, counts the improved performance as evidence that reform centered around local school councils is succeeding. Schools with the biggest test-score gains tend to have better leadership, stronger ties to parents and communities, and more cohesive faculties—the qualities that councils were designed to foster, says Moore. The most-improved schools also tend to have more active local school councils than do less successful schools. These factors were measured in 1994 teacher and student surveys by the Chicago Consortium for School Research.
Role of reform
The School Board’s chief accountability officer, Phil Hansen, applauds the good news, but he notes that leadership, parent connections and faculty teamwork are not unique to Chicago, nor to schools governed by local school councils. They’re part of the national literature about what makes a good school, he says, and they are also qualities that the board’s probation teams work to foster. But Hansen says he doesn’t mean to start a fight over where credit is due. “Let’s all take credit,” he says. “Let’s give the schools credit.”
Researchers at Designs followed the methods of the School Board’s research department, looking at the percentage of students at each school scoring at or above the national average on the Iowa Tests of Basic Skills for each year from 1990 to 1997. The board uses this data to determine which schools have strong trends up or down in achievement, which have weak trends, and which have no trend.
The Designs study added two conditions to make its analysis more stringent than the board’s. First, it counted only schools where at least 90 percent of all eligible students were tested in 1997, thereby excluding any schools that might have artificially boosted scores by not testing low-achieving students. Also, Designs required a steeper climb in test scores than School Board researchers did in defining the top category.
Of the 420 schools that Designs studied:
26 percent (111 schools) raised reading scores “substantially”—by an average of more than 1.5 percentage points per year.
17 percent (73 schools) were “tending up”— gaining an average of 1 to 1.5 percentage points per year.
54 percent (224 schools) showed no consistent trend up or down.
3 percent (12) schools were either “tending down” or “significantly down,” losing more than 1 point per year.
Of the 111 schools that improved substantially, 87 had low achievement levels in 1990, the study’s baseline year; fewer than 40 percent of their students scored above the national average. However, schools that had better test scores in 1990 were likelier to make marked improvement; 38 percent of 1990’s group of medium- to high-achieving schools improved markedly, compared with 25 percent of that year’s low-achieving schools.
The researchers gave extra attention to the schools that started with low achievement levels. They found:
There was a moderately strong racial pattern. Low-achieving schools with test scores that rose substantially tended to have fewer black students than did low-achieving schools with no trend; the much improved schools were 39 percent African American, while the no-trend schools were 72 percent black. However, 30 of the much-improved schools were more than 90 percent black, and the substantially-up schools served proportionately more Latino students (38 percent) than did the no-trend schools (24 percent).
There was a weak class pattern. Eighty-four percent of the students in markedly improved schools were low-income, compared to 93 percent of the students in the no-trend schools. But 30 of the much improved schools had at least 93 percent low-income kids, too.
Schools whose scores went up helped the lowest-achieving kids. Substantially-up schools that started with low achievement saw the percentage of students scoring in the bottom quartile nationally shrink from 43 percent to 29 percent.
Some of the significantly up schools started at a level that would have put them on probation.
The Designs study makes an “important contribution [in] looking at the changes in test scores over time,” says John Easton, deputy director at the Consortium, “but it doesn’t go all the way.” The missing step is “looking at the gains of individual students over time,” says Easton; that’s the aim of the Consortium’s own upcoming study of test-score information, tentatively scheduled for a December release.
Moore recommends that the much-improved schools be given special recognition, even a pseudo-charter status, because he says they have shown how well they handle greater freedom. In return, he says, the School Board could use them as a resource in helping other schools improve.
Meanwhile, the board recently held an awards ceremony for 134 elementary schools and 34 high schools that last year posted a combined increase of at least 10 points in reading and math.