Guest Opinion | Student Testing Cheating – Does PUSD have Georgia Issues?

[Please note: Pasadena Now publishes guest opinions as a service to the community, to allow freedom of expression and diverse points of view the opportunity to reach the public. The authors are solely responsible for the content. Pasadena Now does not edit opinion pieces and is open to receive op/eds from members of the local public on any topic.]

 

Pomeroy’s 2011 suspicions about Roosevelt cheating degenerate to currently defending Roosevelt’s former principal

In 2011, Pasadena Unified School District Board Member Elizabeth Pomeroy’s instinct was that Roosevelt Elementary School Principal Juan Ruelas’ seemingly remarkable achievements raising Roosevelt’s student test scores more likely reflected cheating, not genuine educational achievement. On September 5, 2011, Pomeroy read an article by Los Angeles Times education writer Howard Blume entitled “Cheating on state tests found at two Los Angeles schools.” At 10:27 a.m. that morning, Pomeroy sent an email to PUSD’s Director of Elementary Education Dr. Kathy Onoye asking “What would be the ways to find out whether there is anything amiss in the Roosevelt scores?” Dr. Onoye passed Pomeroy’s inquiry on to the Superintendent, who responded “You can let her know that there is an ongoing investigation of reported irregularities at Roosevelt and that the board will be updated in closed session next week.” Thus, from the test scores alone and without knowing that there was already the ongoing investigation that did determine that Roosevelt was cheating, Pomeroy accurately suspected there was cheating.

As the current PUD Board President, Pomeroy is now the chief apologist for Ruelas. Ruelas was imposed by PUSD Superintendent Brian McDonald on Madison Elementary School as its principal in May 2015 despite objections that he was discriminating against the Madison community by refusing to allow its stakeholders to select 3 applicants from which he would choose the new principal – a privilege he allowed other more wealthy and more anglo communities, including Sierra Madre. McDonald excuses his discriminatory treatment of Madison on the claim that Ruelas’s Roosevelt record of purported educational achievement warranted rejecting any Madison site input into its principal selection decision. Pomeroy today has quieted her suspicions about Ruelas’ purported achievements. She dismissively rejects drawing any conclusions from Roosevelt’s dropping from the 3d highest performing elementary school when Roosevelt got caught cheating down to the 14th rank – which happened in just 4 years. Whatever happened to Pomeroy’s good instincts?

An Atlanta newspaper’s suspicions similar to Pomeroy’s lead to criminal convictions for racketeering

There are well-established best practices to detect cheating. In 2009, the newspaper Atlanta Journal-Constitution was similarly suspicious that Atlanta public school test results reflected cheating, not genuine educational achievement. Its prize-winning report suggested that 44 out of 56 Atlanta schools cheated on the 2009 state student test. The paper’s report set off the events leading to indictments against Atlanta’s Superintendent of Schools and 12 others on criminal racketeering charges. The Superintendent died before trial; 11 of those charged were convicted, with sentences as harsh as 20 years imprisonment.

Test-Score Statistical Analysis was the well-established cheating detection method that first raised the Journal-Constitution’s eyebrows. The U.S. Department of Education’s 2013 “Testing Integrity Symposium: Issues and Recommendations for Best Practice” describes this common detection method: “Test-score analysis examines test scores to see if there have been unusually large gains from the previous year or large relative declines in the subsequent year.” Test score analysis was central to the Atlanta newspaper’s initial exposure of the Atlanta scandal, as reflected by the title of its October 19, 2009, article “Are drastic swings in CRCT scores valid?” Subsequently, both the Journal-Constitution and law enforcement used a number of other cheating-detection methodologies, but recognition of the scandal began by simply looking at the year-to-year test scores and spotting suspicious trends.

Pomeroy’s reasonable suspicions from 2009-2011 test scores that Roosevelt was cheating

What was it about Roosevelt’s test scores that led Pomeroy in 2011 to suspect cheating? Simply looking at the Roosevelt test scores graphed above and knowing the few facts noted on the graph tells the story. What Pomeroy saw was a remarkable rise of test scores during Ruelas’ first two years at Roosevelt. Those scores rocketed Roosevelt to the 3d best test-score ranking among PUSD’s elementary schools – a rapid 92 points rise from the pre-Ruelas overall 796 API score average to the astronomic 888 score in just two years. Pomeroy’s intuition that Ruelas’ meteoric rise was suspicious was well-founded. Unbeknownst to her, PUSD was already investigating this suspicious Roosevelt increase, PUSD had already determined that Roosevelt had cheated on the 2011 API testing, and PUSD was already negotiating with the California Department of Education on the penalty for its cheating. That process ended in January, 2012, with PUSD’s press release announcing that the State withdrew Roosevelt’s Blue Ribbon School nomination because of its cheating on the 2011 tests.

