Mind the Gap: Teachers Grading Their Own Exams
An inner-city schoolteacher attempts to dampen enthusiasm for standardized testing.
As I discussed a few weeks ago, it’s testing time here in New York City high schools. Since mid-May, many teachers of classes that culminate in Regents exams have been preparing for these high-stakes tests. I’ve been as intensively engaged in the work as any other teacher, yet as news continues to emerge on the unreliability of standardized testing, I’ve begun to feel as if I’m engaged in a fraudulent endeavor.
There are problems with the questions being asked, the answers being graded, and the people who are involved. With teachers’ salaries and even jobs now being determined in part by standardized testing scores, these problems need to be urgently addressed.
Many of these tests inquire about a world that is foreign to my students. Exams within the last two years have asked about organic food, horses, hay, and hang-gliding, none of which most of my students are very familiar with. The tests’ emphasis on topics many inner-city students are unaware of recently motivated one Harlem charter school to visit a farm. This wasn’t a field trip. Rather, an innovative form of test prep.
As Eva Moskowitz, the schools’ founder, told The New York Times, “There were passages, literally, about milking, plowing—things that were pretty foreign to Harlem kids. It’s a little bit annoying that there are no passages about the subway, or how crowded the streets are.”
The problems in New York are so widespread even the top education officials in the state have lost confidence in them. As Sol Stern recently reported in City Journal, “Board of Regents chancellor Merryl Tisch and education department commissioner David Steiner, recently concluded that the annual math and English tests for grades three through eight had become unreliable measures of children’s real academic achievement. They are trying to restructure the state’s assessments and recently ordered a study by [Harvard testing expert Daniel] Koretz to measure the extent of score inflation on the tests given in the past several years.”
Just today, the Times came out with a story about score inflation on standardized tests across the country. This news comes just a week after the New York Post published its own expose of bogus grading—for example, “A kid who answers that a two-foot-long skateboard is 48 inches long gets half-credit for adding 24 and 24 instead of the correct 12 plus 12.” Partial credit was also given for blank answers.
Incentive systems that have stressed standardized testing to the detriment of other forms of assessment are partly to blame for such activity. Equally culpable, however, are nonsensical procedures like having teachers grade their own assessments, the very tests for which they can receive a pay raise—or get a pink slip.
On Tuesday afternoon, my students will take the Global History Regents exam. Wednesday morning, my colleagues and I will gather to grade those tests. While the state sends us a rubric to go by, their materials are not exhaustive and there is room for interpretation. That wiggle room, which the more liberal teachers at one school might exploit while more conservative teachers at a neighboring school might not—introduces incredible unreliability to the testing data.
A colleague of mine recently asked New York City Chancellor Joel Klein, a huge proponent of standardized tests and chief cheerleader of the city’s rising scores, what he has done to reform the corrupt practice of teachers grading their own exams. Klein did not seem overly concerned, saying that it was a state issue, and he’s pressing state education leaders to address the system.
For Klein and other education leaders across the country, the ends justify the means. But if the means become corrupt, the ends become invalid.
I’m a huge proponent of teacher and student accountability, of measuring and tracking academic growth. However, the tools we use to accomplish such things are increasingly being exposed as not credible. Thus, the great disparity in state test scores, which teachers can methodically prepare their students for, and the less predictable federal National Assessment of Educational Progress tests.
I support the reform movement’s effort to bring more accountability into the system, but the current practices are untenable. I still haven’t figured out what I’ll tell my students on Tuesday.
Brendan Lowe is a Teach for America corps member who is in his second year of teaching high school in the South Bronx. His dispatch for GOOD appears on Fridays.