Atlanta Journal-Constitution Says Suspicious Student Test Scores Found Nationwide
In the latest example of a newspaper challenging the education
establishment on the issue of student achievement, an investigation by
Atlanta Journal-Constitution reportedly found suspect
test scores in public school districts across the country.
The AJC used open records requests to seek student test data from all 50 states and the District of Columbia. The suspicious test score patterns – which alone is not evidence of cheating – were found in 196 of the nation’s 3,125 largest school districts.
The seven-month investigation, by the AJC and affiliated Cox newspapers, is already getting pushback from some researchers who question the methodology. But the reporters concluded the huge shifts in test scores in cities like Houston and Los Angeles mirror the patterns found in the Atlanta public schools where cheating was determined to have been widespread.
A Georgia investigation into those rapid gains in scores, triggered in large part by the AJC’s reporting in 2008 and 2009, found nearly 200 educators – teachers, principals and regional superintendents – colluded to falsify student achievement on statewide tests. The fallout has been significant, and led to the ouster of Atlanta Superintendent Beverly Hall.
The AJC’s latest findings “are concerning,” U.S. Secretary of Education Arne Duncan said in an emailed statement to the newspaper. “States, districts, schools and testing companies should have sensible safeguards in place to ensure tests accurately reflect student learning.”
According to the AJC, “Overall, 196 of the nation’s 3,125 largest school districts had enough suspect tests that the odds of the results occurring by chance alone were worse than one in 1,000 … For 33 of those districts, the odds were worse than one in a million.”
But Gary Miron, an education professor and researcher at Western Michigan University, wrote in an opinion piece that the AJC’s methodology was flawed, and that “the resulting news story appears to be intended to be alarmist, implying that cheating is rampant in our schools.”
Among Miron’s concerns, published in the Washington Post’s Answer Sheet education blog, was that the investigation did not take into account the impact of high rates of student transiency on achievement. Charter schools represented a high proportion of the schools with alleged test anomalies, and they also have a disproportionate number of students moving in and out over the course of an academic year, Miron said.
The data used for the AJC analysisis also stops at the school level, rather than going all the way down to the individual student level. That makes it impossible to know whether a suspicious results, such as a sudden increase in performance followed by a steep drop-off in scores the following year, actually involved the same group of students, Miron said. (The AJC has since responded to Miron’s criticism, pointing out that many urban districts with high transiency rates did not also have testing anomalies.)
Gary Phillips, a vice president and chief scientist for the nonprofit American Institutes for Research and the newspaper’s advisor for its methodology, told the AJC that “extreme” changes in test scores are like medical tests: “When you find something, you’re supposed to go to the doctor and follow up with a more detailed diagnostic process.”
The next question is how individual districts will respond to the AJC’s investigation, and whether the newspaper’s findings will trigger closer examinations of potentially suspect test scores.
Several districts cited in the AJC report for having particularly high rates of test score anomalies have already spoken up. In the case of Houston’s public schools, the response included a promise a closer look at the findings. In Nashville, district officials said the test score anomalies showed steep declines in achievement, rather than suspicious gains, and were the result of the district’s high rate of student transiency, absenteeism and the large percentage of English language learners, the Tennessean reported. In a written statement, the district also challenged the AJC’s methodology, citing concerns raised by an education researcher at Vanderbilt University.
Los Angeles Unified also made the AJC’s list. That was probably not a surprise to the nation’s second-largest school district, where there have been ongoing issues in recent years related to cheating allegations. A district investigation in 2011 determined teachers were giving students improper access to test materials ahead of the exam, coaching them to provide the correct responses, and even changing answer sheets, according to the Los Angeles Times. Additionally, six charters schools were forced to close in 2010 amid allegations of cheating on statewide exams.
Educators and researchers have argued that the intense emphasis on student test scores, a mandate of the federal No Child Left Behind Law, put pressure on educators to show dramatic (and often unrealistic) gains in student achievement. As a result, they argue, cheating is much more likely to occur.
The LA Times had a story with a similar thesis a few months ago, in which teachers, speaking anonymously, said they would indeed cheat if their jobs were on the line.
That type of cheating – individual teachers taking action to save their own jobs — is a far cry from the kind of high-level, orchestrated malfeasance that reportedly took place in Atlanta. But there is little doubt that public schools are facing significant expectations for student gains.
The AJC story “certainly raises red flags that should lead many individual districts to invest in independent investigations,” said Robert Schaeffer, public education director of Fair Test, an advocacy group that is critical of the overemphasis on standardized tests.
If nothing else, Schaeffer says, the recent cheating scandals “have again demonstrated that overreliance on standardized test scores is a flawed strategy for creating lasting educational reform.”
Atlanta, for better or for worse, has become the reference point for stories about widespread cheating and schools, and it is likely that any gains by its students will be considered suspect for a long time to come. In addition to thousands of schoolchildren whose own learning has been potentially hurt in the process, the loss of the public’s trust in its school system is perhaps one of the scandal’s more disheartening legacies.