This article was originally published in The Notebook. In August 2020, The Notebook became Chalkbeat Philadelphia.
The Notebook recently began sharing content with Education Week, where this piece originally appeared.
As we continue to search for reliable measures to evaluate a school district’s progress and debate the true meaning of test score gains, this sobering story broke in Chicago. A complex analysis of test and other data over 20 years showed that, despite perceptions that school reform in Chicago has led to significant student progress, there was little gain in elementary reading and the racial achievement gap got wider over that period. There was some progress in high schools – an uptick in ACT scores and higher graduation rates – but students really are not more prepared for college, according to the study by the University of Chicago Consortium for School Research.
by Jaclyn Zubrzycki
A new study looking back at nearly 20 years of data on Chicago’s public schools suggests that changes in standards and in test-taking and data-reporting policies over time have led to misconceptions about the city’s progress in improving school and student performance.
Researchers from the Consortium on Chicago School Research, an independent research group based at the University of Chicago, said their analysis indicates that high school performance in the Chicago district improved more than elementary school performance did, contrary to what many local educators may believe, and that the achievement gap between African-American and other students has widened over the past 20 years.
The report details test scores and graduation rates from 1990 to 2009 across three eras: the launch of the city’s landmark school-decentralization effort and Paul G. Vallas’ and Arne Duncan’s separate tenures as chief executive officer of the Chicago schools. Though improvement efforts over those years have led to a number of policy changes, “in general they have built on each other,” said Elaine Allensworth, the chief researcher with the consortium.
Analyzing the data was a project that “took three Ph.D.s three years,” the researchers said in an interview, and the complexity of tracking down and analyzing the data led to one of the report’s key findings: “The publicly reported statistics used to hold schools and districts accountable are not accurate measures of progress.”
Scores can provide a useful snapshot, but the publicly reported data show something different with each change in policy, according to the report.
Stuart Luppescu, the chief psychometrist with the consortium, said the report includes all scores, not just those that are publicly reported, thus offering insights into performance trends across time.
Mr. Luppescu described some of the changes that affected the results: “In 2002-4, there was a break given in the middle of the reading section [of the Iowa Tests of Basic Skills].” A jump in scores at that time may be related to that practice, he added, and renorming over time also led to boosts in elementary school scores.
Former CEO Vallas said that he held off on renorming during his tenure to maintain rigor and had “five years of improved reading and math scores,” and that scores rose 5% after the test was renormed. The researchers agreed. “Students seem to be getting higher scores over time for the same level of skill,” Mr. Luppescu said.
In 2006, Chicago replaced the Iowa Tests of Basic Skills with the ISAT, or Illinois Standards Achievement Test, which Mr. Vallas, the former schools chief, described in an interview with Education Week as an “easier” test.
“Policies are made without a thought to how they will affect comparability of statistics,” Mr. Luppescu said.
The report says these changes in elementary-level tests made it appear “that high schools are less successful [than elementary schools] when, in fact, [high schools] are simply held to a much higher standard.” Tests taken by high school students are based on college-readiness standards, Mr. Luppescu noted.
Jason Cascarino, the chief operating officer of the Chicago Public Education Fund, a private foundation that invests in education programs in the city, said that finding was “the big reveal of this report. For a long time, we have thought that elementary schools were making incremental gains. … But the definition of what is proficient is what matters.”
Raising the Bar
High schools have shown growth in meeting college-ready standards, increasing a full point between 2001 and 2009 on the ACT college-entrance exam. But the report says that most students in the 408,600-student school system come into high school far behind where they need to be. “If we want kids to be college-ready, we need to use college standards [sooner],” Mr. Luppescu said.
The report also finds that dropout rates have declined significantly, though changes in retention policy have caused the rate to fluctuate from year to year.
“In the early 1990s, students who entered Chicago high schools were equally likely to drop out as to graduate,” the report says. “Now they are more than twice as likely to graduate as to drop out.” While that statistic might hint at decreased standards, the rise in ACT scores suggests that academic performance in high schools has continued to rise with the graduation rate, the authors write. The study tracks dropout rates by age cohort rather than grade, which the authors said is a more accurate measure of dropout rates.
Which data are reported publicly has also changed over the course of 20 years, according to the report. “Prior to 2008, students’ test scores could be excluded from public reporting depending on their bilingual or special education status,” it says, and very often were. In the late ’90s, up to 25 percent of student scores were not reported to the public due to that practice.
That omission partially accounts for the consortium’s rosier depiction of 1990s reforms in its 2010 book, Organizing Schools for Improvement: Lessons From Chicago, which also begins its analysis in 1988 rather than 1990, as the current study does. And it makes it difficult to trace student-achievement trends over time.
Further muddying the numbers, the racial and ethnic composition of the district has also changed over time: The percentage of low-performing African-American students decreased, and the percentage of bilingual students whose scores weren’t always reported rose.
Numbers in Context
The report finds that focusing on specific student groups reveals some troubling trends. For one, the achievement gap between the lowest-performing African-American students and all other racial and ethnic groups is greater than the comparable nationwide gap and grew in all three school reform eras the report studied. Ms. Allensworth characterized the growth of that gap as “disheartening.”
Still, the study says, students of similar demographic backgrounds in other Illinois schools actually perform worse on standardized tests and are more likely to drop out. The report “does say Chicago is not doing worse by their kids than other districts as a whole,” Ms. Allensworth said.
The results lend support to the current Chicago school board’s focus on college readiness, said Mr. Cascarino.
Officials from the Chicago school district could not be reached for comment.
But the researchers conclude that there is still work to be done, as “the district has a long way to go before the average student graduates ready to succeed in college,” and that mandates won’t fix all of the school system’s programs.