Under the federal “No Child Left Behind” law, school districts, schools, and teachers face harsh, punitive consequences for failing to do the impossible (consistently increase test scores for 14 years in a row).
It’s no surprise that some administrators and teachers have turned to cheating on the high-stakes tests required by the federal “No Child Left Behind” law.
Follow this link Google News to see recent news articles about the detection of “unusual erasure patterns” on some schools’ tests. (Feel free to modify the search by adding the name of your state.)
There’s a whole art to identifying classrooms and schools where there are an unusually high number of “erasures from wrong to right.” The systems used to identify this form of cheating rely on statistical models (mostly aggregate, not focusing on individual students’ performance compared to past years).
This is not the only type of cheating, but it is the cheapest to detect. There are several types of “erasure cheating,” including a teacher pointing to a student’s wrong answer so that the student changes it, or a teacher or administrator making changes to tests after the students are dismissed. Other forms of cheating are much more expensive to detect and harder to prove.
Detecting “an unusual pattern of erasures” is not proof of cheating, and could represent a particular test-taking strategy (urging students to quickly read and respond to all questions, then use the remaining time to check their answers). If a district, school, or teacher recommends this strategy to students, then those tests should be expected to have a higher ratio of “erasures from incorrect to correct.”
Turn it around: if a district, school, or teacher recommends a different test-taking strategy (“carefully read and give a considered response to each question, even if it means that some questions may be left unanswered when time runs out”), then “erasure cheating” on those tests might be undetectable (with a much lower starting point AND the increased chance of unanswered questions, it’s unlikely that an objective reviewer would discover a statistically significant ratio of “erasures from incorrect to correct).
Why would students use a strategy intended for timed tests for a test without a time limit? Because students are likely to use the same strategies for all tests — and when students choose test-taking strategies, they know they’ll face huge consequences for their scores on college admissions tests (but no consequences for their performance on NCLB tests).
Cheating is a predictable result, if the test results will impact schools and teachers personally. It’s not an ethical or proper response — but the “No Child Left Behind” law was never ethical or legitimate, as an educational policy.
Added July 6: Another standardized-testing scandal makes the news: Atlanta’s improved test scores were based on “rampant, systematic cheating on test scores … ending two years of increasing skepticism over remarkable improvements touted by school leaders. The administration … punished whistle-blowers, hid or manipulated information and altered documents, the investigation found. … the cheating occurred at 44 schools and involved at least 178 teachers and principals, almost half of whom have confessed … A culture of fear, intimidation and retaliation existed in the district, which led to a conspiracy of silence…” http://www.nytimes.com/2011/07/06/education/06atlanta.html