The reasonable inferences of widespread Roosevelt test cheating that have not been investigated, uncovered, nor admitted

The cheating that PUSD caught was in Roosevelt’s 2nd grade and only involved 3 of the 2nd grade students. PUSD’s January 2010 press release disclosing Roosevelt’s cheating and its losing the Blue Ribbon School nomination minimized Roosevelt’s cheating as “minor irregularities” affecting only 1.6% of the students.

Such “minor irregularities” would not likely produce a significant drop in test scores, but the test-score statistical analysis involved in looking at Roosevelt’s chart above indicates the different story that test scores did significantly drop after Roosevelt was caught – not just once, but twice. Each drop is consistent with significant Roosevelt test-cheating that has been ignored.

The 1st significant drop occurred after Roosevelt was caught cheating in 2011. The school had to stop some of its cheating to avoid getting caught again. So its scores significantly decreased in 2012 and 2013 when it had to reduce cheating, causing a decline from the 3rd highest ranking PUSD elementary school to its 6th ranking. The post-2011 significant decline in Roosevelt test scores and District rankings cannot be accounted for by “minor irregularities” with 3 students who constituted only 1.6% of Roosevelt’s students. Rather, the significant decline in test scores after Roosevelt got caught cheating in 2011 indicates there was significantly more test cheating going on which remained uninvestigated and undetected – s0me of which had to be eliminated because the spotlight was on Roosevelt. So PUSD probably discovered and remedied only the tip of Roosevelt’s 2011 cheating; 90% or more of the cheating iceberg remained undetected and uninvestigated.

The 2nd significant drop occurred when the State eliminated paper testing after the 2013 test scores. There was no statewide testing in 2014 as the schools transitioned from API testing to SBAC testing. 2015 testing shifted from paper-and-pencil testing to more secure electronic testing. Electronic testing results go directly from Roosevelt’s electronic terminals to an off-site electronic collection center. In 2015, Roosevelt thus lost control over the paper tests. No longer controlling papers test took away the opportunity to cheat by tampering with the paper tests. After the more secure 2015 testing, Roosevelt plummeted from the 6th best ranking of PUSD elementary schools to ranking 14th. 2015’s 14th ranking probably accurately reflects actual relative educational achievement throughout his tenure as Roosevelt’s principal – i.e., nothing to be particularly proud of and is not an educational miracle. (Test scores comparison between API testing and SBAC testing is not considered valid, but relative rankings have probable validity. The 14th schools 3-year 2011-2013 average API score was 753, which we believe is a reasonable estimate of the API score that reflects Roosevelt’s actual 2015 educational achievement in Ruelas’ last year at Roosevelt.)

Ignoring the cheating scandal doesn’t protect the new Superintendent against his big mistake

McDonald’s rationale for imposing Ruelas on Madison rests on his alleged accomplishment of educational miracles at Roosevelt. The grass-roots coalition objecting to McDonald’s Ruelas decision is asking PUSD to investigate whether Ruelas’ purported achievement at Roosevelt is genuine or based on test cheating. Pomeroy as the current PUSD Board Chair today refuses to embrace any suspicions about Ruelas’s Roosevelt test record. She refuses to apply test-score statistical analysis as she did in 2011. Instead, Pomeroy dogmatically dismisses any criticism of Ruelas or Superintendent McDonald’s decision and leads the Board in refusing to take the same kind of critical look at test scores that led her to correctly discern that there was probably cheating at Roosevelt in 2010.

Pomeroy is not forgetful. She just does not want to face the cheating scandal because she thinks it will undermine Superintendent McDonald. The desire to protect the new Superintendent from public accountability for his first major mistake in unilaterally imposing Ruelas on Madison is corrupting PUSD’s policy-making on test cheating – and on every other public policy issue that touches Madison. PUSD has had the unfortunate experience of musical chairs in the Superintendent’s seat over the past few years, and Pomeroy shares the desire of practically everyone in the community to see McDonald succeed. McDonald has a chance to succeed, but not if saving him is at the cost of prostituting public policy. But that’s what is happening at PUSD. Ignoring cheating on student tests has far worse consequences than honestly facing up to the Superintendent’s Madison mistake and correcting it.

Dale Gronemeier and Skip Hickambottom are local civil rights attorneys who represent the Citizens Council for Empowerment and Justice at Madison, the grassroots organization of parents, teachers, and community members who are demanding that Ruelas be removed as Madison’s principal.