Erasure analysis—the process of examining multiple-choice tests for incorrect answers that were erased and replaced by correct answers—has been in the news a lot lately. I learned about this method of catching cheaters while reading about the Atlanta cheating scandal this summer, and my first thought was how stand-up comedians would use the story. Who ends up with a career counting pencil smudges? What stories do erasure analysts tell at a bar after work?
I kept thinking about comic possibilities as I discovered more about the methods used to catch the Atlanta adults who changed kids’ answers. Former prosecutors with subpoena powers and the help of 60 criminal investigators interrogated teachers and encouraged them to turn in their colleagues, just as detectives on TV shows encourage drug dealers to rat out accomplices. I tried to picture the good-cop/bad-cop routine being used with a teacher who had made 9-year-olds’ answers into bubble C’s instead of bubble B’s.
Still, it’s hard to laugh when you consider how much the Atlanta investigation must have cost, especially when extreme budget pressures nationwide are leading to teacher layoffs and bulging class sizes. Then consider the cost of the tests themselves. It seems that the multiple expenses of administering the tests required by the No Child Left Behind Act are gobbling up scarce educational dollars, thereby undermining the federal law’s stated purpose of improving education.
Yet I understand why NCLB’s authors wanted tests to ensure that children were not being passed on from grade to grade when they hadn’t mastered the basics, and why the law passed Congress with overwhelming bipartisan support a decade ago. Throughout the 1980s and 1990s, I was enraged at the New York City public schools for graduating students who were barely literate. I was teaching English as a second language at the City University of New York, and a significant minority of my students had diplomas from New York high schools. Some had been in the United States since they were in middle school. They were in my classes because they had failed the university’s basic reading and writing tests. Many had also failed both the arithmetic and algebra sections of the math test and lacked an elementary knowledge of history, science, and geography. Most were eager to learn and made noticeable progress. Sometimes, however, they started out so far from where they needed to be that they remained in no-credit classes for several semesters.
When I first heard about NCLB back in 2001, I agreed with the bill’s supporters that schools in poor neighborhoods did students a huge disservice by expecting very little of them. When the bill became law in 2002, I hoped that New York City high school graduates would start entering college with much better skills.
Unfortunately, I’ve seen no decrease over the past decade in the number of New York high school graduates in my classes and no improvement in their academic skills. My personal observations are supported by publicly available data. According to the New York state education department, only 23 percent of all students who graduated from New York City schools in 2009 were prepared for college or careers. According to a March 3, 2011, New York Times article, about 75 percent of all City University of New York community college students in the most recent academic year had failed at least one of the university’s three basic skills tests.
Ironically, efforts by the state education department that were supposed to raise standards seem to have had the opposite effect.
Make demands to raise standards far above their current levels, and there's likely to be cheating, especially when there are severe penalties for not reaching these standards."
In 2000, students could pass the five New York state Regents tests required for graduation with a score of 55; by spring 2012, they will need at least a 65 on all five tests. But according to a Feb. 19 New York Post article, a Regents score of 65 doesn’t mean 65 percent. Instead, the Post reported, scores are based on an adjustable scale, which has changed over time. For example, in 2003, a score of 65 on the math Regents meant that 61.2 percent of a student’s answers were correct. Yet now a student needs only 30 points out of 87 to get a score of 65 on the math Regents. In other words, with just 34.5 percent of the answers correct, the student passes the math test.
The same article quotes the current chancellor of the state board of regents as saying that students with weak skills shouldn’t be handed diplomas, as they are “locked into a life with no choices.” I agree. But I also wonder—if only 23 percent of all graduates in 2009 really should have graduated, if students in the class of 2012 have similar skills, and if the rules are changed so that high school diplomas are handed only to those ready for college or careers, what will happen to that other 77 percent? Will they stay in high school until they do have adequate skills? If so, how will already-strapped high schools find classrooms and teachers for all these extra students? Or will these students be pushed into already-overtaxed high-school-equivalency programs?
In theory, virtually everyone favors the concept of raising standards in the schools. But the practical details are incredibly difficult, especially in districts where many students have very weak skills. Make demands to raise standards far above their current levels, and there’s likely to be cheating, especially when there are severe penalties for not reaching those standards. Yet without some form of accountability, students will continue graduating from high school without the skills that their diplomas are supposed to signify.
Nevertheless, accountability isn’t an end for its own sake. It exists to make sure that educational goals don’t fall by the wayside. Discussion of erasure analysis is not as important as discussion of how to improve instruction so that today’s elementary school students don’t eventually end up in remedial college classes. Improving test security is less important than figuring out what to do with the many current high school students whose academic skills are very, very far from where they should be.
Finding a way to help these students reach the standard of “college and/or career ready” is an extremely challenging task; the best way to make sure that it doesn’t happen is to threaten teachers and administrators with sanctions if they don’t show dramatic gains on test scores. That strategy will only lead to more sleight-of-hand tricks like calling students proficient in math when they manage to answer one-third of the questions on a math exam correctly. And this type of trick, instigated by those at the very top of the educational hierarchy, is at least as corrupt as changing answers on a multiple-choice test.
A version of this article appeared in the October 12, 2011 edition of Education Week as Testing and Cheating: A Tragicomedy