(Note: I originally wrote this piece several months ago, when the Chicago teacher’s strike was ongoing. In true slacker/procrastinator fashion, I held off on posting it, because I was going to submit it to a few sites for publication, which I never actually got around to doing. Then the strike was resolved, and it wasn’t topical anymore. However, I think the main points are still pretty sound, and I don’t really feel like re-writing it, so here’s the original draft, posted for your enjoyment.)
Analysis of the Chicago teacher’s strike has re-ignited the ongoing debate about merit pay for teachers, particularly in regards to standardized test scores. Rather than examine the benefits of merit pay, I’d like to take a minute to discuss some of the logical fallacies that become apparent when this issue is discussed in detail.
Teachers’ unions and their apologists (usually those on the left) generally oppose merit-based pay systems, not necessarily in principle, but due to more practical concerns of measurement techniques. The primary way of measuring “merit” is by performance on standardized tests. The unions and apologists will tell you that this measurement technique is faulty. That it fails to control for socioeconomic conditions. That standardized tests are biased and not an accurate reflection of the quality of education a student receives. They paint a picture of a nightmare scenario where normally excellent, kind-hearted teachers will be forced to stop really educating their students because, fearing for their jobs, they are now entirely focused on “teaching to the test” and drilling their students with completely useless knowledge on obscure and archaic topics like reading comprehension and basic math. A large percentage of the public seems to generally agree with this line of thought, that standardized testing is a poor measurement of teacher performance.
Meanwhile, the message we receive from politicians (from both parties) is that American students are “falling behind.” This being an election year, we can safely assume that we will be hearing speech after speech decrying the sorry state of American schools, with one candidate suggesting it means we need to spend more money, and the other candidate proposing it means we need some other type of reform. But neither candidate would suggest that our schools aren’t falling behind. The rhetoric isn’t limited to election campaigning either. Feel free to look up any Presidential state of the union address, and you can be assured that at some point, a reference will be made to the declining quality of American schools. This rarely disputed narrative, cited by numerous studies, maintains that the educational quality of American schools continues to fall in comparison to other “developed” or “western” nations, and that we are even on the verge of being surpassed by countries such as China or India (perish the thought!) How exactly do we know this is the case? How do we arrive at the conclusions such as “In math literacy, the United States ranks 25th among OECD countries?” Why, by performance on standardized tests, of course. How else can we be expected to measure student performance globally? A large percentage of the public seems to generally agree with this line of thought as well, that standardized testing is a legitimate way to measure the continuing decline of the quality of education in America.
Notice the inconsistency. The average American tends to believe that standardized testing is a poor way of measuring the performance of individual schools or individual teachers, but is a perfectly legitimate way to measure the performance of nations. If you suggest that poor test performance indicates a teacher is doing something wrong, you’re being unfair because there are probably a large number of factors outside of the teacher’s control contributing to the results. Meanwhile, poor test performance on a national level is clearly an obvious signal that we need to “invest in the next generation,” (as in, spend more money that we don’t have). Anyone who suggests that on a national level there may be valid socioeconomic or cultural reasons that our students perform poorly is simply making excuses. At first glance, this line of thinking may appear to be rational. After all, at the national level you are dealing with a much larger sample size in which case variables such as economic well-being, local standards and regulations, etc. are likely to average out. However, simply comparing nation against nation does not control for these variables. Statistics such as population size, population density, poverty rates, ethnic diversity, and all the other potential variables in play vary wildly across different nations. Is it fair to compare the United States to South Korea, a much smaller and less diverse society? Is it fair to compare the United States to China, which is much larger and has much more endemic poverty? The same objections of unfairness that apply on the micro level apply on the macro level as well.
Eventually, we are going to have to decide. We cannot continue to cherry-pick, acknowledging test results when we want to demand more funding, and ignoring test results when we want to evaluate teacher performance. Either standardized testing conveys valuable information about the quality of education students receive, or it doesn’t.