The Dark Side of High Stakes Testing

A low test score isn't just bad news for a student, it can affect teacher bonuses, school funding, and more. In today's post, Dan Edmonds discusses the reality of cheating on high stakes tests by teachers, administrators and institutions of higher learning.

Last week, I touched on the recent Long Island SAT cheating scandal, and how it reflected potential issues in our system of high stakes tests like the SAT, ACT, Regents, CRCT, or GRE.

Today, I want to look at how high stakes testing is having a very real impact on the behavior of teachers, administrators, and institutions of higher learning.

Let's start with the scandal last year in Georgia, where 178 teachers and administrators were implicated in what is thought to be the largest confirmed cheating scandal in US history. In this case, teachers, principals, and administrators worked together not only to fix answers on the CRCT, but also to hide any traces of the improprieties, and even to intimidate or punish whistle-blowers. The question becomes: why would teachers and administrators help students to cheat on a test (or, as the case was in Georgia, cheat on behalf of students who might not even be aware the cheating is going on).

I've heard stories of similar cheating on a much smaller scale here in New York; I had a student a few years ago who talked about how the proctor in her school for her Regents exams--herself a teacher in the school--helped the students to answer some of the tougher questions on the test. Why? Because we judge teachers, principals, schools, even school districts based on the results of these high stakes tests. Their results can determine everything from merit pay and bonuses to extra funding from the state. In some cases they can even lead to the closing of schools that are deemed to be "failing."

With all this at stake--including jobs, promotions, and professional reputation--is it any wonder that some teachers are tempted to cheat?

Less serious but far more common than actual cheating is the tendency for teachers, as a result of these policies, to "teach to the test." In other words, teachers in public schools are increasingly under pressure to modify their curricula in an attempt to get better scores for their students on these high stakes tests that are used to evaluate student and teachers alike; this is potentially problematic, as there are serious questions about how teaching to the test impacts the quality of the education students receive. The question doesn't have a simple answer, but it's one worthy of consideration.

I'm personally of the opinion that over-emphasis on the results of high stakes tests is likely to particularly impact the education of what I'd call the outlier students: those students who are so weak that they're unlikely to pass these tests (losses to be written off), and those students so strong that their passing the tests is a foregone conclusion. The former may not receive the remediation and personal attention they need to thrive; the latter will likely not be given the challenges they need to reach their full potential. Unquestionably, there are many complexities that go into meeting the needs of students with differing abilities levels. That said, when high stakes tests are over-emphasized, teachers are given clear incentives to teach to the students who are close to passing the test, and to ignore the needs of the students who excel or the students who are sure to fail. Evidence seems particularly strong that high-achieving students may be receiving the short shrift.

Fittingly, the very day that I published my blog last week, the news broke that Claremont McKenna had, for years, been padding its SAT scores. Why would such a prestigious school, which just recently cracked the list of the top 10 liberals arts colleges, resort to this kind of manipulation? Quite possibly to help them, for instance, to crack the top 10 in the US News and World Report!

In the same way that K-12 institutions are judged based on the performances of their students on high stakes tests like the Regents or the CRCT, average SAT (and ACT) scores are used to help rank the competitiveness of colleges and universities. This has long impacted admissions decisions and criteria at colleges, universities, and even graduate schools. We now know this has, in at least one case, led a college administrator to knowingly falsify information.

So what's the bottom line here? The fact is that standardized tests are here to stay. How do we mitigate some of the negative impact of these tests, an impact grounded in the fact that we use them not only to judge individual students, but also to judge teachers, administrators, and even institutions? I don't have a quick and easy answer, but I firmly believe that both the health and integrity of education in America depends on our wrestling with these questions.

Previously: The First Rule of Test Scores

Next: Infographic: How Google is Changing Our Memory